inst/sgm/wbc19-aluru.R
In PriceLab/TrenaProjectArabidopsisRoot: explore transporters in Arabidopsis roots
library(GSEABase)
library(GOstats)
library(GO.db)
library(Category)
library(trena)
library(trenaSGM)
library(TrenaProjectArabidopsisRoot)
library(org.At.tair.db)
#------------------------------------------------------------------------------------------------------------------------
geneOntology <- function(orf)
{
tbl.go <- select(org.At.tair.db, keytype="TAIR", keys="AT3G59060", columns="GO")
return(select(GO.db, keys=tbl.go$GO, columns=c("TERM"))$TERM)
} # geneOntology
#------------------------------------------------------------------------------------------------------------------------
runTests <- function()
{
test_geneOntology()
test_goEnrichment()
} # runTests
#------------------------------------------------------------------------------------------------------------------------
test_geneOntology <- function()
{
message(sprintf("--- test_geneOntology"))
geneSymbol <- "PIF5"
orfName <- getGeneNames(tp, geneSymbol)$orf
tbl.go <- geneOntology(orfName)
} # test_geneOntology
#------------------------------------------------------------------------------------------------------------------------
goEnrichment <- function(orfNames, maxCategories)
{
gene.universe = character(0)
go.params <- new("GOHyperGParams", geneIds=orfNames,
universeGeneIds=gene.universe, annotation = "org.At.tair.db",
ontology = 'BP', pvalueCutoff = 0.01, conditional = FALSE,
testDirection = "over")
message(sprintf("about to calculate GO enrichment"))
go.bp.hgr <- hyperGTest(go.params)
tbl.go <- summary(go.bp.hgr)
if(nrow(tbl.go) > maxCategories)
tbl.go <- head(tbl.go, n=maxCategories)
message(sprintf("about to look up genes for each of %d GO categories", nrow(tbl.go)))
suppressMessages(
geneSymbols <- lapply(tbl.go$GOBPID,
function(goTerm){
if(!goTerm %in% keys(org.At.tair.db, keytype="GOALL")) {
printf("failed goTerm: %s", goTerm)
return ("")
}
keepers <- intersect(orfNames,select(org.At.tair.db, keys=goTerm, keytype="GOALL", columns="TAIR")$TAIR)
keeper.geneSymbols <- unlist(lapply(keepers, function(orf) getGeneNames(tp,orf)$symbol))
paste(keeper.geneSymbols, collapse=";")
}))
tbl.go$genes <- unlist(geneSymbols)
tbl.go
} # goEnrichment
#------------------------------------------------------------------------------------------------------------------------
test_goEnrichment <- function()
{
message(sprintf("--- test_goEnrichment"))
orfNames <- c("AT3G55130", "AT3G59060", "AT4G00050", "AT4G36540", "AT5G62165", "AT5G18830","AT5G60200",
"AT1G53160", "AT1G09530", "AT3G15270","AT1G80840")
tbl.enrichment <- goEnrichment(orfNames, maxCategories=5) # 10 is more typical
} # test_goEnrichment
#------------------------------------------------------------------------------------------------------------------------
if(!exists("tp")) {
message(sprintf("--- creating instance of TrenaProjectArabidopsisRoot"))
tp <- TrenaProjectArabidopsisRoot();
}
genome <- "tair10"
getExpressionMatrixNames(tp)
mtx <- getExpressionMatrix(tp, "aluru.18617x1938.orfRowNames")
targetGene <- canonicalizeName(tp, "WBC19")
#--------------------------------------------------------------------------
# first model: only 18 TFs with >= 95% motif match in short root-specific
# ATAC-seq about 2k upstream of WBC19
#--------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20438126,20438543),
end=c(20438246,20438984),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x <- build(builder)
lapply(x, dim)
tbl.model <- x$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model$symbol <- tf.symbols
# tbl.model:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT3G59060 0.25182969 7.586023e-257 0.6738150 1301.6266 0.17793969 0.5077358 NA PIF5
# 6 AT4G00050 0.25848974 1.790158e-129 0.5765029 738.1581 0.23987441 0.4916015 NA UNE10
# 7 AT4G36540 0.20257033 1.016042e-88 0.5333334 701.2759 0.19764529 0.4806764 NA BEE2
# 10 AT5G62165 -0.09238406 6.961638e-41 -0.4991596 513.0297 -0.11703297 -0.3849297 NA AGL42
# 9 AT5G60200 -0.10874331 4.796574e-21 -0.3877858 347.2974 -0.14396653 -0.3551172 NA DOF5.3
# 8 AT5G18830 -0.25172200 6.186848e-25 -0.3776184 331.5325 -0.28773203 -0.3622195 NA SPL7
# 2 AT1G53160 0.06447171 4.696022e-21 0.3917817 298.7136 0.09052672 0.3554472 NA SPL4
# 1 AT1G09530 0.01684827 1.433186e-05 0.4333553 256.6698 0.06233615 0.3551013 NA PIF3
# 4 AT3G15270 0.07932286 2.032701e-26 0.4072103 243.6807 0.10121717 0.3111859 NA SPL5
# 3 AT1G80840 0.00000000 9.856132e-01 0.1509693 175.8583 0.01323395 0.1306107 NA WRKY40
#------------------------------------------------------------------------------
# second model: 16 TFs with >= 95% motif match within 500kb (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20437000,20434000),
end=c(20437500,20434500),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x2 <- build(builder)
lapply(x2, dim)
tbl.model.2 <- x2$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.2$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.2$symbol <- tf.symbols
# tbl.model.2
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT4G14465 -0.46401927 1.919310e-162 -0.5628267 1331.0327 -0.35798325 -0.3313744 NA AHL20
# 10 AT5G62165 -0.26067798 8.382164e-104 -0.4991596 871.5585 -0.22696959 -0.3849297 NA AGL42
# 9 AT5G60200 -0.15703209 1.020370e-22 -0.3877858 573.3732 -0.18761945 -0.3551172 NA DOF5.3
# 6 AT5G13080 -0.13021371 1.794286e-21 -0.3233200 352.6772 -0.13125359 -0.2561005 NA WRKY75
# 7 AT5G46350 0.26386017 1.382743e-16 0.2225107 333.4381 0.25837985 0.1982006 NA WRKY8
# 4 AT2G46130 -0.06135211 2.192553e-12 -0.3191933 326.8847 -0.08656127 -0.2546452 NA WRKY43
# 8 AT5G49520 -0.05820226 6.901810e-05 -0.2456283 322.7049 -0.11555394 -0.2372729 NA WRKY48
# 1 AT1G64620 0.05390654 1.121712e-02 -0.1025043 269.3130 0.10217146 -0.1231658 NA AT1G64620
# 3 AT2G37590 0.00000000 1.613487e-01 -0.2068212 250.8909 -0.09618521 -0.1980168 NA DOF2.4
# 2 AT1G80840 0.02794550 1.308692e-02 0.1509693 242.7149 0.08498833 0.1306107 NA WRKY40
#------------------------------------------------------------------------------
# third model: 36 TFs with >= 95% motif match within 1000kb (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20434500,20436000),
end=c(20435500,20437000),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x3 <- build(builder)
lapply(x3, dim)
tbl.model.3 <- x2$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.3$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.3$symbol <- tf.symbols
# tbl.model.3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT4G14465 -0.46401927 1.919310e-162 -0.5628267 1331.0327 -0.35798325 -0.3313744 NA AHL20
# 10 AT5G62165 -0.26067798 8.382164e-104 -0.4991596 871.5585 -0.22696959 -0.3849297 NA AGL42
# 9 AT5G60200 -0.15703209 1.020370e-22 -0.3877858 573.3732 -0.18761945 -0.3551172 NA DOF5.3
# 6 AT5G13080 -0.13021371 1.794286e-21 -0.3233200 352.6772 -0.13125359 -0.2561005 NA WRKY75
# 7 AT5G46350 0.26386017 1.382743e-16 0.2225107 333.4381 0.25837985 0.1982006 NA WRKY8
# 4 AT2G46130 -0.06135211 2.192553e-12 -0.3191933 326.8847 -0.08656127 -0.2546452 NA WRKY43
# 8 AT5G49520 -0.05820226 6.901810e-05 -0.2456283 322.7049 -0.11555394 -0.2372729 NA WRKY48
# 1 AT1G64620 0.05390654 1.121712e-02 -0.1025043 269.3130 0.10217146 -0.1231658 NA AT1G64620
# 3 AT2G37590 0.00000000 1.613487e-01 -0.2068212 250.8909 -0.09618521 -0.1980168 NA DOF2.4
# 2 AT1G80840 0.02794550 1.308692e-02 0.1509693 242.7149 0.08498833 0.1306107 NA WRKY40
# side note: AGL42 is within the top 5 among all 3 models (repressor)
# another note: AGL40 is the first one among models 2 & 3
# make sure to review the others as well
#------------------------------------------------------------------------------
# fourth model: 19 TFs with >= 95% motif match within 1200kb (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432000,20433210),
end=c(20433200,20434410),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
#side note: next model, go back to around the third model's region.
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x4 <- build(builder)
lapply(x4, dim)
tbl.model.4 <- x4$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.4$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.4$symbol <- tf.symbols
# tbl.model.4
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 7 AT4G14465 -0.63306716 1.919310e-162 -0.5628267 1372.8482 -0.44939720 -0.3313744 NA AHL20
# 9 AT5G60200 -0.23040752 9.592281e-32 -0.3877858 673.3668 -0.24152133 -0.3551172 NA DOF5.3
# 1 AT1G52150 -0.01352807 3.660479e-02 -0.2822424 493.6272 -0.10277135 -0.3354909 NA ATHB-15
# 4 AT2G22430 0.53927909 3.009286e-43 0.3242114 481.4268 0.48990652 0.2671454 NA ATHB-6
# 10 AT5G65590 0.19838070 5.449207e-39 0.3535122 454.8653 0.19894077 0.2939522 NA AT5G65590 aka SCAP1
# 6 AT3G01470 0.00000000 9.838639e-01 -0.2254802 327.7194 -0.10693530 -0.1848806 NA ATHB-1
# 2 AT1G64620 0.04547039 1.688527e-02 -0.1025043 292.8133 0.10958571 -0.1231658 NA AT1G64620
# 5 AT2G37590 0.00000000 5.330965e-01 -0.2068212 261.5152 -0.08311396 -0.1980168 NA DOF2.4
# 8 AT5G39660 -0.03144685 5.430603e-03 -0.2213106 254.0163 -0.12676544 -0.1426431 NA CDF2
# 3 AT1G80840 0.00000000 3.945345e-01 0.1509693 252.7218 0.03454721 0.1306107 NA WRKY40
#------------------------------------------------------------------------------
# fifth model: 30 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(22532400,22538210),
end=c(22533900,22539910),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x5 <- build(builder)
lapply(x5, dim)
tbl.model.5 <- x5$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.5$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.5$symbol <- tf.symbols
# tbl.model.5
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 -0.45126281 1.919310e-162 -0.5628267 1217.9599 -0.33094480 -0.3313744 NA AHL20
# 4 AT2G01760 0.35079119 1.811735e-68 0.4259098 591.9681 0.31450068 0.3974638 NA ARR14
# 9 AT4G31920 -0.41681921 2.657399e-51 -0.3963221 578.9558 -0.38395182 -0.3279602 NA ARR10
# 12 AT5G60200 -0.21672612 2.056503e-33 -0.3877858 526.6647 -0.22047585 -0.3551172 NA DOF5.3
# 2 AT1G67710 -0.04803401 3.653619e-14 -0.3991527 319.9506 -0.09684702 -0.3220879 NA ARR11
# 13 AT5G65590 0.14140722 6.260262e-34 0.3535122 297.9758 0.15024399 0.2939522 NA AT5G65590 aka SCAP1
# 10 AT4G40060 0.00000000 2.658003e-01 -0.2832086 235.4189 -0.01086425 -0.2614751 NA ATHB-16
# 1 AT1G64620 0.03707026 1.295114e-02 -0.1025043 207.6641 0.10161573 -0.1231658 NA AT1G64620
# 5 AT2G37590 -0.04281820 3.412511e-03 -0.2068212 200.8512 -0.14316299 -0.1980168 NA DOF2.4
# 11 AT5G39660 -0.08074957 4.412580e-06 -0.2213106 191.9243 -0.14452532 -0.1426431 NA CDF2
#------------------------------------------------------------------------------
# sixth model: 29 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(13032400,13037100),
end=c(13033400,13038100),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x6 <- build(builder)
lapply(x6, dim)
tbl.model.6 <- x6$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.6$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.6$symbol <- tf.symbols
# tbl.model.6
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 -0.38623478 1.919310e-162 -0.5628267 1072.8117 -0.29493570 -0.3313744 NA AHL20
# 15 AT5G62165 -0.17274948 8.382164e-104 -0.4991596 660.4390 -0.16416343 -0.3849297 NA AGL42
# 9 AT4G31920 -0.30060591 3.462460e-41 -0.3963221 465.5176 -0.30569923 -0.3279602 NA ARR10
# 14 AT5G60200 -0.12411684 3.034644e-23 -0.3877858 402.2469 -0.16687658 -0.3551172 NA DOF5.3
# 1 AT1G17950 -0.05301677 3.260175e-34 -0.4291068 352.9953 -0.08665296 -0.3200268 NA MYB52
# 7 AT2G42200 0.17331619 1.024211e-34 0.3494224 256.2832 0.17743521 0.3171417 NA SPL9
# 5 AT2G22430 0.34975419 6.445556e-38 0.3242114 246.3073 0.31453242 0.2671454 NA ATHB-6
# 11 AT5G02320 -0.17835622 1.354184e-32 -0.3254427 218.6656 -0.17944006 -0.3079975 NA MYB3R-5
# 16 AT5G65590 0.10574985 1.332595e-33 0.3535122 216.4876 0.12160481 0.2939522 NA AT5G65590 aka SCAP1
# 12 AT5G11510 -0.06323905 1.388600e-09 -0.2168312 177.7344 -0.09146526 -0.2150125 NA MYB3R-4
#------------------------------------------------------------------------------------------------------------------------------
# Get high expression data
wbc19.vector <- mtx[c("AT3G55130", "AT3G59060"),]
plotWBC19 <- plot((mtx[c("AT3G55130"),]), (mtx[c("AT3G59060"),]), main= "WBC19 Expression in Roots")
new.wbc.vector <- t(wbc19.vector)
number1 <- new.wbc.vector[(new.wbc.vector[,1] > 2),]
number2 <- number1[(number1[,2] > 2),]
plot(number2, main="High WBC19 Expression in Roots")
highExpressionWBC19 <- t(number2)
rownames(highExpressionWBC19)
mtx2 <- highExpressionWBC19
mtx2.colnames <- colnames(mtx2)
mtx2 <- mtx[,mtx2.colnames]
# created matrix of high expression data
# mtx3 has 169 samples
ordered.number2<- number2[order(number2[,1], number2[,2], decreasing=TRUE),]
# allows you to view sample names with highest expression
ordered.number2.50 <- ordered.number2[1:50,]
evenHigherExpressionWBC19 <- t(ordered.number2.50)
rownames(evenHigherExpressionWBC19)
mtx3.colnames <- colnames(evenHigherExpressionWBC19)
mtx3 <- mtx[,mtx3.colnames]
#build smaller matrix with even higher expression of WBC19
# top 1-3,7 samples are from E-GEOD-24348
# Experiment: Transcription profiling by array of Arabidopsis mutant for nas4x after iron deprivation
# top 4-6,9-10 samples are from experiment 489 from NASC arrays:
# Experiment: Investigating the molecular response of Arabidopsis to combined
# nematode and dehydration stress using the Affymetrix ATH1 microarray.
# top 8 sample is from Experiment #490 from NASC arrays:
# Experiment: Cell-specific nitrogen responses in the Arabidopsis root
#------------------------------------------------------------------------------------------------------------------------------
#Modeling with high expression data:
#------------------------------------------------------------------------------
# seventh model: 32 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(22632400,22638210),
end=c(22633900,22639910),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx2,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x7 <- build(builder)
lapply(x7, dim)
tbl.model.7 <- x7$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.7$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.7$symbol <- tf.symbols
# tbl.model.7
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT3G61150 1.844280e-01 6.937774e-06 0.3380855 6.106435 0.13379306 0.3369922 NA HDG1
# 11 AT5G16560 1.117630e-02 2.977540e-02 0.2091608 5.729552 0.03780677 0.2138357 NA KAN
# 13 AT5G60200 -7.479631e-02 1.935798e-03 -0.2496830 5.132500 -0.08682439 -0.3265924 NA DOF5.3
# 7 AT2G37590 -1.357610e-01 5.089497e-04 -0.2755380 4.436605 -0.14352526 -0.3060042 NA DOF2.4
# 6 AT2G22430 4.522021e-02 2.389368e-02 0.2079384 4.399553 0.10152916 0.2029337 NA ATHB-6
# 4 AT1G73360 8.112228e-03 4.952906e-02 0.1015424 4.388088 0.05521869 0.2124932 NA EDT1
# 12 AT5G25830 3.193348e-02 1.265326e-02 0.1590170 3.715073 0.04473851 0.1023321 NA GATA12
# 9 AT3G54810 -4.519279e-02 1.264134e-02 -0.1717809 3.307638 -0.07737611 -0.1361245 NA GATA8
# 14 AT5G62940 0.000000e+00 8.643032e-01 -0.1254659 2.961352 -0.02415076 -0.1940754 NA DOF5.6
# 8 AT2G46130 -7.883023e-05 6.007031e-02 -0.1284293 2.786565 -0.02130100 -0.1615932 NA WRKY43
#------------------------------------------------------------------------------------------------------------------------
#use smaller mtx (mtx3) for even higher expression data
#------------------------------------------------------------------------------
# eighth model: 79 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(18632400,18638210),
end=c(18633900,18639710),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x8 <- build(builder)
lapply(x8, dim)
tbl.model.8 <- x8$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.8$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.8$symbol <- tf.symbols
# tbl.model.8
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G25550 -0.015443951 0.02172242 -0.3272428 0.9130473 -0.027804138 -0.29190876 NA AT1G25550
# 8 AT1G74930 0.000000000 0.83419203 -0.2764086 0.4672606 -0.007247037 -0.27875150 NA ERF018
# 33 AT5G65210 0.000000000 0.13996296 -0.2391059 0.4449350 -0.025328820 -0.16869148 NA TGA1
# 31 AT5G62165 0.000000000 0.24719669 0.2466379 0.3578841 0.017652083 0.25051621 NA AGL42
# 4 AT1G61110 0.000000000 0.04399367 0.2745188 0.3098864 0.027149326 0.36998800 NA anac025
# 3 AT1G53160 0.000000000 0.51715079 0.1885299 0.2887157 0.022863901 0.18703481 NA SPL4
# 20 AT3G54810 -0.005432485 0.03930759 -0.2940763 0.2704288 -0.043393328 -0.20384154 NA GATA8
# 29 AT5G39610 0.000000000 0.09783550 -0.2399027 0.2386434 -0.012143183 -0.27452581 NA ATNAC2
# 18 AT3G06740 0.029844783 0.01895477 0.3307985 0.2139190 0.045933432 0.27352397 NA GATA15
# 32 AT5G62940 0.000000000 0.20621717 0.1423515 0.2113919 0.018586458 0.08821128 NA DOF5.6
#------------------------------------------------------------------------------
# ninth model: 24 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(13532400,13538210),
end=c(13533400,13539210),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x9 <- build(builder)
lapply(x9, dim)
tbl.model.9 <- x9$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.9$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.9$symbol <- tf.symbols
# tbl.model.9
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT3G54810 -0.060234503 0.03603976 -0.2940763 0.6984862 -0.063738119 -0.20384154 NA GATA8
# 12 AT5G62940 0.000000000 0.24576433 0.1423515 0.6378016 0.028938147 0.08821128 NA DOF5.6
# 7 AT4G32890 0.003008057 0.10549938 0.2345584 0.6144417 0.027627205 0.19164466 NA GATA9
# 5 AT3G06740 0.101587678 0.01895477 0.3307985 0.6094221 0.073410073 0.27352397 NA GATA15
# 2 AT1G53160 0.000000000 0.34003156 0.1885299 0.5776807 0.026310324 0.18703481 NA SPL4
# 10 AT5G25830 0.000000000 0.25792875 0.1818998 0.5764563 0.015273616 0.07130852 NA GATA12
# 9 AT5G16560 0.000000000 0.56268364 0.1300772 0.5381874 0.008683821 0.15668667 NA KAN
# 8 AT5G02840 0.000000000 0.12336829 -0.1900216 0.3819700 -0.043429621 -0.11750300 NA LCL1
# 13 AT5G65590 0.000000000 0.45605691 -0.1758766 0.3539472 -0.009495115 -0.30208884 NA AT5G65590 aka SCAP1
# 3 AT1G64620 0.000000000 0.15354621 0.2198560 0.3528297 0.049762578 0.21949580 NA AT1G64620
#------------------------------------------------------------------------------
# 10th model: 27 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20434000,20435500),
end=c(20435000,20436500),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x10 <- build(builder)
lapply(x10, dim)
tbl.model.10 <- x10$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.10$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.10$symbol <- tf.symbols
# tbl.model.10
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT5G46760 0.000000000 0.72496950 0.1201494 0.5582605 0.02368710 0.18722689 NA AT5G46760
# 5 AT3G06740 0.099267109 0.01895477 0.3307985 0.5543676 0.06976204 0.27352397 NA GATA15
# 12 AT5G62165 0.003440748 0.08922270 0.2466379 0.5489340 0.02126300 0.25051621 NA AGL42
# 4 AT2G36270 0.000000000 0.21446488 -0.2409433 0.5174508 -0.01353082 -0.11510204 NA ABI5
# 13 AT5G62940 0.000000000 0.23139773 0.1423515 0.5087755 0.02582457 0.08821128 NA DOF5.6
# 7 AT3G54810 -0.054040991 0.03603976 -0.2940763 0.4856329 -0.04923703 -0.20384154 NA GATA8
# 1 AT1G53160 0.000000000 0.26685569 0.1885299 0.4703625 0.02634593 0.18703481 NA SPL4
# 6 AT3G47500 0.000000000 0.27552762 0.2092642 0.4636700 0.02766993 0.29901561 NA CDF3
# 10 AT5G16560 0.000000000 0.60334479 0.1300772 0.4396909 0.01021718 0.15668667 NA KAN
# 2 AT1G64620 0.000000000 0.13975644 0.2198560 0.4252454 0.06111624 0.21949580 NA AT1G64620
#------------------------------------------------------------------------------
# 11th model: 19 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20433010,20439400),
end=c(20434010,20440400),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x11 <- build(builder)
lapply(x11, dim)
tbl.model.11 <- x11$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.11$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.11$symbol <- tf.symbols
# tbl.model.11
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 1 AT1G25550 -0.022746363 0.02172242 -0.3272428 1.2660687 -0.039665255 -0.29190876 NA AT1G25550
# 5 AT3G54810 -0.025411789 0.04902443 -0.2940763 0.6545343 -0.051632292 -0.20384154 NA GATA8
# 10 AT5G62165 0.001002055 0.10201106 0.2466379 0.5926299 0.021996337 0.25051621 NA AGL42
# 11 AT5G62940 0.000000000 0.32551922 0.1423515 0.5173541 0.026794172 0.08821128 NA DOF5.6
# 6 AT4G14465 -0.029122906 0.02643625 -0.3113506 0.5150485 -0.027251150 -0.28038415 NA AHL20
# 4 AT3G06740 0.091388846 0.01895477 0.3307985 0.5034917 0.072256535 0.27352397 NA GATA15
# 2 AT1G64620 0.000000000 0.11792728 0.2198560 0.4519734 0.060525966 0.21949580 NA AT1G64620
# 9 AT5G16560 0.000000000 0.73158073 0.1300772 0.3951998 0.008438453 0.15668667 NA KAN
# 12 AT5G65590 0.000000000 0.74044625 -0.1758766 0.3658925 -0.009455525 -0.30208884 NA AT5G65590 aka SCAP1
# 8 AT5G03790 0.000000000 0.19902620 -0.1720009 0.3526549 -0.013799422 -0.20096038 NA ATHB-51
#------------------------------------------------------------------------------
# 12th model: 63 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20431010,20436400),
end=c(20432010,20437400),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x12 <- build(builder)
lapply(x12, dim)
tbl.model.12 <- x12$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.12$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.12$symbol <- tf.symbols
# tbl.model.12
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 17 AT4G17880 0.093382268 0.0008222258 0.4580876 1.2723674 0.04412493 0.4253061 NA AT4G17880
# 3 AT1G22985 -0.033356269 0.0099813571 -0.3858113 0.7122849 -0.04852291 -0.3998559 NA AT1G22985
# 13 AT3G15210 0.000000000 0.7570313477 -0.2229774 0.3615262 -0.01276041 -0.1080912 NA ERF4
# 23 AT5G07680 0.000000000 0.0966796673 -0.2597433 0.3409373 -0.01130599 -0.1699400 NA ANAC080
# 19 AT4G32800 0.000000000 0.0847828944 0.2511761 0.2765392 0.01980436 0.2693397 NA AT4G32800
# 15 AT4G14465 -0.018735267 0.0139181058 -0.3113506 0.2307609 -0.02534347 -0.2803842 NA AHL20
# 14 AT4G00050 0.000000000 0.5831800806 0.1559773 0.2282816 0.01596675 0.2100840 NA UNE10
# 26 AT5G43410 0.003556295 0.0252531286 0.2610044 0.2255470 0.01595780 0.3809364 NA AT5G43410
# 27 AT5G46760 0.000000000 0.8295434551 0.1201494 0.2081812 0.02031843 0.1872269 NA AT5G46760
# 29 AT5G62165 0.000000000 0.0632480856 0.2466379 0.1992148 0.01515128 0.2505162 NA AGL42
#------------------------------------------------------------------------------
# 13th model: 52 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20430000,20435500),
end=c(20431500,20437000),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x13 <- build(builder)
lapply(x13, dim)
tbl.model.13 <- x13$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.13$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.13$symbol <- tf.symbols
# tbl.model.13
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G22985 -0.072674574 0.005655387 -0.3858113 0.8585397 -0.05207582 -0.39985594 NA AT1G22985
# 11 AT3G15210 0.000000000 0.195240490 -0.2229774 0.4950218 -0.01760155 -0.10809124 NA ERF4
# 20 AT5G07680 0.000000000 0.126299481 -0.2597433 0.4152436 -0.01300549 -0.16993998 NA ANAC080
# 15 AT4G32800 0.000000000 0.060636290 0.2511761 0.3392273 0.02260782 0.26933974 NA AT4G32800
# 23 AT5G46760 0.000000000 0.402790027 0.1201494 0.2764723 0.02720795 0.18722689 NA AT5G46760
# 10 AT3G06740 0.004794159 0.044481922 0.3307985 0.2722499 0.04303362 0.27352397 NA GATA15
# 7 AT1G69780 0.000000000 0.211205052 -0.1558149 0.2606508 -0.01934952 -0.24043217 NA ATHB13
# 25 AT5G62940 0.000000000 0.322875673 0.1423515 0.2501957 0.01630500 0.08821128 NA DOF5.6
# 12 AT3G47500 0.000000000 0.807130465 0.2092642 0.2484925 0.01951010 0.29901561 NA CDF3
# 22 AT5G43410 0.000000000 0.052223010 0.2610044 0.2283790 0.01637408 0.38093637 NA AT5G43410
#no regulatory regions yet
#------------------------------------------------------------------------------
# 14th model: 42 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20423050,20427384),
end=c(20424550,20428884),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x14 <- build(builder)
lapply(x14, dim)
tbl.model.14 <- x14$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.14$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.14$symbol <- tf.symbols
# tbl.model.14
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT4G14465 -0.032987913 0.02604191 -0.3113506 0.5811695 -0.032779964 -0.28038415 NA AHL20
# 8 AT3G47500 0.000000000 0.10256329 0.2092642 0.5284426 0.038234129 0.29901561 NA CDF3
# 3 AT1G69780 0.000000000 0.47126347 -0.1558149 0.4558065 -0.020464919 -0.24043217 NA ATHB13
# 17 AT5G62940 0.000000000 0.44899545 0.1423515 0.4496875 0.019987754 0.08821128 NA DOF5.6
# 7 AT3G06740 0.070921388 0.01895477 0.3307985 0.4338415 0.057434221 0.27352397 NA GATA15
# 11 AT4G17460 0.009649258 0.04073222 0.3055506 0.4336755 0.024513398 0.24264106 NA HAT1
# 2 AT1G64620 0.000000000 0.11742320 0.2198560 0.4100220 0.060567525 0.21949580 NA AT1G64620
# 12 AT4G37180 -0.025091221 0.05295391 -0.2804712 0.3833676 -0.044010154 -0.29142857 NA AT4G37180
# 15 AT5G16560 0.000000000 0.95486682 0.1300772 0.3681782 0.006399682 0.15668667 NA KAN
# 5 AT2G46680 0.000000000 0.10509509 -0.2534576 0.3379815 -0.017638285 -0.26261705 NA ATHB-7
#---------------------------------------------------------------------------------------
# 15th model: 20 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
#---------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432106,20434107),
end=c(20434106,20434607),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x15 <- build(builder)
lapply(x15, dim)
tbl.model.15 <- x15$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.15$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.15$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT3G54810 -0.039787610 0.04365073 -0.2940763 0.6846830 -0.06001076 -0.20384154 NA GATA8
# 5 AT3G06740 0.124765638 0.01895477 0.3307985 0.6674693 0.08295289 0.27352397 NA GATA15
# 10 AT5G62940 0.000000000 0.41461791 0.1423515 0.6670734 0.02197803 0.08821128 NA DOF5.6
# 7 AT4G14465 -0.035092534 0.02604191 -0.3113506 0.6431163 -0.03020385 -0.28038415 NA AHL20
# 2 AT1G69780 0.000000000 0.91517751 -0.1558149 0.6375511 -0.01767389 -0.24043217 NA ATHB13
# 1 AT1G64620 0.003497219 0.12537262 0.2198560 0.6301531 0.06288908 0.21949580 NA AT1G64620
# 9 AT5G16560 0.000000000 0.54146029 0.1300772 0.5653065 0.01288556 0.15668667 NA KAN
# 11 AT5G65590 0.000000000 0.53350580 -0.1758766 0.4449672 -0.01152127 -0.30208884 NA AT5G65590 aka SCAP1
# 3 AT1G80840 0.000000000 0.44366311 -0.1730834 0.3949913 -0.01562107 -0.14900360 NA WRKY40
# 4 AT3G01470 0.000000000 0.85935301 -0.1515070 0.3905654 -0.01787426 -0.13373349 NA ATHB-1
#----------------------------------------------------------------------------------------
# 16th model: 20 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# same region as #15 on atac.seq; Use bigger matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432112,20434113),
end=c(20434112,20434613),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x16 <- build(builder)
lapply(x16, dim)
tbl.model.16 <- x16$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.16$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.16$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 7 AT4G14465 -0.631679182 1.919310e-162 -0.5628267 1392.8188 -0.45797827 -0.3313744 NA AHL20
# 9 AT5G60200 -0.226439315 9.592281e-32 -0.3877858 663.6387 -0.24379551 -0.3551172 NA DOF5.3
# 1 AT1G52150 -0.007069971 3.660479e-02 -0.2822424 510.0015 -0.10189183 -0.3354909 NA ATHB-15
# 4 AT2G22430 0.534924218 3.009286e-43 0.3242114 484.3623 0.49642149 0.2671454 NA ATHB-6
# 10 AT5G65590 0.196341104 5.449207e-39 0.3535122 453.4247 0.20074867 0.2939522 NA AT5G65590 aka SCAP1
# 6 AT3G01470 0.000000000 9.838639e-01 -0.2254802 318.4881 -0.10405340 -0.1848806 NA ATHB-1
# 2 AT1G64620 0.032505573 1.688527e-02 -0.1025043 291.1077 0.11394430 -0.1231658 NA AT1G64620
# 5 AT2G37590 0.000000000 5.330965e-01 -0.2068212 260.5657 -0.08101563 -0.1980168 NA DOF2.4
# 8 AT5G39660 -0.026584270 5.430603e-03 -0.2213106 253.8812 -0.12570706 -0.1426431 NA CDF2
# 3 AT1G80840 0.000000000 3.945345e-01 0.1509693 251.0913 0.03387059 0.1306107 NA WRKY40
#with mtx3:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT3G54810 -0.03167221 0.04365073 -0.2940763 0.8106728 -0.05785286 -0.20384154 NA GATA8
# 2 AT1G69780 0.00000000 0.91517751 -0.1558149 0.6819254 -0.01714176 -0.24043217 NA ATHB13
# 10 AT5G62940 0.00000000 0.41461791 0.1423515 0.6513227 0.02122323 0.08821128 NA DOF5.6
# 7 AT4G14465 -0.02996968 0.02604191 -0.3113506 0.6449731 -0.02901716 -0.28038415 NA AHL20
# 5 AT3G06740 0.10799807 0.01895477 0.3307985 0.6422672 0.07991416 0.27352397 NA GATA15
# 9 AT5G16560 0.00000000 0.54146029 0.1300772 0.5825086 0.01236389 0.15668667 NA KAN
# 1 AT1G64620 0.00000000 0.12537262 0.2198560 0.5129569 0.06036687 0.21949580 NA AT1G64620
# 11 AT5G65590 0.00000000 0.53350580 -0.1758766 0.4958118 -0.01116943 -0.30208884 NA AT5G65590 aka SCAP1
# 8 AT5G03790 0.00000000 0.18322862 -0.1720009 0.4236046 -0.01437446 -0.20096038 NA ATHB-51
# 3 AT1G80840 0.00000000 0.44366311 -0.1730834 0.4227791 -0.01505340 -0.14900360 NA WRKY40
#----------------------------------------------------------------------------------------
# 17th model: 28 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# same region as #15 on atac.seq; Use smaller matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432112,20434113),
end=c(20434112,20436113),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x17 <- build(builder)
lapply(x17, dim)
tbl.model.17 <- x17$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.17$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.17$symbol <- tf.symbols
#with mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT3G54810 -0.03315773 0.08019973 -0.2644196 8.106659 -0.16569641 -0.20000000 NA GATA8
# 6 AT2G37590 0.19950526 0.04200000 0.2887062 6.680566 0.24873045 0.30727491 NA DOF2.4
# 14 AT5G65590 0.00000000 0.24946301 -0.1392564 6.477805 -0.03827349 -0.32513806 NA AT5G65590
# 9 AT4G14465 -0.06861681 0.03802656 -0.2942996 6.385959 -0.06734547 -0.17166867 NA AHL20
# 11 AT5G18830 0.00000000 0.12425412 -0.2049080 6.275481 -0.23378909 -0.09176471 NA SPL7
# 4 AT1G69780 0.00000000 0.95502237 -0.2050077 4.572243 -0.04328460 0.05690276 NA ATHB13
#----------------------------------------------------------------------------------------
# 18th model: 104 TFs with >= 95% motif match within 5000 upstream, 5000 downstream of TSS
# same region as #15 on atac.seq; Use bigger matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20429112,20434113),
end=c(20434112,20439113),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x18 <- build(builder)
lapply(x18, dim)
tbl.model.18 <- x18$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.18$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.18$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 30 AT3G59060 0.20660505 7.586023e-257 0.6738150 1106.31242 0.09978381 0.5077358 NA PIF5
# 33 AT4G14465 -0.14559575 7.720037e-103 -0.5628267 537.79490 -0.09335849 -0.3313744 NA AHL20
# 32 AT4G00050 0.21111517 1.790158e-129 0.5765029 451.14938 0.14417470 0.4916015 NA UNE10
# 38 AT4G36540 0.16916609 2.128292e-89 0.5333334 329.98386 0.12713276 0.4806764 NA BEE2
# 23 AT3G01220 -0.08784450 1.524911e-46 -0.5233056 273.24071 -0.08700164 -0.3849471 NA ATHB20
# 51 AT5G62165 -0.03805766 3.058215e-24 -0.4991596 217.69778 -0.05817641 -0.3849297 NA AGL42
# 9 AT1G53160 0.02257309 1.136179e-12 0.3917817 92.34605 0.04437246 0.3554472 NA SPL4
# 50 AT5G60200 -0.02179131 5.127391e-12 -0.3877858 91.62244 -0.05630662 -0.3551172 NA DOF5.3
# 43 AT5G18830 -0.12205317 5.265642e-17 -0.3776184 82.37603 -0.15105459 -0.3622195 NA SPL7
# 35 AT4G31920 0.00000000 9.739117e-01 -0.3963221 82.05983 -0.08120219 -0.3279602 NA ARR10
#----------------------------------------------------------------------------------------
# 19th model: 98 TFs with >= 95% motif match within 5000 upstream, 5000 downstream of TSS
# same region as previous on atac.seq; Use smaller matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20429112,20436288),
end=c(20434112,20441288),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x19 <- build(builder)
lapply(x19, dim)
tbl.model.19 <- x19$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.19$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.19$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 27 AT3G59060 0.20847974 7.586023e-257 0.6738150 1132.98732 0.10504491 0.5077358 NA PIF5
# 30 AT4G14465 -0.14831565 7.720037e-103 -0.5628267 527.16775 -0.09513729 -0.3313744 NA AHL20
# 29 AT4G00050 0.21213270 1.790158e-129 0.5765029 467.57087 0.14878185 0.4916015 NA UNE10
# 21 AT3G01220 -0.08679869 1.524911e-46 -0.5233056 329.26474 -0.09229795 -0.3849471 NA ATHB20
# 34 AT4G36540 0.16911396 2.128292e-89 0.5333334 320.70080 0.12903040 0.4806764 NA BEE2
# 47 AT5G62165 -0.03733050 3.058215e-24 -0.4991596 185.77783 -0.06039029 -0.3849297 NA AGL42
# 46 AT5G60200 -0.02111709 5.127391e-12 -0.3877858 91.65985 -0.05759636 -0.3551172 NA DOF5.3
# 39 AT5G18830 -0.12003641 5.265642e-17 -0.3776184 86.14901 -0.15167134 -0.3622195 NA SPL7
# 9 AT1G53160 0.02216566 1.136179e-12 0.3917817 84.55421 0.04472416 0.3554472 NA SPL4
# 32 AT4G31920 0.00000000 9.427982e-01 -0.3963221 73.34116 -0.07899182 -0.3279602 NA ARR10
# with smaller matrix mtx3:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 29 AT4G17880 0.06445869 0.0008222258 0.4580876 0.9450456 0.034282571 0.4253061 NA AT4G17880
# 7 AT1G25550 0.00000000 0.0328008880 -0.3272428 0.5391818 -0.019705100 -0.2919088 NA AT1G25550
# 40 AT5G37020 0.10286850 0.0010999014 0.4500930 0.5314113 0.073007047 0.3855462 NA ARF8
# 5 AT1G22985 -0.01027799 0.0142236209 -0.3858113 0.3790130 -0.038876945 -0.3998559 NA AT1G22985
# 37 AT5G07680 0.00000000 0.0518950825 -0.2597433 0.3328196 -0.010513774 -0.1699400 NA ANAC080
# 22 AT3G15270 0.00000000 0.0593857696 0.3030271 0.2860886 0.013592940 0.2473469 NA SPL5
# 21 AT3G15210 0.00000000 0.7331515557 -0.2229774 0.1963197 -0.007838267 -0.1080912 NA ERF4
# 14 AT1G74930 0.00000000 0.6960000957 -0.2764086 0.1911578 -0.004877530 -0.2787515 NA ERF018
# 41 AT5G43410 0.00000000 0.0281112451 0.2610044 0.1895305 0.016059188 0.3809364 NA AT5G43410
# 32 AT4G32800 0.00000000 0.0749800426 0.2511761 0.1830661 0.018230405 0.2693397 NA AT4G32800
#------------------------------------------------------------------------------------------------------------------------
#Model transcription factors found and see what you find! Remember to use both the bigger and smaller matrix
#----------------------------------------------------------------------------------------
# 20th model: 48 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AHL20")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(8318970,8320971),
end=c(8320970,8322971),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AHL20",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x20 <- build(builder)
lapply(x20, dim)
tbl.model.20 <- x20$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.20$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.20$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT1G69780 0.48600063 9.251529e-08 0.6717750 10.353287 0.1155951 0.4410564 NA ATHB13
# 16 AT3G01470 0.00000000 6.915034e-01 0.4376918 3.907703 0.0631113 0.3113085 NA ATHB-1
# 23 AT4G40060 0.00000000 9.979775e-02 0.3992646 3.388234 0.1189144 0.2845138 NA ATHB-16
# 12 AT2G42200 0.00000000 3.466663e-02 -0.4381118 3.112813 -0.0665653 -0.2523409 NA SPL9
# 30 AT5G56270 -0.04441607 4.814786e-03 -0.3738075 2.231266 -0.1599834 -0.4034094 NA WRKY2
# 9 AT2G22430 0.00000000 7.772254e-02 -0.4260024 2.090012 -0.1382915 -0.4021609 NA ATHB-6
#----------------------------------------------------------------------------------------
# 21st model: 66 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "PIF5")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(21826187,21828188),
end=c(21828187,21828688),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="PIF5",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x21 <- build(builder)
lapply(x21, dim)
tbl.model.21 <- x21$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.21$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.21$symbol <- tf.symbols
#with bigger matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 28 AT4G36540 0.31729995 7.689933e-146 0.5413038 1301.6056 0.22207458 0.4579084 NA BEE2
# 36 AT5G62165 -0.20517737 9.475748e-170 -0.5731729 1169.2908 -0.15495633 -0.4259736 NA AGL42
# 24 AT4G31920 -0.33374588 7.894880e-84 -0.4761674 558.2853 -0.31075981 -0.3834930 NA ARR10
# 22 AT3G60490 -0.12682864 8.515071e-76 -0.4383758 403.6812 -0.09674127 -0.3911913 NA AT3G60490
# 35 AT5G60200 -0.16877910 5.074558e-47 -0.4158650 359.4931 -0.16318164 -0.3608002 NA DOF5.3
# 31 AT5G25830 -0.21821714 6.054286e-93 -0.4750765 342.0631 -0.20263815 -0.3635412 NA GATA12
# 1 AT1G17950 -0.08224473 1.652507e-97 -0.4788226 332.4711 -0.07033979 -0.3746657 NA MYB52
# 6 AT1G53170 0.17270085 1.592222e-70 0.4460720 267.3134 0.13643533 0.3760144 NA ERF8
# 21 AT3G57600 0.16206402 1.123689e-74 0.4555560 236.9006 0.13824439 0.3921747 NA AT3G57600
# 20 AT3G16770 0.16564353 1.974600e-35 0.3417913 179.2040 0.17261444 0.3205095 NA RAP2-3
#with small matrix:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 1 AT1G06160 0.000000000 0.035329865 0.2210415 2.217605 0.03002289 0.03442977 NA ERF094
# 10 AT2G22430 0.073019124 0.006538555 0.4177917 1.888240 0.07946312 0.38141657 NA ATHB-6
# 30 AT5G62000 0.032641170 0.002867844 0.4235871 1.837604 0.03986967 0.44460984 NA ARF2
# 3 AT1G28370 0.080778283 0.001006354 0.4512119 1.677484 0.03366724 0.46448980 NA ERF11
# 23 AT5G08130 -0.169730460 0.004738418 -0.3754061 1.591577 -0.13639390 -0.27308523 NA BIM1
# 32 AT5G65590 0.007382096 0.017227683 0.3482814 1.426294 0.03293105 0.32187275 NA AT5G65590 aka SCAP1
# 18 AT3G23240 0.000000000 0.457686819 0.3319695 1.276688 0.01306984 0.35001200 NA ERF1B
# 9 AT1G80840 0.000000000 0.920894439 0.4096893 1.226684 0.02722907 0.38285714 NA WRKY40
# 17 AT3G16770 0.005335604 0.022086968 0.2903235 1.150150 0.03497298 0.22698679 NA RAP2-3
# 16 AT3G15210 0.000000000 0.359911361 0.3449356 1.053483 0.03197530 0.42770708 NA ERF4
#----------------------------------------------------------------------------------------
# 22nd model: 26 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "UNE10")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(15862,17863),
end=c(17862,18363),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="UNE10",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x22 <- build(builder)
lapply(x22, dim)
tbl.model.22 <- x22$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.22$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.22$symbol <- tf.symbols
#with bigger matrix:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 -0.2352269 4.954686e-106 -0.4678254 376.5201 -0.194942574 -0.3154342 NA AHL20
# 11 AT5G62165 -0.1605824 4.050781e-102 -0.4629587 339.9943 -0.142173176 -0.3505407 NA AGL42
# 10 AT5G60200 -0.1010548 2.685066e-24 -0.3526809 175.9870 -0.122245187 -0.3010236 NA DOF5.3
# 12 AT5G65590 0.1221619 8.165302e-37 0.3390436 171.3479 0.119279678 0.3121283 NA AT5G65590 aka SCAP1
# 2 AT1G52150 0.0000000 7.388956e-02 -0.2542754 155.6862 -0.055204319 -0.2906103 NA ATHB-15
# 4 AT2G22430 0.2062901 8.318967e-26 0.2863719 125.5836 0.224416183 0.2595819 NA ATHB-6
# 3 AT1G63480 -0.1759239 1.151318e-17 -0.1640331 114.9688 -0.167926041 -0.1373603 NA AT1G63480
# 6 AT2G46590 0.0000000 3.175261e-01 -0.1714005 112.2792 0.007433777 -0.1765432 NA DAG2
# 9 AT4G40060 0.0000000 5.794193e-01 -0.1947899 103.9999 -0.049056939 -0.2129665 NA ATHB-16
# 7 AT3G01470 0.0000000 1.467616e-01 -0.1471854 100.7755 0.026399529 -0.1529602 NA ATHB-1
#with smaller matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT4G40060 0.38676339 0.0003898992 0.4823011 3.436711 0.18352202 0.4773589 NA ATHB-16
# 12 AT5G39660 0.19433705 0.0006938689 0.4612668 2.995625 0.12643245 0.3785354 NA CDF2
# 1 AT1G47655 0.00000000 0.2856213499 0.2405810 2.256596 0.03568010 0.2492677 NA AT1G47655
# 9 AT4G14465 0.00000000 0.2353579376 0.2489781 1.789956 0.03119151 0.1388235 NA AHL20
# 13 AT5G60200 0.00000000 0.0452494962 0.1770031 1.516223 0.10228559 0.1669628 NA DOF5.3
# 7 AT3G01470 0.00000000 0.4349516377 0.1756762 1.469617 0.04019154 0.1840576 NA ATHB-1
# 8 AT3G06740 0.02947161 0.0326759802 0.2637456 1.373348 0.13936260 0.2711710 NA GATA15
# 4 AT1G63480 0.00000000 0.2213847405 -0.3395003 1.337998 -0.07550148 -0.2319808 NA AT1G63480
# 14 AT5G62165 0.00000000 0.0541638997 -0.3317078 1.224800 -0.03967390 -0.1822329 NA AGL42
# 11 AT5G13180 0.00000000 0.1539337710 0.1325205 1.158471 0.06127619 0.1330612 NA ANAC083
#----------------------------------------------------------------------------------------
# 23rd model: 29 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "BEE2")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(17241698,17243699),
end=c(17243698,17244199),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="BEE2",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x23 <- build(builder)
lapply(x23, dim)
tbl.model.23 <- x23$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.23$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.23$symbol <- tf.symbols
# with bigger matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 15 AT5G62165 -0.14654224 8.781013e-62 -0.3672063 386.6629 -0.11614451 -0.28922974 NA AGL42
# 9 AT4G31920 -0.32184470 8.196087e-52 -0.3513091 343.8435 -0.28954908 -0.32828247 NA ARR10
# 1 AT1G53160 0.14257230 3.025479e-64 0.3721841 309.8696 0.12612429 0.40741137 NA SPL4
# 4 AT2G02540 0.16268244 1.387200e-67 0.3798380 261.2975 0.14506505 0.36150086 NA ATHB21
# 14 AT5G60200 0.00000000 5.376887e-02 -0.1971474 205.7040 -0.04499575 -0.25056895 NA DOF5.3
# 3 AT1G69780 0.19690706 1.589081e-17 0.1621895 201.0077 0.18118475 0.15222777 NA ATHB13
# 17 AT5G65310 0.27520837 4.586593e-20 0.1561768 173.0371 0.23328707 0.17642863 NA ATHB-5
# 10 AT4G32890 -0.13829996 7.193778e-26 -0.2485245 159.3232 -0.12355251 -0.25224104 NA GATA9
# 11 AT5G16560 -0.04288807 2.020327e-07 -0.1283945 148.8646 -0.06416597 -0.08825239 NA KAN
# 13 AT5G49520 -0.06936169 1.606297e-13 -0.2128161 148.5657 -0.08714175 -0.27128479 NA WRKY48
#with smaller matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G69780 0.475263698 9.071561e-12 0.7900089 6.4627143 0.14877009 0.6877791 NA ATHB13
# 12 AT4G31920 0.000000000 4.422050e-01 -0.2539478 1.7797840 -0.07590570 -0.2144058 NA ARR10
# 16 AT5G60200 0.000000000 6.627797e-01 -0.3210403 1.5411595 -0.06501800 -0.2618487 NA DOF5.3
# 14 AT5G16560 0.009221995 4.351465e-03 0.4371744 1.5307764 0.04171660 0.5045378 NA KAN
# 6 AT2G22430 0.000000000 8.916344e-01 -0.5026041 1.4761931 -0.11028331 -0.4729412 NA ATHB-6
# 3 AT1G63480 0.012537756 5.041249e-03 0.1390710 1.3466403 0.06012580 0.2322689 NA AT1G63480
# 5 AT1G80840 0.000000000 8.335613e-01 -0.4675158 1.1657589 -0.03904494 -0.3576951 NA WRKY40
# 7 AT2G46830 0.000000000 8.288264e-01 -0.2435563 0.9598848 -0.01694524 -0.1338295 NA CCA1
# 8 AT3G01470 0.000000000 6.096893e-01 0.3443222 0.7905707 0.04480747 0.3185114 NA ATHB-1
# 10 AT3G15270 0.000000000 5.967193e-01 -0.3595525 0.7188257 -0.02229638 -0.1956783 NA SPL5
#----------------------------------------------------------------------------------------
# 24th model: 47 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AGL42")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(24963074,24965075),
end=c(24965074,24967075),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AGL42",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x24 <- build(builder)
lapply(x24, dim)
tbl.model.24 <- x24$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.24$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.24$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT3G54810 0.00000000 0.084785117 -0.2644196 6.188577 -0.146332585 -0.200000000 NA GATA8
# 10 AT3G15270 0.07540106 0.006998546 0.3772227 4.869652 0.069474696 0.198559424 NA SPL5
# 24 AT5G65590 0.00000000 0.518332449 -0.1392564 4.552919 -0.018034682 -0.325138055 NA AT5G65590
# 7 AT2G37590 0.00000000 0.058326562 0.2887062 3.853160 0.223484191 0.307274910 NA DOF2.4
# 1 AT1G47655 0.00000000 0.179587983 -0.2823909 3.703040 -0.053497594 -0.145450180 NA AT1G47655
# 20 AT5G18830 0.00000000 0.122141996 -0.2049080 3.656628 -0.178760199 -0.091764706 NA SPL7
# 8 AT2G46130 0.00000000 0.268403642 0.1314209 3.474132 0.019499434 -0.006386555 NA WRKY43
# 17 AT4G40060 -0.38767399 0.002192146 -0.4233002 3.322573 -0.190071649 -0.348283313 NA ATHB-16
# 12 AT4G14465 0.00000000 0.824461667 -0.2942996 3.096320 -0.032989690 -0.171668667 NA AHL20
# 5 AT1G69780 0.00000000 0.480709469 -0.2050077 2.786546 -0.004923472 0.056902761 NA ATHB13
#----------------------------------------------------------------------------------------
# 25th model: 49 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "SPL7")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(6274114,6276115),
end=c(6276114,6278115),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="SPL7",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x25 <- build(builder)
lapply(x25, dim)
tbl.model.25 <- x25$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.25$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.25$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 18 AT4G34000 0.077602497 0.000558144 0.4708817 0.4809283 0.040375745 0.47716687 NA ABF3
# 10 AT2G46830 0.000000000 0.276421291 0.2673371 0.4729973 0.009602261 0.33474190 NA CCA1
# 30 AT5G65590 0.000000000 0.034660547 -0.2762644 0.4410077 -0.018100695 -0.22487395 NA AT5G65590
# 8 AT2G46590 0.005972524 0.023016935 0.2930937 0.3572919 0.022259839 -0.02722689 NA DAG2
# 21 AT5G03790 0.000000000 0.037914342 0.2938346 0.2258425 0.016411370 0.30919568 NA ATHB-51
# 11 AT3G01220 0.000000000 0.784879006 -0.1079931 0.2041922 -0.001507687 -0.08657863 NA ATHB20
#----------------------------------------------------------------------------------------
# 26th model: 42 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB20")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(71598,73599),
end=c(73598,74099),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB20",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x26 <- build(builder)
lapply(x26, dim)
tbl.model.26 <- x26$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.26$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.26$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT4G31920 0.41772711 3.406814e-101 0.4580854 381.0669 0.34747429 0.4738247 NA ARR10
# 14 AT5G25830 0.19961820 1.318223e-92 0.4445244 363.0023 0.17044102 0.4215165 NA GATA12
# 8 AT4G16110 0.43145786 7.786821e-97 0.4506042 361.5957 0.35814863 0.4004716 NA ARR2
# 3 AT1G69780 0.38328304 3.096823e-74 0.3917870 241.0333 0.29992065 0.3124582 NA ATHB13
# 4 AT2G22430 -0.33991092 1.593083e-39 -0.3110957 123.9119 -0.27593860 -0.2621734 NA ATHB-6
# 11 AT4G40060 0.16606026 1.371820e-26 0.2986074 121.7611 0.16935473 0.3205553 NA ATHB-16
# 20 AT5G60200 0.06673388 5.472103e-20 0.2569738 118.5619 0.06216835 0.2385958 NA DOF5.3
# 1 AT1G52150 0.00000000 9.359605e-01 0.2043231 112.4625 0.02076009 0.2827836 NA ATHB-15
# 19 AT5G56270 -0.11981506 3.764489e-21 -0.2444904 105.9166 -0.15869759 -0.2278204 NA WRKY2
# 12 AT5G12870 0.01968777 2.114678e-11 0.2610498 104.6406 0.04900232 0.2855275 NA MYB46
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 19 AT5G62000 -0.2862536 0.000191864 -0.5037641 13.112046 -0.11058991 -0.49848739 NA ARF2
# 8 AT2G23320 0.0000000 0.454605160 0.1141251 8.021538 0.09733246 0.06160864 NA WRKY15
# 4 AT1G64620 0.0000000 0.156291095 0.2507090 6.958213 0.17797144 0.17887155 NA AT1G64620
# 9 AT2G37590 0.1879416 0.017642741 0.3330993 6.609022 0.34492495 0.33618247 NA DOF2.4
# 7 AT2G22430 0.0000000 0.670812321 -0.3491920 6.598530 -0.10096907 -0.38477791 NA ATHB-6
# 18 AT5G60200 0.0000000 0.681120586 -0.2310342 4.441324 -0.10450002 -0.31130852 NA DOF5.3
# 13 AT4G37180 0.0000000 0.955245566 -0.1257275 4.421769 -0.01297474 -0.11769508 NA AT4G37180
# 2 AT1G08010 0.0000000 0.711167545 0.2947692 3.809657 0.10327200 0.28384154 NA GATA11
# 5 AT1G69780 0.0000000 0.200130094 0.3239425 3.625423 0.08096407 0.24504202 NA ATHB13
# 3 AT1G52150 0.0000000 0.779012361 -0.2248019 3.554240 -0.10040755 -0.24504202 NA ATHB-15
#----------------------------------------------------------------------------------------
# 27th model: 43 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB-15")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(19407912,19409913),
end=c(19409912,19410413),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB-15",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x27 <- build(builder)
lapply(x27, dim)
tbl.model.27 <- x27$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.27$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.27$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 12 AT5G60200 0.21340936 2.382589e-140 0.5291238 180.25009 0.15742201 0.5104056 NA DOF5.3
# 11 AT5G59780 0.12134591 9.156398e-89 0.4541901 134.24507 0.10896906 0.5001797 NA MYB59
# 9 AT5G12870 0.10519445 1.720585e-64 0.4073818 113.38724 0.09952470 0.3741304 NA MYB46
# 3 AT1G64620 0.17991695 1.974772e-73 0.4309093 104.51558 0.17044489 0.3959355 NA AT1G64620
# 6 AT2G37590 0.05159066 9.584809e-12 0.3314857 79.61398 0.09905095 0.3236278 NA DOF2.4
# 1 AT1G19850 0.05712985 7.918163e-11 0.2524720 57.30713 0.07933547 0.2915014 NA ARF5
# 7 AT3G01470 0.00000000 9.624542e-02 0.1885680 52.84702 -0.02167611 0.3273231 NA ATHB-1
# 15 AT5G65310 0.02036068 2.148224e-03 0.1811693 49.68740 0.07261525 0.2085411 NA ATHB-5
# 13 AT5G62165 0.00000000 9.062988e-01 0.2863012 42.93329 0.01439715 0.3785580 NA AGL42
# 4 AT1G80840 -0.03598329 1.281630e-13 -0.1940822 39.17065 -0.04269959 -0.2079710 NA WRKY40
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 20 AT5G62940 0.10801195 0.0006322369 0.4679888 0.8921942 0.065451480 0.135846339 NA DOF5.6
# 1 AT1G19850 0.00000000 0.0500041441 0.2097378 0.8527553 0.040026434 0.374597839 NA ARF5
# 12 AT4G17460 0.00000000 0.6630970577 -0.1509035 0.7576329 -0.002103981 0.044321729 NA HAT1
# 16 AT5G40350 0.04182587 0.0018847096 0.4204592 0.7557620 0.028697198 0.386506603 NA MYB24
# 21 AT5G65310 0.06294694 0.0005086301 0.4738772 0.6652933 0.042665227 0.151500600 NA ATHB-5
# 4 AT2G22430 0.00000000 0.7015060035 0.2227988 0.4832946 0.016838179 0.122785114 NA ATHB-6
# 8 AT3G01470 0.00000000 0.8829875080 -0.1533313 0.3471583 -0.014339995 0.001776711 NA ATHB-1
# 5 AT2G42200 0.00000000 0.8771819918 0.1964781 0.3297144 0.012000219 0.330804322 NA SPL9
# 10 AT3G15500 0.00000000 0.1752398645 -0.1873589 0.3038921 -0.014458599 -0.170900360 NA NAC055
# 2 AT1G53160 0.00000000 0.6867735006 0.1634598 0.2932131 0.016955481 0.004945978 NA SPL4
#----------------------------------------------------------------------------------------
# 28th model: 72 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB6")
tbl.atac <- data.frame(chrom=rep("Chr2", 2),
start=c(9522469,9526470),
end=c(9526469,9526970),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB6",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x28 <- build(builder)
lapply(x28, dim)
tbl.model.28 <- x28$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.28$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.28$symbol <- tf.symbols
#with bigger matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 26 AT5G25830 -0.09483151 1.140345e-60 -0.3607472 43.16806 -0.06685793 -0.33905727 NA GATA12
# 11 AT3G01470 0.22466705 1.211813e-22 0.1711951 42.22054 0.17356091 0.15181251 NA ATHB-1
# 17 AT3G59060 0.01548082 7.926326e-40 0.3216937 34.07054 0.01428551 0.24764231 NA PIF5
# 19 AT4G14465 -0.01061005 7.291409e-04 -0.1975364 28.95192 -0.01726298 -0.06137936 NA AHL20
# 27 AT5G39660 -0.12605465 1.095966e-23 -0.2464525 24.99268 -0.10654957 -0.21454257 NA CDF2
# 5 AT1G53170 0.05656884 4.104794e-15 0.2205160 24.07504 0.04516757 0.22032783 NA ERF8
# 3 AT1G50640 0.03543838 9.952316e-13 0.1508054 21.46977 0.05594105 0.11563041 NA ERF3
# 1 AT1G25550 -0.09614162 2.697151e-09 -0.1148429 21.07468 -0.10490239 -0.06847917 NA AT1G25550
# 30 AT5G62165 -0.01626665 2.727340e-15 -0.2656049 20.32351 -0.01924904 -0.24787569 NA AGL42
# 24 AT4G37180 0.03308813 3.732311e-08 0.1568264 16.61269 0.04429686 0.16640427 NA AT4G37180
#with smaller matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 25 AT3G59060 0.00000000 6.013292e-03 0.4177917 1.4925109 0.04271919 0.3814166 NA PIF5
# 2 AT1G06160 0.00000000 1.376525e-02 -0.3846738 1.4547389 -0.01850690 -0.5465066 NA ERF094
# 16 AT2G37590 -0.25457628 1.593588e-04 -0.5006159 0.6920075 -0.13581099 -0.3996639 NA DOF2.4
# 32 AT5G16560 0.00000000 4.887221e-01 -0.4302894 0.5912776 -0.01893960 -0.4443217 NA KAN
# 9 AT1G53170 0.03921353 5.176469e-04 0.5012652 0.5717675 0.03725612 0.5334454 NA ERF8
# 5 AT1G28370 0.02500702 5.105083e-05 0.5403776 0.5613470 0.01986683 0.4971429 NA ERF11
# 27 AT4G31800 0.04091388 3.129935e-04 0.5047127 0.5010410 0.02207024 0.4797599 NA WRKY18
# 28 AT4G36540 -0.05821874 3.747159e-04 -0.5026041 0.4860311 -0.04020065 -0.4729412 NA BEE2
# 34 AT5G25830 0.00000000 6.219192e-03 -0.3594084 0.4298235 -0.02886218 -0.4058103 NA GATA12
# 23 AT3G23240 0.00000000 1.194863e-01 0.4357799 0.3651550 0.02055299 0.4164706 NA ERF1B
#----------------------------------------------------------------------------------------
# 29th model: 50 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "GATA8")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20294956,20296957),
end=c(20296956,20298957),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="GATA8",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x29 <- build(builder)
lapply(x29, dim)
tbl.model.29 <- x29$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.29$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.29$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 19 AT4G37260 0.1625935 8.031956e-07 0.6331905 1.4551616 0.07785364 0.5953902 NA MYB73
# 18 AT4G36930 0.1319216 5.388445e-05 0.5527549 1.1165373 0.07551803 0.5863625 NA SPT
# 27 AT5G62165 0.0000000 2.469158e-02 -0.2644196 0.9213861 -0.02436430 -0.2000000 NA AGL42
# 28 AT5G62940 0.0000000 4.853429e-01 -0.2108329 0.8014314 -0.01950336 -0.1296038 NA DOF5.6
# 21 AT5G16560 0.0000000 2.791695e-02 0.2876827 0.5131881 0.02265711 0.2155582 NA KAN
# 15 AT4G14465 0.0000000 2.568635e-01 0.4068371 0.5001389 0.02587057 0.3986074 NA AHL20
#----------------------------------------------------------------------------------------
# 30th model: 27 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "GATA15")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(2124658,2126659),
end=c(2126658,2127159),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="GATA15",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x30 <- build(builder)
lapply(x30, dim)
tbl.model.30 <- x30$model[1:9,]
tf.symbols <- unlist(lapply(tbl.model.30$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.30$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 1 AT1G42990 -0.33508042 1.351745e-54 -0.3428298 136.45933 -0.28399798 -0.33740759 NA BZIP60
# 3 AT4G16110 -0.11672350 3.434959e-07 -0.1189034 135.80445 -0.11646256 -0.11692962 NA ARR2
# 2 AT2G22430 -0.10248342 1.859476e-08 -0.1354756 101.82278 -0.10607455 -0.09941991 NA ATHB-6
# 4 AT4G31920 0.07838795 4.844631e-05 0.1150265 93.92049 0.08577908 0.10381406 NA ARR10
# 5 AT5G39660 -0.04991222 3.873510e-04 -0.1007136 93.64231 -0.06195625 -0.12276830 NA CDF2
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G80840 -0.1237662 1.995854e-05 -0.5639998 1.7139994 -0.05265239 -0.5307690 NA WRKY40
# 7 AT5G49520 0.0000000 8.609518e-01 -0.3637126 1.1599551 -0.02557242 -0.3389277 NA WRKY48
# 5 AT2G22430 0.0000000 8.372288e-01 -0.2135799 0.9022090 -0.02643684 -0.2063435 NA ATHB-6
# 1 AT1G42990 0.0000000 3.054324e-01 -0.3711281 0.8352476 -0.05579421 -0.3108358 NA BZIP60
# 9 AT5G65590 0.0000000 2.991368e-01 -0.1733936 0.6920649 -0.01608158 -0.1782996 NA AT5G65590
# 8 AT5G60200 0.0000000 8.251771e-01 -0.1577859 0.5924579 -0.02278387 -0.1964513 NA DOF5.3
# 2 AT1G49560 0.0000000 7.170586e-01 -0.1130704 0.5691463 -0.02138932 -0.1672069 NA AT1G49560
# 3 AT1G77920 0.0000000 3.590363e-01 -0.1727587 0.5415838 -0.02227929 -0.1709525 NA AT1G77920
# 6 AT5G39660 0.0000000 6.216311e-01 -0.1109419 0.4314407 -0.01836529 -0.1417081 NA CDF2
#----------------------------------------------------------------------------------------
# 31st model: 29 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "DOF5.3")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(24239077,24241078),
end=c(24241077,24241578),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="DOF5.3",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x31 <- build(builder)
lapply(x31, dim)
tbl.model.31 <- x31$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.31$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.31$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 0.214155150 4.916909e-114 0.4832782 239.50481 0.15668213 0.44778587 NA AHL20
# 6 AT2G37590 0.357741795 1.568545e-90 0.4384045 188.61084 0.31449635 0.42497559 NA DOF2.4
# 2 AT1G64620 0.211534579 1.091095e-58 0.3815155 165.40192 0.19713607 0.41092741 NA AT1G64620
# 11 AT5G62165 0.070304535 7.690478e-39 0.3851670 130.60854 0.07760712 0.34801100 NA AGL42
# 4 AT2G18380 0.080615872 2.501859e-21 0.2916731 94.26407 0.09837057 0.28940058 NA GATA20
# 9 AT4G31920 0.000000000 6.329018e-02 0.1446562 78.76389 -0.04578754 0.20417106 NA ARR10
# 12 AT5G62940 0.072017247 3.750608e-13 0.2035832 77.73078 0.08849514 0.17916316 NA DOF5.6
# 10 AT5G25830 0.007394708 1.987848e-03 0.2516686 66.08657 0.04835927 0.26882677 NA GATA12
# 5 AT2G22430 0.000000000 1.158494e-01 -0.1197237 64.66402 0.03705229 -0.09882664 NA ATHB-6
# 3 AT1G80840 -0.001951971 1.292859e-02 -0.1299032 57.67901 -0.02536580 -0.13267635 NA WRKY40
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G80840 0.038016905 0.01034647 0.3594898 0.9927881 0.02919905 0.38986795 NA WRKY40
# 14 AT5G62940 0.051637742 0.01879464 0.3284058 0.9061619 0.05611126 0.25570228 NA DOF5.6
# 11 AT4G31920 0.000000000 0.64088629 -0.1142101 0.7730006 -0.02855464 -0.06228091 NA ARR10
# 2 AT1G64620 0.000000000 0.09435306 -0.2271095 0.6201841 -0.05787497 -0.16600240 NA AT1G64620
# 12 AT5G16560 0.000000000 0.23323122 -0.2672106 0.5990362 -0.01935544 -0.36211285 NA KAN
# 10 AT4G14465 0.000000000 0.87438075 -0.2027642 0.5986355 -0.01076157 -0.18943577 NA AHL20
# 9 AT3G15270 0.000000000 0.15020663 0.2614142 0.5646575 0.01586613 0.24081633 NA SPL5
# 13 AT5G62165 0.003856408 0.05513555 0.2602714 0.4449524 0.02247415 0.28172869 NA AGL42
# 8 AT3G06740 0.000000000 0.69841228 -0.1577859 0.4156874 -0.02077967 -0.19645130 NA GATA15
# 5 AT2G18380 0.000000000 0.35127576 0.1476492 0.4000534 0.01118307 0.02809124 NA GATA20
#----------------------------------------------------------------------------------------
# 32nd model: 47 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "DOF1.8")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(24005287,24007288),
end=c(24007287,24007788),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="DOF1.8",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x32 <- build(builder)
lapply(x32, dim)
tbl.model.32 <- x32$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.32$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.32$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT5G60200 0.187105963 3.251037e-68 0.3815155 122.88080 0.143461584 0.41092741 NA DOF5.3
# 6 AT2G37590 0.195499776 6.374630e-33 0.3108857 108.78777 0.191921494 0.30756576 NA DOF2.4
# 9 AT4G37790 0.065707743 8.327646e-10 0.1851465 70.06470 0.073489682 0.21960667 NA HAT22
# 12 AT5G62940 0.023270984 3.497178e-04 0.1522472 61.25121 0.041208829 0.15454709 NA DOF5.6
# 4 AT1G75080 0.000000000 5.510751e-01 -0.1180820 59.31977 0.004938775 -0.09912931 NA BZR1
# 2 AT1G50640 0.020920599 4.737249e-03 0.1408610 57.08185 0.051081277 0.18047068 NA ERF3
# 7 AT4G14465 0.009766064 3.472301e-06 0.2410530 55.09684 0.029934282 0.27975820 NA AHL20
# 8 AT4G16110 0.000000000 6.983879e-01 0.1330114 50.03913 0.013573475 0.21742204 NA ARR2
# 5 AT2G23340 0.002045172 5.886952e-02 0.1147439 46.57284 0.024416134 0.12424610 NA DEAR3
# 3 AT1G53170 -0.032093620 1.722291e-05 -0.1726655 44.97815 -0.040581825 -0.17985127 NA ERF8
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT1G75080 0.000000000 0.07045807 0.2569291 0.8972444 0.065996112 0.26530612 NA BZR1
# 18 AT5G65210 0.000000000 0.06127087 0.2385765 0.5332703 0.045748444 0.19788715 NA TGA1
# 15 AT5G43410 0.008428567 0.02544334 0.3191712 0.5219050 0.022891551 0.28931573 NA AT5G43410
# 8 AT2G22430 0.000000000 0.05771801 -0.2717833 0.5065601 -0.038136257 -0.23697479 NA ATHB-6
# 12 AT3G23230 0.007146285 0.02207873 0.3231495 0.4882881 0.015521526 0.32235294 NA AT3G23230
# 2 AT1G19210 0.000000000 0.19981887 -0.1720897 0.3969417 -0.014377444 -0.09147659 NA AT1G19210
# 17 AT5G60200 0.000000000 0.09561391 -0.2271095 0.3778582 -0.047719742 -0.16600240 NA DOF5.3
# 1 AT1G06160 0.000000000 0.67509515 0.1302599 0.3752255 0.003923286 0.06055222 NA ERF094
# 14 AT4G37790 0.000000000 0.09713631 0.2282825 0.3720663 0.066419448 0.26492197 NA HAT22
# 13 AT4G16750 -0.016138806 0.02399706 -0.3204023 0.3581817 -0.026578976 -0.34588235 NA AT4G16750
#----------------------------------------------------------------------------------------
# 33rd model: 35 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AT5G65590")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(26210083,26212084),
end=c(26212083,26214084),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AT5G65590",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x33 <- build(builder)
lapply(x33, dim)
tbl.model.33 <- x33$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.33$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.33$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT2G01760 0.4033589 0.0005560655 0.4710025 9.069182 0.19982640 0.387851140 NA ARR14
# 21 AT5G02460 0.0000000 0.1215603958 0.1929624 6.025120 0.06524601 0.006194478 NA AT5G02460
# 14 AT3G50410 0.1970301 0.0043926087 0.3994689 5.597807 0.16865208 0.348955582 NA OBP1
# 16 AT4G16750 0.0000000 0.6450941741 0.2044169 5.379886 0.03300583 0.279231693 NA AT4G16750
# 9 AT2G42200 0.0000000 0.6635019982 0.2290212 4.347205 0.04489868 0.158799520 NA SPL9
# 15 AT4G14465 0.0000000 0.4708751326 0.1463848 4.095072 0.03802711 0.289699880 NA AHL20
#----------------------------------------------------------------------------------------
# 34th model: 28 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "MYC4")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(9931701,9933702),
end=c(9933701,9934202),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="MYC4",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x34 <- build(builder)
lapply(x34, dim)
tbl.model.34 <- x34$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.34$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.34$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT5G62000 0.03293216 2.059567e-05 0.1116754 95.07190 0.04064757 0.05220811 NA ARF2
# 3 AT3G47500 0.10271069 8.896262e-15 0.1744021 88.43036 0.09811551 0.15830419 NA CDF3
# 2 AT3G06740 0.10785022 2.374480e-09 0.1340536 88.02321 0.10871368 0.11134859 NA GATA15
# 5 AT5G39660 0.09237757 1.068600e-12 0.1690606 87.22162 0.09327233 0.12990382 NA CDF2
# 4 AT4G14465 0.07248091 3.511876e-15 0.1774471 84.26109 0.06375901 0.20562914 NA AHL20
# 7 AT5G62165 0.00709379 2.461137e-02 0.1026428 83.80349 0.01739119 0.12144408 NA AGL42
# 1 AT1G49560 0.10077876 1.331365e-10 0.1442764 75.67232 0.09751514 0.13397775 NA AT1G49560
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT3G06740 0.40927216 0.0002450054 0.4965225 3.209359 0.21101236 0.5101203 NA GATA15
# 6 AT3G47500 0.10391120 0.0138938349 0.3601382 2.830953 0.11856463 0.4494118 NA CDF3
# 3 AT2G22430 -0.13576117 0.0131700679 -0.3625533 2.570702 -0.10784324 -0.2595438 NA ATHB-6
# 8 AT5G16560 0.00000000 0.0756487998 0.3501211 2.274969 0.04497062 0.3777671 NA KAN
# 9 AT5G39660 0.09777027 0.0079118978 0.3307048 2.254376 0.07885702 0.1379592 NA CDF2
# 11 AT5G62940 0.00000000 0.1484288502 0.1361369 1.653449 0.05301587 0.1466026 NA DOF5.6
# 2 AT1G64620 0.00000000 0.2869265118 0.2155694 1.575104 0.07660245 0.1059784 NA AT1G64620
# 4 AT2G37590 0.00000000 0.1679122457 0.2146886 1.373392 0.10137605 0.1597599 NA DOF2.4
# 10 AT5G60200 -0.01546326 0.0570073327 -0.2469768 1.316233 -0.11839511 -0.1868427 NA DOF5.3
# 1 AT1G49560 0.00000000 0.7869961954 -0.2202242 1.166599 -0.05085925 -0.2133493 NA AT1G49560
#----------------------------------------------------------------------------------------
# 35th model: 48 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AT1G25550")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(8974643,8976644),
end=c(8976643,8977143),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AT1G25550",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x35 <- build(builder)
lapply(x35, dim)
tbl.model.35 <- x35$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.35$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.35$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G37180 0.1931458598 1.021758e-48 0.3242890 94.59979 0.163751975 0.30083819 NA AT4G37180
# 3 AT1G80840 0.1066981489 4.115582e-43 0.3056062 82.41714 0.088806587 0.30201622 NA WRKY40
# 4 AT2G22430 -0.1848360668 3.291719e-12 -0.1148429 69.93445 -0.168051347 -0.06847917 NA ATHB-6
# 7 AT4G36540 0.0522095676 4.995605e-09 0.1523035 60.67394 0.054368338 0.16142347 NA BEE2
# 6 AT3G06740 -0.0842381150 5.731672e-09 -0.1691969 55.23117 -0.089748937 -0.15614525 NA GATA15
# 9 AT4G38620 0.0151246394 2.285397e-05 0.1541935 45.27454 0.024560415 0.11727319 NA MYB4
# 5 AT2G44840 0.0000000000 4.091548e-01 0.1630523 43.35259 0.002397413 0.14819748 NA ERF13
# 10 AT5G49520 0.0161834476 1.974998e-03 0.1682330 42.05875 0.030072592 0.16698822 NA WRKY48
# 1 AT1G08000 -0.0340783357 2.511792e-05 -0.1266205 38.88968 -0.028013797 -0.15010301 NA GATA10
# 2 AT1G08010 -0.0009570583 1.000000e+00 -0.1266205 38.43543 -0.027596918 -0.15010301 NA GATA11
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 9 AT3G06740 -0.299907577 6.439146e-05 -0.5342536 1.8577021 -0.15109740 -0.5212130 NA GATA15
# 4 AT1G80840 0.060545281 4.300083e-04 0.5025548 1.7573561 0.04383603 0.5300840 NA WRKY40
# 7 AT2G44840 0.000000000 8.838719e-01 0.3936641 1.1996919 0.02183806 0.4226170 NA ERF13
# 21 AT5G16560 0.000000000 2.603671e-01 -0.2740614 1.1184633 -0.02773111 -0.3031453 NA KAN
# 11 AT3G54810 0.009909955 2.564185e-02 0.2725065 0.9697879 0.08001210 0.2621369 NA GATA8
# 13 AT4G00050 0.000000000 8.322380e-01 -0.1452447 0.9422956 -0.01770847 -0.1826170 NA UNE10
# 12 AT3G59060 0.000000000 9.752550e-01 0.1444829 0.8469312 0.01177003 0.2466747 NA PIF5
# 14 AT4G14465 0.000000000 1.443136e-01 0.1358985 0.8118220 0.02571154 0.1400720 NA AHL20
# 5 AT2G01760 0.000000000 8.058925e-01 0.1600451 0.7753703 0.03482297 0.2307323 NA ARR14
# 22 AT5G42630 0.013153562 1.765845e-02 0.2956409 0.7692770 0.02849112 0.2895078 NA ATS
#----------------------------------------------------------------------------------------
# 36th model: 47 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "WRKY40")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(30381833,30383834),
end=c(30383833,30384334),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WRKY40",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x36 <- build(builder)
lapply(x36, dim)
tbl.model.36 <- x36$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.36$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.36$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT4G31800 0.51525632 4.964320e-198 0.6100412 702.2947 0.38318834 0.5895222 NA WRKY18
# 11 AT5G49520 0.20422611 4.091283e-24 0.3789303 312.8799 0.21953091 0.3569045 NA WRKY48
# 5 AT4G14465 -0.16862619 5.477636e-25 -0.2222530 157.1492 -0.12068589 -0.1928905 NA AHL20
# 8 AT4G40060 -0.08719258 2.176891e-11 -0.1802024 142.2125 -0.11461538 -0.1942995 NA ATHB-16
# 1 AT1G52150 -0.08624685 1.354613e-16 -0.1940822 138.6247 -0.10105451 -0.2079710 NA ATHB-15
# 7 AT4G31920 -0.02157285 2.252443e-03 -0.2124314 126.4829 -0.07640374 -0.2480439 NA ARR10
# 4 AT3G54810 0.05375731 6.408844e-07 0.1900307 121.3865 0.11027880 0.1646632 NA GATA8
# 12 AT5G60200 -0.04284976 4.494388e-07 -0.1299032 116.7443 -0.06072735 -0.1326763 NA DOF5.3
# 3 AT3G01470 0.00000000 4.510046e-02 -0.2207950 112.6439 -0.08703087 -0.2035526 NA ATHB-1
# 13 AT5G62165 0.00000000 2.372011e-01 -0.1474643 107.6476 -0.01547067 -0.1469365 NA AGL42
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 17 AT5G49520 0.30946554 2.050576e-08 0.6956844 16.702504 0.15112743 0.6376471 NA WRKY48
# 11 AT4G31800 0.14937448 7.223464e-06 0.6214864 8.312755 0.10574067 0.5374790 NA WRKY18
# 7 AT3G06740 -0.42878420 1.995821e-05 -0.5639998 5.991193 -0.31067054 -0.5307690 NA GATA15
# 9 AT3G50410 0.00000000 1.557519e-02 0.3614604 4.713995 0.12785945 0.4128211 NA OBP1
# 15 AT5G16560 0.00000000 8.691337e-02 -0.4332895 3.753391 -0.05814861 -0.5171188 NA KAN
# 2 AT1G69780 -0.02171626 3.770456e-03 -0.4954645 2.945440 -0.10626775 -0.4257863 NA ATHB13
# 3 AT2G22430 0.00000000 7.659822e-02 0.4055689 2.699469 0.12850483 0.3892917 NA ATHB-6
# 16 AT5G39660 0.00000000 9.408914e-01 0.1232301 2.434100 0.04390803 0.2660744 NA CDF2
# 5 AT2G46830 0.00000000 5.002095e-01 0.1257317 1.782495 0.01610316 0.2545498 NA CCA1
# 19 AT5G60200 0.00000000 1.777865e-01 0.3594898 1.689535 0.13931685 0.3898679 NA DOF5.3
#----------------------------------------------------------------------------------------
# 37th model: 63 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "HDG1")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(22628768,22630769),
end=c(22630768,22631269),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="HDG1",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x37 <- build(builder)
lapply(x37, dim)
tbl.model.37 <- x37$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.37$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.37$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 15 AT4G14465 -0.154765390 2.596098e-115 -0.4856727 237.18354 -0.09040606 -0.4259938 NA AHL20
# 28 AT5G62165 -0.143733663 8.555141e-99 -0.4655235 194.10680 -0.10233955 -0.4206713 NA AGL42
# 4 AT1G53170 0.111083778 4.538903e-39 0.3885351 123.54931 0.10857248 0.3412866 NA ERF8
# 27 AT5G60200 -0.003436034 3.476019e-03 -0.3182049 94.32478 -0.03456264 -0.3386358 NA DOF5.3
# 14 AT3G16770 0.095642215 1.287627e-23 0.2962935 89.88464 0.09575802 0.3110551 NA RAP2-3
# 23 AT5G42630 0.000000000 4.896472e-01 -0.2929949 85.81616 -0.02404363 -0.2472149 NA ATS
# 2 AT1G43160 -0.038001593 1.805203e-35 -0.3635980 72.15085 -0.03897048 -0.3642711 NA RAP2-6
# 19 AT5G16560 0.098398516 1.032615e-20 0.2124910 70.89221 0.10234046 0.1777708 NA KAN
# 1 AT1G22985 -0.168688436 1.072297e-27 -0.2819869 65.52232 -0.13920117 -0.2649606 NA AT1G22985
# 26 AT5G59780 0.000000000 9.785459e-01 -0.3170906 61.38923 -0.02360199 -0.3422669 NA MYB59
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G19850 -0.15160166 0.0001122043 -0.5191075 2.7433970 -0.07520242 -0.6030732 NA ARF5
# 13 AT3G06740 0.04224990 0.0058084445 0.3901419 1.2922685 0.07388667 0.4125909 NA GATA15
# 5 AT1G28370 -0.02436093 0.0025452370 -0.4350029 0.8931289 -0.02253084 -0.4381753 NA ERF11
# 10 AT2G33810 0.00000000 0.2602239950 -0.2489882 0.6723020 -0.01894787 -0.3287875 NA SPL3
# 20 AT4G14465 0.00000000 0.4723964614 0.3278602 0.6388994 0.01115103 0.3110204 NA AHL20
# 25 AT5G16560 0.00530587 0.0085767546 0.3717312 0.5917463 0.02276504 0.2917167 NA KAN
# 19 AT3G54810 0.09148248 0.0025801359 0.4166885 0.4226205 0.08816604 0.4079232 NA GATA8
# 7 AT1G50640 0.00000000 0.0297311200 -0.1685020 0.3406988 -0.04450800 -0.1623529 NA ERF3
# 15 AT3G15270 0.00000000 0.6326714126 -0.2212659 0.3184976 -0.01069675 -0.1128932 NA SPL5
# 16 AT3G16770 0.00000000 0.0438706191 0.2069915 0.3161507 0.02201526 0.1376711 NA RAP2-3
#----------------------------------------------------------------------------------------
# 38th model: 39 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "OBP3")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20525197,20527198),
end=c(20527197,20527698),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="OBP3",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x38 <- build(builder)
lapply(x38, dim)
tbl.model.38 <- x38$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.38$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.38$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G52150 0.30032344 7.613967e-139 0.5288124 294.84890 0.21362574 0.52041400 NA ATHB-15
# 20 AT5G60200 0.19500629 1.186395e-142 0.5328210 278.27108 0.15273043 0.49603968 NA DOF5.3
# 13 AT4G14465 0.21858158 9.676539e-138 0.5271657 245.22635 0.12879812 0.49331909 NA AHL20
# 8 AT2G03500 0.13145435 3.818057e-36 0.3335143 98.22692 0.11723378 0.34131853 NA AT2G03500
# 21 AT5G62165 0.06357558 1.869604e-32 0.4095408 85.46927 0.06649715 0.42466574 NA AGL42
# 11 AT2G37590 0.06264759 1.633947e-13 0.3418685 70.43422 0.11443598 0.32495722 NA DOF2.4
# 19 AT5G42630 0.00000000 6.830392e-01 0.2720615 69.88591 0.01116839 0.29793409 NA ATS
# 17 AT5G16560 -0.02219210 2.446977e-05 -0.1551953 69.29002 -0.05366961 -0.08497416 NA KAN
# 4 AT1G64620 0.03786746 1.173939e-09 0.3421884 67.12839 0.08976062 0.33598918 NA AT1G64620
# 12 AT3G01470 0.00000000 6.418641e-01 0.2890155 66.15996 0.04800213 0.34502592 NA ATHB-1
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G70920 0.00000000 0.0549188851 -0.3220010 2.4308873 -0.02119084 -0.29114046 NA ATHB18
# 16 AT5G42630 0.00000000 0.2775032985 -0.2703003 1.5977141 -0.01979709 -0.08235294 NA ATS
# 6 AT2G01930 0.07720251 0.0096555402 0.3393510 1.5553011 0.06560244 0.24206483 NA BPC1
# 3 AT1G63480 -0.08890034 0.0027604541 -0.4203422 1.4083491 -0.08789321 -0.33714286 NA AT1G63480
# 2 AT1G52150 0.00000000 0.5393362014 0.1994989 1.2331230 0.04341727 -0.06151261 NA ATHB-15
# 11 AT4G17460 0.00000000 0.0360625159 -0.3557152 1.2322980 -0.04324252 -0.20364946 NA HAT1
# 14 AT4G37790 0.00000000 0.0431609962 -0.3080342 1.0922844 -0.10320802 -0.21286915 NA HAT22
# 20 AT5G65310 0.12564916 0.0005152703 0.4734606 0.9821801 0.06209034 0.14468187 NA ATHB-5
# 19 AT5G62940 0.00000000 0.1403350982 0.3512926 0.9386974 0.06213477 0.12297719 NA DOF5.6
# 7 AT2G03500 0.00000000 0.9089545184 0.1293064 0.7918994 0.01891004 0.17051621 NA AT2G03500
#----------------------------------------------------------------------------------------
# 39th model: 34 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "DOF2.4")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(15767292,15769293),
end=c(15769292,15771292),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="DOF2.4",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x39 <- build(builder)
lapply(x39, dim)
tbl.model.39 <- x39$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.39$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.39$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT2G22430 -0.151795953 0.000213524 -0.5006159 0.7358094 -0.074575421 -0.3996639 NA ATHB-6
# 1 AT1G24260 0.000000000 0.228304076 0.2310974 0.4672936 0.011216871 0.1334454 NA SEP3
# 9 AT3G47500 0.000000000 0.262294024 -0.2275964 0.4037608 -0.020777263 -0.2905642 NA CDF3
# 18 AT5G62165 0.002932115 0.031305813 0.2887062 0.3832981 0.018771658 0.3072749 NA AGL42
# 13 AT5G39610 0.000000000 0.154705994 -0.3126121 0.3506971 -0.019748291 -0.3306122 NA ATNAC2
# 6 AT2G46130 0.000000000 0.396256146 -0.1520034 0.3352320 -0.007884915 -0.1622569 NA WRKY43
#----------------------------------------------------------------------------------------
# 40th model: TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB13")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(26257165,26259166),
end=c(26259165,26261165),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB13",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x40 <- build(builder)
lapply(x40, dim)
tbl.model.40 <- x40$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.40$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.40$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT4G14465 0.19343639 9.251529e-08 0.6717750 6.812816 0.07842681 0.44105642 NA AHL20
# 12 AT4G31920 -0.02028084 8.951545e-03 -0.4447952 3.867082 -0.11947308 -0.21046819 NA ARR10
# 8 AT3G01470 0.19255800 2.463540e-04 0.5321812 2.562428 0.11928985 0.35750300 NA ATHB-1
# 18 AT5G60200 0.00000000 8.904937e-01 -0.3163195 1.875129 -0.06562144 -0.16686675 NA DOF5.3
# 5 AT2G22430 -0.01002657 1.165891e-02 -0.4936664 1.565940 -0.12507994 -0.62084034 NA ATHB-6
# 17 AT5G59780 0.00000000 5.213319e-01 0.3437988 1.378598 0.03966587 0.06314526 NA MYB59
#------------------------------------------------------------------------------------------------------------------------
if(!interactive())
runTests()
PriceLab/TrenaProjectArabidopsisRoot documentation built on Oct. 30, 2019, 10:42 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
library(GSEABase)
library(GOstats)
library(GO.db)
library(Category)
library(trena)
library(trenaSGM)
library(TrenaProjectArabidopsisRoot)
library(org.At.tair.db)
#------------------------------------------------------------------------------------------------------------------------
geneOntology <- function(orf)
{
tbl.go <- select(org.At.tair.db, keytype="TAIR", keys="AT3G59060", columns="GO")
return(select(GO.db, keys=tbl.go$GO, columns=c("TERM"))$TERM)
} # geneOntology
#------------------------------------------------------------------------------------------------------------------------
runTests <- function()
{
test_geneOntology()
test_goEnrichment()
} # runTests
#------------------------------------------------------------------------------------------------------------------------
test_geneOntology <- function()
{
message(sprintf("--- test_geneOntology"))
geneSymbol <- "PIF5"
orfName <- getGeneNames(tp, geneSymbol)$orf
tbl.go <- geneOntology(orfName)
} # test_geneOntology
#------------------------------------------------------------------------------------------------------------------------
goEnrichment <- function(orfNames, maxCategories)
{
gene.universe = character(0)
go.params <- new("GOHyperGParams", geneIds=orfNames,
universeGeneIds=gene.universe, annotation = "org.At.tair.db",
ontology = 'BP', pvalueCutoff = 0.01, conditional = FALSE,
testDirection = "over")
message(sprintf("about to calculate GO enrichment"))
go.bp.hgr <- hyperGTest(go.params)
tbl.go <- summary(go.bp.hgr)
if(nrow(tbl.go) > maxCategories)
tbl.go <- head(tbl.go, n=maxCategories)
message(sprintf("about to look up genes for each of %d GO categories", nrow(tbl.go)))
suppressMessages(
geneSymbols <- lapply(tbl.go$GOBPID,
function(goTerm){
if(!goTerm %in% keys(org.At.tair.db, keytype="GOALL")) {
printf("failed goTerm: %s", goTerm)
return ("")
}
keepers <- intersect(orfNames,select(org.At.tair.db, keys=goTerm, keytype="GOALL", columns="TAIR")$TAIR)
keeper.geneSymbols <- unlist(lapply(keepers, function(orf) getGeneNames(tp,orf)$symbol))
paste(keeper.geneSymbols, collapse=";")
}))
tbl.go$genes <- unlist(geneSymbols)
tbl.go
} # goEnrichment
#------------------------------------------------------------------------------------------------------------------------
test_goEnrichment <- function()
{
message(sprintf("--- test_goEnrichment"))
orfNames <- c("AT3G55130", "AT3G59060", "AT4G00050", "AT4G36540", "AT5G62165", "AT5G18830","AT5G60200",
"AT1G53160", "AT1G09530", "AT3G15270","AT1G80840")
tbl.enrichment <- goEnrichment(orfNames, maxCategories=5) # 10 is more typical
} # test_goEnrichment
#------------------------------------------------------------------------------------------------------------------------
if(!exists("tp")) {
message(sprintf("--- creating instance of TrenaProjectArabidopsisRoot"))
tp <- TrenaProjectArabidopsisRoot();
}
genome <- "tair10"
getExpressionMatrixNames(tp)
mtx <- getExpressionMatrix(tp, "aluru.18617x1938.orfRowNames")
targetGene <- canonicalizeName(tp, "WBC19")
#--------------------------------------------------------------------------
# first model: only 18 TFs with >= 95% motif match in short root-specific
# ATAC-seq about 2k upstream of WBC19
#--------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20438126,20438543),
end=c(20438246,20438984),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x <- build(builder)
lapply(x, dim)
tbl.model <- x$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model$symbol <- tf.symbols
# tbl.model:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT3G59060 0.25182969 7.586023e-257 0.6738150 1301.6266 0.17793969 0.5077358 NA PIF5
# 6 AT4G00050 0.25848974 1.790158e-129 0.5765029 738.1581 0.23987441 0.4916015 NA UNE10
# 7 AT4G36540 0.20257033 1.016042e-88 0.5333334 701.2759 0.19764529 0.4806764 NA BEE2
# 10 AT5G62165 -0.09238406 6.961638e-41 -0.4991596 513.0297 -0.11703297 -0.3849297 NA AGL42
# 9 AT5G60200 -0.10874331 4.796574e-21 -0.3877858 347.2974 -0.14396653 -0.3551172 NA DOF5.3
# 8 AT5G18830 -0.25172200 6.186848e-25 -0.3776184 331.5325 -0.28773203 -0.3622195 NA SPL7
# 2 AT1G53160 0.06447171 4.696022e-21 0.3917817 298.7136 0.09052672 0.3554472 NA SPL4
# 1 AT1G09530 0.01684827 1.433186e-05 0.4333553 256.6698 0.06233615 0.3551013 NA PIF3
# 4 AT3G15270 0.07932286 2.032701e-26 0.4072103 243.6807 0.10121717 0.3111859 NA SPL5
# 3 AT1G80840 0.00000000 9.856132e-01 0.1509693 175.8583 0.01323395 0.1306107 NA WRKY40
#------------------------------------------------------------------------------
# second model: 16 TFs with >= 95% motif match within 500kb (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20437000,20434000),
end=c(20437500,20434500),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x2 <- build(builder)
lapply(x2, dim)
tbl.model.2 <- x2$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.2$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.2$symbol <- tf.symbols
# tbl.model.2
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT4G14465 -0.46401927 1.919310e-162 -0.5628267 1331.0327 -0.35798325 -0.3313744 NA AHL20
# 10 AT5G62165 -0.26067798 8.382164e-104 -0.4991596 871.5585 -0.22696959 -0.3849297 NA AGL42
# 9 AT5G60200 -0.15703209 1.020370e-22 -0.3877858 573.3732 -0.18761945 -0.3551172 NA DOF5.3
# 6 AT5G13080 -0.13021371 1.794286e-21 -0.3233200 352.6772 -0.13125359 -0.2561005 NA WRKY75
# 7 AT5G46350 0.26386017 1.382743e-16 0.2225107 333.4381 0.25837985 0.1982006 NA WRKY8
# 4 AT2G46130 -0.06135211 2.192553e-12 -0.3191933 326.8847 -0.08656127 -0.2546452 NA WRKY43
# 8 AT5G49520 -0.05820226 6.901810e-05 -0.2456283 322.7049 -0.11555394 -0.2372729 NA WRKY48
# 1 AT1G64620 0.05390654 1.121712e-02 -0.1025043 269.3130 0.10217146 -0.1231658 NA AT1G64620
# 3 AT2G37590 0.00000000 1.613487e-01 -0.2068212 250.8909 -0.09618521 -0.1980168 NA DOF2.4
# 2 AT1G80840 0.02794550 1.308692e-02 0.1509693 242.7149 0.08498833 0.1306107 NA WRKY40
#------------------------------------------------------------------------------
# third model: 36 TFs with >= 95% motif match within 1000kb (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20434500,20436000),
end=c(20435500,20437000),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x3 <- build(builder)
lapply(x3, dim)
tbl.model.3 <- x2$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.3$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.3$symbol <- tf.symbols
# tbl.model.3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT4G14465 -0.46401927 1.919310e-162 -0.5628267 1331.0327 -0.35798325 -0.3313744 NA AHL20
# 10 AT5G62165 -0.26067798 8.382164e-104 -0.4991596 871.5585 -0.22696959 -0.3849297 NA AGL42
# 9 AT5G60200 -0.15703209 1.020370e-22 -0.3877858 573.3732 -0.18761945 -0.3551172 NA DOF5.3
# 6 AT5G13080 -0.13021371 1.794286e-21 -0.3233200 352.6772 -0.13125359 -0.2561005 NA WRKY75
# 7 AT5G46350 0.26386017 1.382743e-16 0.2225107 333.4381 0.25837985 0.1982006 NA WRKY8
# 4 AT2G46130 -0.06135211 2.192553e-12 -0.3191933 326.8847 -0.08656127 -0.2546452 NA WRKY43
# 8 AT5G49520 -0.05820226 6.901810e-05 -0.2456283 322.7049 -0.11555394 -0.2372729 NA WRKY48
# 1 AT1G64620 0.05390654 1.121712e-02 -0.1025043 269.3130 0.10217146 -0.1231658 NA AT1G64620
# 3 AT2G37590 0.00000000 1.613487e-01 -0.2068212 250.8909 -0.09618521 -0.1980168 NA DOF2.4
# 2 AT1G80840 0.02794550 1.308692e-02 0.1509693 242.7149 0.08498833 0.1306107 NA WRKY40
# side note: AGL42 is within the top 5 among all 3 models (repressor)
# another note: AGL40 is the first one among models 2 & 3
# make sure to review the others as well
#------------------------------------------------------------------------------
# fourth model: 19 TFs with >= 95% motif match within 1200kb (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432000,20433210),
end=c(20433200,20434410),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
#side note: next model, go back to around the third model's region.
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x4 <- build(builder)
lapply(x4, dim)
tbl.model.4 <- x4$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.4$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.4$symbol <- tf.symbols
# tbl.model.4
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 7 AT4G14465 -0.63306716 1.919310e-162 -0.5628267 1372.8482 -0.44939720 -0.3313744 NA AHL20
# 9 AT5G60200 -0.23040752 9.592281e-32 -0.3877858 673.3668 -0.24152133 -0.3551172 NA DOF5.3
# 1 AT1G52150 -0.01352807 3.660479e-02 -0.2822424 493.6272 -0.10277135 -0.3354909 NA ATHB-15
# 4 AT2G22430 0.53927909 3.009286e-43 0.3242114 481.4268 0.48990652 0.2671454 NA ATHB-6
# 10 AT5G65590 0.19838070 5.449207e-39 0.3535122 454.8653 0.19894077 0.2939522 NA AT5G65590 aka SCAP1
# 6 AT3G01470 0.00000000 9.838639e-01 -0.2254802 327.7194 -0.10693530 -0.1848806 NA ATHB-1
# 2 AT1G64620 0.04547039 1.688527e-02 -0.1025043 292.8133 0.10958571 -0.1231658 NA AT1G64620
# 5 AT2G37590 0.00000000 5.330965e-01 -0.2068212 261.5152 -0.08311396 -0.1980168 NA DOF2.4
# 8 AT5G39660 -0.03144685 5.430603e-03 -0.2213106 254.0163 -0.12676544 -0.1426431 NA CDF2
# 3 AT1G80840 0.00000000 3.945345e-01 0.1509693 252.7218 0.03454721 0.1306107 NA WRKY40
#------------------------------------------------------------------------------
# fifth model: 30 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(22532400,22538210),
end=c(22533900,22539910),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x5 <- build(builder)
lapply(x5, dim)
tbl.model.5 <- x5$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.5$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.5$symbol <- tf.symbols
# tbl.model.5
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 -0.45126281 1.919310e-162 -0.5628267 1217.9599 -0.33094480 -0.3313744 NA AHL20
# 4 AT2G01760 0.35079119 1.811735e-68 0.4259098 591.9681 0.31450068 0.3974638 NA ARR14
# 9 AT4G31920 -0.41681921 2.657399e-51 -0.3963221 578.9558 -0.38395182 -0.3279602 NA ARR10
# 12 AT5G60200 -0.21672612 2.056503e-33 -0.3877858 526.6647 -0.22047585 -0.3551172 NA DOF5.3
# 2 AT1G67710 -0.04803401 3.653619e-14 -0.3991527 319.9506 -0.09684702 -0.3220879 NA ARR11
# 13 AT5G65590 0.14140722 6.260262e-34 0.3535122 297.9758 0.15024399 0.2939522 NA AT5G65590 aka SCAP1
# 10 AT4G40060 0.00000000 2.658003e-01 -0.2832086 235.4189 -0.01086425 -0.2614751 NA ATHB-16
# 1 AT1G64620 0.03707026 1.295114e-02 -0.1025043 207.6641 0.10161573 -0.1231658 NA AT1G64620
# 5 AT2G37590 -0.04281820 3.412511e-03 -0.2068212 200.8512 -0.14316299 -0.1980168 NA DOF2.4
# 11 AT5G39660 -0.08074957 4.412580e-06 -0.2213106 191.9243 -0.14452532 -0.1426431 NA CDF2
#------------------------------------------------------------------------------
# sixth model: 29 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(13032400,13037100),
end=c(13033400,13038100),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x6 <- build(builder)
lapply(x6, dim)
tbl.model.6 <- x6$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.6$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.6$symbol <- tf.symbols
# tbl.model.6
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 -0.38623478 1.919310e-162 -0.5628267 1072.8117 -0.29493570 -0.3313744 NA AHL20
# 15 AT5G62165 -0.17274948 8.382164e-104 -0.4991596 660.4390 -0.16416343 -0.3849297 NA AGL42
# 9 AT4G31920 -0.30060591 3.462460e-41 -0.3963221 465.5176 -0.30569923 -0.3279602 NA ARR10
# 14 AT5G60200 -0.12411684 3.034644e-23 -0.3877858 402.2469 -0.16687658 -0.3551172 NA DOF5.3
# 1 AT1G17950 -0.05301677 3.260175e-34 -0.4291068 352.9953 -0.08665296 -0.3200268 NA MYB52
# 7 AT2G42200 0.17331619 1.024211e-34 0.3494224 256.2832 0.17743521 0.3171417 NA SPL9
# 5 AT2G22430 0.34975419 6.445556e-38 0.3242114 246.3073 0.31453242 0.2671454 NA ATHB-6
# 11 AT5G02320 -0.17835622 1.354184e-32 -0.3254427 218.6656 -0.17944006 -0.3079975 NA MYB3R-5
# 16 AT5G65590 0.10574985 1.332595e-33 0.3535122 216.4876 0.12160481 0.2939522 NA AT5G65590 aka SCAP1
# 12 AT5G11510 -0.06323905 1.388600e-09 -0.2168312 177.7344 -0.09146526 -0.2150125 NA MYB3R-4
#------------------------------------------------------------------------------------------------------------------------------
# Get high expression data
wbc19.vector <- mtx[c("AT3G55130", "AT3G59060"),]
plotWBC19 <- plot((mtx[c("AT3G55130"),]), (mtx[c("AT3G59060"),]), main= "WBC19 Expression in Roots")
new.wbc.vector <- t(wbc19.vector)
number1 <- new.wbc.vector[(new.wbc.vector[,1] > 2),]
number2 <- number1[(number1[,2] > 2),]
plot(number2, main="High WBC19 Expression in Roots")
highExpressionWBC19 <- t(number2)
rownames(highExpressionWBC19)
mtx2 <- highExpressionWBC19
mtx2.colnames <- colnames(mtx2)
mtx2 <- mtx[,mtx2.colnames]
# created matrix of high expression data
# mtx3 has 169 samples
ordered.number2<- number2[order(number2[,1], number2[,2], decreasing=TRUE),]
# allows you to view sample names with highest expression
ordered.number2.50 <- ordered.number2[1:50,]
evenHigherExpressionWBC19 <- t(ordered.number2.50)
rownames(evenHigherExpressionWBC19)
mtx3.colnames <- colnames(evenHigherExpressionWBC19)
mtx3 <- mtx[,mtx3.colnames]
#build smaller matrix with even higher expression of WBC19
# top 1-3,7 samples are from E-GEOD-24348
# Experiment: Transcription profiling by array of Arabidopsis mutant for nas4x after iron deprivation
# top 4-6,9-10 samples are from experiment 489 from NASC arrays:
# Experiment: Investigating the molecular response of Arabidopsis to combined
# nematode and dehydration stress using the Affymetrix ATH1 microarray.
# top 8 sample is from Experiment #490 from NASC arrays:
# Experiment: Cell-specific nitrogen responses in the Arabidopsis root
#------------------------------------------------------------------------------------------------------------------------------
#Modeling with high expression data:
#------------------------------------------------------------------------------
# seventh model: 32 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(22632400,22638210),
end=c(22633900,22639910),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx2,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x7 <- build(builder)
lapply(x7, dim)
tbl.model.7 <- x7$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.7$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.7$symbol <- tf.symbols
# tbl.model.7
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT3G61150 1.844280e-01 6.937774e-06 0.3380855 6.106435 0.13379306 0.3369922 NA HDG1
# 11 AT5G16560 1.117630e-02 2.977540e-02 0.2091608 5.729552 0.03780677 0.2138357 NA KAN
# 13 AT5G60200 -7.479631e-02 1.935798e-03 -0.2496830 5.132500 -0.08682439 -0.3265924 NA DOF5.3
# 7 AT2G37590 -1.357610e-01 5.089497e-04 -0.2755380 4.436605 -0.14352526 -0.3060042 NA DOF2.4
# 6 AT2G22430 4.522021e-02 2.389368e-02 0.2079384 4.399553 0.10152916 0.2029337 NA ATHB-6
# 4 AT1G73360 8.112228e-03 4.952906e-02 0.1015424 4.388088 0.05521869 0.2124932 NA EDT1
# 12 AT5G25830 3.193348e-02 1.265326e-02 0.1590170 3.715073 0.04473851 0.1023321 NA GATA12
# 9 AT3G54810 -4.519279e-02 1.264134e-02 -0.1717809 3.307638 -0.07737611 -0.1361245 NA GATA8
# 14 AT5G62940 0.000000e+00 8.643032e-01 -0.1254659 2.961352 -0.02415076 -0.1940754 NA DOF5.6
# 8 AT2G46130 -7.883023e-05 6.007031e-02 -0.1284293 2.786565 -0.02130100 -0.1615932 NA WRKY43
#------------------------------------------------------------------------------------------------------------------------
#use smaller mtx (mtx3) for even higher expression data
#------------------------------------------------------------------------------
# eighth model: 79 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(18632400,18638210),
end=c(18633900,18639710),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x8 <- build(builder)
lapply(x8, dim)
tbl.model.8 <- x8$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.8$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.8$symbol <- tf.symbols
# tbl.model.8
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G25550 -0.015443951 0.02172242 -0.3272428 0.9130473 -0.027804138 -0.29190876 NA AT1G25550
# 8 AT1G74930 0.000000000 0.83419203 -0.2764086 0.4672606 -0.007247037 -0.27875150 NA ERF018
# 33 AT5G65210 0.000000000 0.13996296 -0.2391059 0.4449350 -0.025328820 -0.16869148 NA TGA1
# 31 AT5G62165 0.000000000 0.24719669 0.2466379 0.3578841 0.017652083 0.25051621 NA AGL42
# 4 AT1G61110 0.000000000 0.04399367 0.2745188 0.3098864 0.027149326 0.36998800 NA anac025
# 3 AT1G53160 0.000000000 0.51715079 0.1885299 0.2887157 0.022863901 0.18703481 NA SPL4
# 20 AT3G54810 -0.005432485 0.03930759 -0.2940763 0.2704288 -0.043393328 -0.20384154 NA GATA8
# 29 AT5G39610 0.000000000 0.09783550 -0.2399027 0.2386434 -0.012143183 -0.27452581 NA ATNAC2
# 18 AT3G06740 0.029844783 0.01895477 0.3307985 0.2139190 0.045933432 0.27352397 NA GATA15
# 32 AT5G62940 0.000000000 0.20621717 0.1423515 0.2113919 0.018586458 0.08821128 NA DOF5.6
#------------------------------------------------------------------------------
# ninth model: 24 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(13532400,13538210),
end=c(13533400,13539210),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x9 <- build(builder)
lapply(x9, dim)
tbl.model.9 <- x9$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.9$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.9$symbol <- tf.symbols
# tbl.model.9
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT3G54810 -0.060234503 0.03603976 -0.2940763 0.6984862 -0.063738119 -0.20384154 NA GATA8
# 12 AT5G62940 0.000000000 0.24576433 0.1423515 0.6378016 0.028938147 0.08821128 NA DOF5.6
# 7 AT4G32890 0.003008057 0.10549938 0.2345584 0.6144417 0.027627205 0.19164466 NA GATA9
# 5 AT3G06740 0.101587678 0.01895477 0.3307985 0.6094221 0.073410073 0.27352397 NA GATA15
# 2 AT1G53160 0.000000000 0.34003156 0.1885299 0.5776807 0.026310324 0.18703481 NA SPL4
# 10 AT5G25830 0.000000000 0.25792875 0.1818998 0.5764563 0.015273616 0.07130852 NA GATA12
# 9 AT5G16560 0.000000000 0.56268364 0.1300772 0.5381874 0.008683821 0.15668667 NA KAN
# 8 AT5G02840 0.000000000 0.12336829 -0.1900216 0.3819700 -0.043429621 -0.11750300 NA LCL1
# 13 AT5G65590 0.000000000 0.45605691 -0.1758766 0.3539472 -0.009495115 -0.30208884 NA AT5G65590 aka SCAP1
# 3 AT1G64620 0.000000000 0.15354621 0.2198560 0.3528297 0.049762578 0.21949580 NA AT1G64620
#------------------------------------------------------------------------------
# 10th model: 27 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20434000,20435500),
end=c(20435000,20436500),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x10 <- build(builder)
lapply(x10, dim)
tbl.model.10 <- x10$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.10$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.10$symbol <- tf.symbols
# tbl.model.10
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT5G46760 0.000000000 0.72496950 0.1201494 0.5582605 0.02368710 0.18722689 NA AT5G46760
# 5 AT3G06740 0.099267109 0.01895477 0.3307985 0.5543676 0.06976204 0.27352397 NA GATA15
# 12 AT5G62165 0.003440748 0.08922270 0.2466379 0.5489340 0.02126300 0.25051621 NA AGL42
# 4 AT2G36270 0.000000000 0.21446488 -0.2409433 0.5174508 -0.01353082 -0.11510204 NA ABI5
# 13 AT5G62940 0.000000000 0.23139773 0.1423515 0.5087755 0.02582457 0.08821128 NA DOF5.6
# 7 AT3G54810 -0.054040991 0.03603976 -0.2940763 0.4856329 -0.04923703 -0.20384154 NA GATA8
# 1 AT1G53160 0.000000000 0.26685569 0.1885299 0.4703625 0.02634593 0.18703481 NA SPL4
# 6 AT3G47500 0.000000000 0.27552762 0.2092642 0.4636700 0.02766993 0.29901561 NA CDF3
# 10 AT5G16560 0.000000000 0.60334479 0.1300772 0.4396909 0.01021718 0.15668667 NA KAN
# 2 AT1G64620 0.000000000 0.13975644 0.2198560 0.4252454 0.06111624 0.21949580 NA AT1G64620
#------------------------------------------------------------------------------
# 11th model: 19 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20433010,20439400),
end=c(20434010,20440400),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x11 <- build(builder)
lapply(x11, dim)
tbl.model.11 <- x11$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.11$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.11$symbol <- tf.symbols
# tbl.model.11
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 1 AT1G25550 -0.022746363 0.02172242 -0.3272428 1.2660687 -0.039665255 -0.29190876 NA AT1G25550
# 5 AT3G54810 -0.025411789 0.04902443 -0.2940763 0.6545343 -0.051632292 -0.20384154 NA GATA8
# 10 AT5G62165 0.001002055 0.10201106 0.2466379 0.5926299 0.021996337 0.25051621 NA AGL42
# 11 AT5G62940 0.000000000 0.32551922 0.1423515 0.5173541 0.026794172 0.08821128 NA DOF5.6
# 6 AT4G14465 -0.029122906 0.02643625 -0.3113506 0.5150485 -0.027251150 -0.28038415 NA AHL20
# 4 AT3G06740 0.091388846 0.01895477 0.3307985 0.5034917 0.072256535 0.27352397 NA GATA15
# 2 AT1G64620 0.000000000 0.11792728 0.2198560 0.4519734 0.060525966 0.21949580 NA AT1G64620
# 9 AT5G16560 0.000000000 0.73158073 0.1300772 0.3951998 0.008438453 0.15668667 NA KAN
# 12 AT5G65590 0.000000000 0.74044625 -0.1758766 0.3658925 -0.009455525 -0.30208884 NA AT5G65590 aka SCAP1
# 8 AT5G03790 0.000000000 0.19902620 -0.1720009 0.3526549 -0.013799422 -0.20096038 NA ATHB-51
#------------------------------------------------------------------------------
# 12th model: 63 TFs with >= 95% motif match within 1000 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20431010,20436400),
end=c(20432010,20437400),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x12 <- build(builder)
lapply(x12, dim)
tbl.model.12 <- x12$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.12$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.12$symbol <- tf.symbols
# tbl.model.12
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 17 AT4G17880 0.093382268 0.0008222258 0.4580876 1.2723674 0.04412493 0.4253061 NA AT4G17880
# 3 AT1G22985 -0.033356269 0.0099813571 -0.3858113 0.7122849 -0.04852291 -0.3998559 NA AT1G22985
# 13 AT3G15210 0.000000000 0.7570313477 -0.2229774 0.3615262 -0.01276041 -0.1080912 NA ERF4
# 23 AT5G07680 0.000000000 0.0966796673 -0.2597433 0.3409373 -0.01130599 -0.1699400 NA ANAC080
# 19 AT4G32800 0.000000000 0.0847828944 0.2511761 0.2765392 0.01980436 0.2693397 NA AT4G32800
# 15 AT4G14465 -0.018735267 0.0139181058 -0.3113506 0.2307609 -0.02534347 -0.2803842 NA AHL20
# 14 AT4G00050 0.000000000 0.5831800806 0.1559773 0.2282816 0.01596675 0.2100840 NA UNE10
# 26 AT5G43410 0.003556295 0.0252531286 0.2610044 0.2255470 0.01595780 0.3809364 NA AT5G43410
# 27 AT5G46760 0.000000000 0.8295434551 0.1201494 0.2081812 0.02031843 0.1872269 NA AT5G46760
# 29 AT5G62165 0.000000000 0.0632480856 0.2466379 0.1992148 0.01515128 0.2505162 NA AGL42
#------------------------------------------------------------------------------
# 13th model: 52 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20430000,20435500),
end=c(20431500,20437000),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x13 <- build(builder)
lapply(x13, dim)
tbl.model.13 <- x13$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.13$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.13$symbol <- tf.symbols
# tbl.model.13
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G22985 -0.072674574 0.005655387 -0.3858113 0.8585397 -0.05207582 -0.39985594 NA AT1G22985
# 11 AT3G15210 0.000000000 0.195240490 -0.2229774 0.4950218 -0.01760155 -0.10809124 NA ERF4
# 20 AT5G07680 0.000000000 0.126299481 -0.2597433 0.4152436 -0.01300549 -0.16993998 NA ANAC080
# 15 AT4G32800 0.000000000 0.060636290 0.2511761 0.3392273 0.02260782 0.26933974 NA AT4G32800
# 23 AT5G46760 0.000000000 0.402790027 0.1201494 0.2764723 0.02720795 0.18722689 NA AT5G46760
# 10 AT3G06740 0.004794159 0.044481922 0.3307985 0.2722499 0.04303362 0.27352397 NA GATA15
# 7 AT1G69780 0.000000000 0.211205052 -0.1558149 0.2606508 -0.01934952 -0.24043217 NA ATHB13
# 25 AT5G62940 0.000000000 0.322875673 0.1423515 0.2501957 0.01630500 0.08821128 NA DOF5.6
# 12 AT3G47500 0.000000000 0.807130465 0.2092642 0.2484925 0.01951010 0.29901561 NA CDF3
# 22 AT5G43410 0.000000000 0.052223010 0.2610044 0.2283790 0.01637408 0.38093637 NA AT5G43410
#no regulatory regions yet
#------------------------------------------------------------------------------
# 14th model: 42 TFs with >= 95% motif match within 1500 (each) promoter.
#------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20423050,20427384),
end=c(20424550,20428884),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x14 <- build(builder)
lapply(x14, dim)
tbl.model.14 <- x14$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.14$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.14$symbol <- tf.symbols
# tbl.model.14
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT4G14465 -0.032987913 0.02604191 -0.3113506 0.5811695 -0.032779964 -0.28038415 NA AHL20
# 8 AT3G47500 0.000000000 0.10256329 0.2092642 0.5284426 0.038234129 0.29901561 NA CDF3
# 3 AT1G69780 0.000000000 0.47126347 -0.1558149 0.4558065 -0.020464919 -0.24043217 NA ATHB13
# 17 AT5G62940 0.000000000 0.44899545 0.1423515 0.4496875 0.019987754 0.08821128 NA DOF5.6
# 7 AT3G06740 0.070921388 0.01895477 0.3307985 0.4338415 0.057434221 0.27352397 NA GATA15
# 11 AT4G17460 0.009649258 0.04073222 0.3055506 0.4336755 0.024513398 0.24264106 NA HAT1
# 2 AT1G64620 0.000000000 0.11742320 0.2198560 0.4100220 0.060567525 0.21949580 NA AT1G64620
# 12 AT4G37180 -0.025091221 0.05295391 -0.2804712 0.3833676 -0.044010154 -0.29142857 NA AT4G37180
# 15 AT5G16560 0.000000000 0.95486682 0.1300772 0.3681782 0.006399682 0.15668667 NA KAN
# 5 AT2G46680 0.000000000 0.10509509 -0.2534576 0.3379815 -0.017638285 -0.26261705 NA ATHB-7
#---------------------------------------------------------------------------------------
# 15th model: 20 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
#---------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432106,20434107),
end=c(20434106,20434607),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x15 <- build(builder)
lapply(x15, dim)
tbl.model.15 <- x15$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.15$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.15$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT3G54810 -0.039787610 0.04365073 -0.2940763 0.6846830 -0.06001076 -0.20384154 NA GATA8
# 5 AT3G06740 0.124765638 0.01895477 0.3307985 0.6674693 0.08295289 0.27352397 NA GATA15
# 10 AT5G62940 0.000000000 0.41461791 0.1423515 0.6670734 0.02197803 0.08821128 NA DOF5.6
# 7 AT4G14465 -0.035092534 0.02604191 -0.3113506 0.6431163 -0.03020385 -0.28038415 NA AHL20
# 2 AT1G69780 0.000000000 0.91517751 -0.1558149 0.6375511 -0.01767389 -0.24043217 NA ATHB13
# 1 AT1G64620 0.003497219 0.12537262 0.2198560 0.6301531 0.06288908 0.21949580 NA AT1G64620
# 9 AT5G16560 0.000000000 0.54146029 0.1300772 0.5653065 0.01288556 0.15668667 NA KAN
# 11 AT5G65590 0.000000000 0.53350580 -0.1758766 0.4449672 -0.01152127 -0.30208884 NA AT5G65590 aka SCAP1
# 3 AT1G80840 0.000000000 0.44366311 -0.1730834 0.3949913 -0.01562107 -0.14900360 NA WRKY40
# 4 AT3G01470 0.000000000 0.85935301 -0.1515070 0.3905654 -0.01787426 -0.13373349 NA ATHB-1
#----------------------------------------------------------------------------------------
# 16th model: 20 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# same region as #15 on atac.seq; Use bigger matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432112,20434113),
end=c(20434112,20434613),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x16 <- build(builder)
lapply(x16, dim)
tbl.model.16 <- x16$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.16$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.16$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 7 AT4G14465 -0.631679182 1.919310e-162 -0.5628267 1392.8188 -0.45797827 -0.3313744 NA AHL20
# 9 AT5G60200 -0.226439315 9.592281e-32 -0.3877858 663.6387 -0.24379551 -0.3551172 NA DOF5.3
# 1 AT1G52150 -0.007069971 3.660479e-02 -0.2822424 510.0015 -0.10189183 -0.3354909 NA ATHB-15
# 4 AT2G22430 0.534924218 3.009286e-43 0.3242114 484.3623 0.49642149 0.2671454 NA ATHB-6
# 10 AT5G65590 0.196341104 5.449207e-39 0.3535122 453.4247 0.20074867 0.2939522 NA AT5G65590 aka SCAP1
# 6 AT3G01470 0.000000000 9.838639e-01 -0.2254802 318.4881 -0.10405340 -0.1848806 NA ATHB-1
# 2 AT1G64620 0.032505573 1.688527e-02 -0.1025043 291.1077 0.11394430 -0.1231658 NA AT1G64620
# 5 AT2G37590 0.000000000 5.330965e-01 -0.2068212 260.5657 -0.08101563 -0.1980168 NA DOF2.4
# 8 AT5G39660 -0.026584270 5.430603e-03 -0.2213106 253.8812 -0.12570706 -0.1426431 NA CDF2
# 3 AT1G80840 0.000000000 3.945345e-01 0.1509693 251.0913 0.03387059 0.1306107 NA WRKY40
#with mtx3:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT3G54810 -0.03167221 0.04365073 -0.2940763 0.8106728 -0.05785286 -0.20384154 NA GATA8
# 2 AT1G69780 0.00000000 0.91517751 -0.1558149 0.6819254 -0.01714176 -0.24043217 NA ATHB13
# 10 AT5G62940 0.00000000 0.41461791 0.1423515 0.6513227 0.02122323 0.08821128 NA DOF5.6
# 7 AT4G14465 -0.02996968 0.02604191 -0.3113506 0.6449731 -0.02901716 -0.28038415 NA AHL20
# 5 AT3G06740 0.10799807 0.01895477 0.3307985 0.6422672 0.07991416 0.27352397 NA GATA15
# 9 AT5G16560 0.00000000 0.54146029 0.1300772 0.5825086 0.01236389 0.15668667 NA KAN
# 1 AT1G64620 0.00000000 0.12537262 0.2198560 0.5129569 0.06036687 0.21949580 NA AT1G64620
# 11 AT5G65590 0.00000000 0.53350580 -0.1758766 0.4958118 -0.01116943 -0.30208884 NA AT5G65590 aka SCAP1
# 8 AT5G03790 0.00000000 0.18322862 -0.1720009 0.4236046 -0.01437446 -0.20096038 NA ATHB-51
# 3 AT1G80840 0.00000000 0.44366311 -0.1730834 0.4227791 -0.01505340 -0.14900360 NA WRKY40
#----------------------------------------------------------------------------------------
# 17th model: 28 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# same region as #15 on atac.seq; Use smaller matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20432112,20434113),
end=c(20434112,20436113),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x17 <- build(builder)
lapply(x17, dim)
tbl.model.17 <- x17$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.17$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.17$symbol <- tf.symbols
#with mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT3G54810 -0.03315773 0.08019973 -0.2644196 8.106659 -0.16569641 -0.20000000 NA GATA8
# 6 AT2G37590 0.19950526 0.04200000 0.2887062 6.680566 0.24873045 0.30727491 NA DOF2.4
# 14 AT5G65590 0.00000000 0.24946301 -0.1392564 6.477805 -0.03827349 -0.32513806 NA AT5G65590
# 9 AT4G14465 -0.06861681 0.03802656 -0.2942996 6.385959 -0.06734547 -0.17166867 NA AHL20
# 11 AT5G18830 0.00000000 0.12425412 -0.2049080 6.275481 -0.23378909 -0.09176471 NA SPL7
# 4 AT1G69780 0.00000000 0.95502237 -0.2050077 4.572243 -0.04328460 0.05690276 NA ATHB13
#----------------------------------------------------------------------------------------
# 18th model: 104 TFs with >= 95% motif match within 5000 upstream, 5000 downstream of TSS
# same region as #15 on atac.seq; Use bigger matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20429112,20434113),
end=c(20434112,20439113),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x18 <- build(builder)
lapply(x18, dim)
tbl.model.18 <- x18$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.18$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.18$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 30 AT3G59060 0.20660505 7.586023e-257 0.6738150 1106.31242 0.09978381 0.5077358 NA PIF5
# 33 AT4G14465 -0.14559575 7.720037e-103 -0.5628267 537.79490 -0.09335849 -0.3313744 NA AHL20
# 32 AT4G00050 0.21111517 1.790158e-129 0.5765029 451.14938 0.14417470 0.4916015 NA UNE10
# 38 AT4G36540 0.16916609 2.128292e-89 0.5333334 329.98386 0.12713276 0.4806764 NA BEE2
# 23 AT3G01220 -0.08784450 1.524911e-46 -0.5233056 273.24071 -0.08700164 -0.3849471 NA ATHB20
# 51 AT5G62165 -0.03805766 3.058215e-24 -0.4991596 217.69778 -0.05817641 -0.3849297 NA AGL42
# 9 AT1G53160 0.02257309 1.136179e-12 0.3917817 92.34605 0.04437246 0.3554472 NA SPL4
# 50 AT5G60200 -0.02179131 5.127391e-12 -0.3877858 91.62244 -0.05630662 -0.3551172 NA DOF5.3
# 43 AT5G18830 -0.12205317 5.265642e-17 -0.3776184 82.37603 -0.15105459 -0.3622195 NA SPL7
# 35 AT4G31920 0.00000000 9.739117e-01 -0.3963221 82.05983 -0.08120219 -0.3279602 NA ARR10
#----------------------------------------------------------------------------------------
# 19th model: 98 TFs with >= 95% motif match within 5000 upstream, 5000 downstream of TSS
# same region as previous on atac.seq; Use smaller matrix
#----------------------------------------------------------------------------------------
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20429112,20436288),
end=c(20434112,20441288),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WBC19",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x19 <- build(builder)
lapply(x19, dim)
tbl.model.19 <- x19$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.19$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.19$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 27 AT3G59060 0.20847974 7.586023e-257 0.6738150 1132.98732 0.10504491 0.5077358 NA PIF5
# 30 AT4G14465 -0.14831565 7.720037e-103 -0.5628267 527.16775 -0.09513729 -0.3313744 NA AHL20
# 29 AT4G00050 0.21213270 1.790158e-129 0.5765029 467.57087 0.14878185 0.4916015 NA UNE10
# 21 AT3G01220 -0.08679869 1.524911e-46 -0.5233056 329.26474 -0.09229795 -0.3849471 NA ATHB20
# 34 AT4G36540 0.16911396 2.128292e-89 0.5333334 320.70080 0.12903040 0.4806764 NA BEE2
# 47 AT5G62165 -0.03733050 3.058215e-24 -0.4991596 185.77783 -0.06039029 -0.3849297 NA AGL42
# 46 AT5G60200 -0.02111709 5.127391e-12 -0.3877858 91.65985 -0.05759636 -0.3551172 NA DOF5.3
# 39 AT5G18830 -0.12003641 5.265642e-17 -0.3776184 86.14901 -0.15167134 -0.3622195 NA SPL7
# 9 AT1G53160 0.02216566 1.136179e-12 0.3917817 84.55421 0.04472416 0.3554472 NA SPL4
# 32 AT4G31920 0.00000000 9.427982e-01 -0.3963221 73.34116 -0.07899182 -0.3279602 NA ARR10
# with smaller matrix mtx3:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 29 AT4G17880 0.06445869 0.0008222258 0.4580876 0.9450456 0.034282571 0.4253061 NA AT4G17880
# 7 AT1G25550 0.00000000 0.0328008880 -0.3272428 0.5391818 -0.019705100 -0.2919088 NA AT1G25550
# 40 AT5G37020 0.10286850 0.0010999014 0.4500930 0.5314113 0.073007047 0.3855462 NA ARF8
# 5 AT1G22985 -0.01027799 0.0142236209 -0.3858113 0.3790130 -0.038876945 -0.3998559 NA AT1G22985
# 37 AT5G07680 0.00000000 0.0518950825 -0.2597433 0.3328196 -0.010513774 -0.1699400 NA ANAC080
# 22 AT3G15270 0.00000000 0.0593857696 0.3030271 0.2860886 0.013592940 0.2473469 NA SPL5
# 21 AT3G15210 0.00000000 0.7331515557 -0.2229774 0.1963197 -0.007838267 -0.1080912 NA ERF4
# 14 AT1G74930 0.00000000 0.6960000957 -0.2764086 0.1911578 -0.004877530 -0.2787515 NA ERF018
# 41 AT5G43410 0.00000000 0.0281112451 0.2610044 0.1895305 0.016059188 0.3809364 NA AT5G43410
# 32 AT4G32800 0.00000000 0.0749800426 0.2511761 0.1830661 0.018230405 0.2693397 NA AT4G32800
#------------------------------------------------------------------------------------------------------------------------
#Model transcription factors found and see what you find! Remember to use both the bigger and smaller matrix
#----------------------------------------------------------------------------------------
# 20th model: 48 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AHL20")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(8318970,8320971),
end=c(8320970,8322971),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AHL20",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x20 <- build(builder)
lapply(x20, dim)
tbl.model.20 <- x20$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.20$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.20$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT1G69780 0.48600063 9.251529e-08 0.6717750 10.353287 0.1155951 0.4410564 NA ATHB13
# 16 AT3G01470 0.00000000 6.915034e-01 0.4376918 3.907703 0.0631113 0.3113085 NA ATHB-1
# 23 AT4G40060 0.00000000 9.979775e-02 0.3992646 3.388234 0.1189144 0.2845138 NA ATHB-16
# 12 AT2G42200 0.00000000 3.466663e-02 -0.4381118 3.112813 -0.0665653 -0.2523409 NA SPL9
# 30 AT5G56270 -0.04441607 4.814786e-03 -0.3738075 2.231266 -0.1599834 -0.4034094 NA WRKY2
# 9 AT2G22430 0.00000000 7.772254e-02 -0.4260024 2.090012 -0.1382915 -0.4021609 NA ATHB-6
#----------------------------------------------------------------------------------------
# 21st model: 66 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "PIF5")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(21826187,21828188),
end=c(21828187,21828688),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="PIF5",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x21 <- build(builder)
lapply(x21, dim)
tbl.model.21 <- x21$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.21$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.21$symbol <- tf.symbols
#with bigger matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 28 AT4G36540 0.31729995 7.689933e-146 0.5413038 1301.6056 0.22207458 0.4579084 NA BEE2
# 36 AT5G62165 -0.20517737 9.475748e-170 -0.5731729 1169.2908 -0.15495633 -0.4259736 NA AGL42
# 24 AT4G31920 -0.33374588 7.894880e-84 -0.4761674 558.2853 -0.31075981 -0.3834930 NA ARR10
# 22 AT3G60490 -0.12682864 8.515071e-76 -0.4383758 403.6812 -0.09674127 -0.3911913 NA AT3G60490
# 35 AT5G60200 -0.16877910 5.074558e-47 -0.4158650 359.4931 -0.16318164 -0.3608002 NA DOF5.3
# 31 AT5G25830 -0.21821714 6.054286e-93 -0.4750765 342.0631 -0.20263815 -0.3635412 NA GATA12
# 1 AT1G17950 -0.08224473 1.652507e-97 -0.4788226 332.4711 -0.07033979 -0.3746657 NA MYB52
# 6 AT1G53170 0.17270085 1.592222e-70 0.4460720 267.3134 0.13643533 0.3760144 NA ERF8
# 21 AT3G57600 0.16206402 1.123689e-74 0.4555560 236.9006 0.13824439 0.3921747 NA AT3G57600
# 20 AT3G16770 0.16564353 1.974600e-35 0.3417913 179.2040 0.17261444 0.3205095 NA RAP2-3
#with small matrix:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 1 AT1G06160 0.000000000 0.035329865 0.2210415 2.217605 0.03002289 0.03442977 NA ERF094
# 10 AT2G22430 0.073019124 0.006538555 0.4177917 1.888240 0.07946312 0.38141657 NA ATHB-6
# 30 AT5G62000 0.032641170 0.002867844 0.4235871 1.837604 0.03986967 0.44460984 NA ARF2
# 3 AT1G28370 0.080778283 0.001006354 0.4512119 1.677484 0.03366724 0.46448980 NA ERF11
# 23 AT5G08130 -0.169730460 0.004738418 -0.3754061 1.591577 -0.13639390 -0.27308523 NA BIM1
# 32 AT5G65590 0.007382096 0.017227683 0.3482814 1.426294 0.03293105 0.32187275 NA AT5G65590 aka SCAP1
# 18 AT3G23240 0.000000000 0.457686819 0.3319695 1.276688 0.01306984 0.35001200 NA ERF1B
# 9 AT1G80840 0.000000000 0.920894439 0.4096893 1.226684 0.02722907 0.38285714 NA WRKY40
# 17 AT3G16770 0.005335604 0.022086968 0.2903235 1.150150 0.03497298 0.22698679 NA RAP2-3
# 16 AT3G15210 0.000000000 0.359911361 0.3449356 1.053483 0.03197530 0.42770708 NA ERF4
#----------------------------------------------------------------------------------------
# 22nd model: 26 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "UNE10")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(15862,17863),
end=c(17862,18363),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="UNE10",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x22 <- build(builder)
lapply(x22, dim)
tbl.model.22 <- x22$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.22$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.22$symbol <- tf.symbols
#with bigger matrix:
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 -0.2352269 4.954686e-106 -0.4678254 376.5201 -0.194942574 -0.3154342 NA AHL20
# 11 AT5G62165 -0.1605824 4.050781e-102 -0.4629587 339.9943 -0.142173176 -0.3505407 NA AGL42
# 10 AT5G60200 -0.1010548 2.685066e-24 -0.3526809 175.9870 -0.122245187 -0.3010236 NA DOF5.3
# 12 AT5G65590 0.1221619 8.165302e-37 0.3390436 171.3479 0.119279678 0.3121283 NA AT5G65590 aka SCAP1
# 2 AT1G52150 0.0000000 7.388956e-02 -0.2542754 155.6862 -0.055204319 -0.2906103 NA ATHB-15
# 4 AT2G22430 0.2062901 8.318967e-26 0.2863719 125.5836 0.224416183 0.2595819 NA ATHB-6
# 3 AT1G63480 -0.1759239 1.151318e-17 -0.1640331 114.9688 -0.167926041 -0.1373603 NA AT1G63480
# 6 AT2G46590 0.0000000 3.175261e-01 -0.1714005 112.2792 0.007433777 -0.1765432 NA DAG2
# 9 AT4G40060 0.0000000 5.794193e-01 -0.1947899 103.9999 -0.049056939 -0.2129665 NA ATHB-16
# 7 AT3G01470 0.0000000 1.467616e-01 -0.1471854 100.7755 0.026399529 -0.1529602 NA ATHB-1
#with smaller matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT4G40060 0.38676339 0.0003898992 0.4823011 3.436711 0.18352202 0.4773589 NA ATHB-16
# 12 AT5G39660 0.19433705 0.0006938689 0.4612668 2.995625 0.12643245 0.3785354 NA CDF2
# 1 AT1G47655 0.00000000 0.2856213499 0.2405810 2.256596 0.03568010 0.2492677 NA AT1G47655
# 9 AT4G14465 0.00000000 0.2353579376 0.2489781 1.789956 0.03119151 0.1388235 NA AHL20
# 13 AT5G60200 0.00000000 0.0452494962 0.1770031 1.516223 0.10228559 0.1669628 NA DOF5.3
# 7 AT3G01470 0.00000000 0.4349516377 0.1756762 1.469617 0.04019154 0.1840576 NA ATHB-1
# 8 AT3G06740 0.02947161 0.0326759802 0.2637456 1.373348 0.13936260 0.2711710 NA GATA15
# 4 AT1G63480 0.00000000 0.2213847405 -0.3395003 1.337998 -0.07550148 -0.2319808 NA AT1G63480
# 14 AT5G62165 0.00000000 0.0541638997 -0.3317078 1.224800 -0.03967390 -0.1822329 NA AGL42
# 11 AT5G13180 0.00000000 0.1539337710 0.1325205 1.158471 0.06127619 0.1330612 NA ANAC083
#----------------------------------------------------------------------------------------
# 23rd model: 29 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "BEE2")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(17241698,17243699),
end=c(17243698,17244199),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="BEE2",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x23 <- build(builder)
lapply(x23, dim)
tbl.model.23 <- x23$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.23$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.23$symbol <- tf.symbols
# with bigger matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 15 AT5G62165 -0.14654224 8.781013e-62 -0.3672063 386.6629 -0.11614451 -0.28922974 NA AGL42
# 9 AT4G31920 -0.32184470 8.196087e-52 -0.3513091 343.8435 -0.28954908 -0.32828247 NA ARR10
# 1 AT1G53160 0.14257230 3.025479e-64 0.3721841 309.8696 0.12612429 0.40741137 NA SPL4
# 4 AT2G02540 0.16268244 1.387200e-67 0.3798380 261.2975 0.14506505 0.36150086 NA ATHB21
# 14 AT5G60200 0.00000000 5.376887e-02 -0.1971474 205.7040 -0.04499575 -0.25056895 NA DOF5.3
# 3 AT1G69780 0.19690706 1.589081e-17 0.1621895 201.0077 0.18118475 0.15222777 NA ATHB13
# 17 AT5G65310 0.27520837 4.586593e-20 0.1561768 173.0371 0.23328707 0.17642863 NA ATHB-5
# 10 AT4G32890 -0.13829996 7.193778e-26 -0.2485245 159.3232 -0.12355251 -0.25224104 NA GATA9
# 11 AT5G16560 -0.04288807 2.020327e-07 -0.1283945 148.8646 -0.06416597 -0.08825239 NA KAN
# 13 AT5G49520 -0.06936169 1.606297e-13 -0.2128161 148.5657 -0.08714175 -0.27128479 NA WRKY48
#with smaller matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G69780 0.475263698 9.071561e-12 0.7900089 6.4627143 0.14877009 0.6877791 NA ATHB13
# 12 AT4G31920 0.000000000 4.422050e-01 -0.2539478 1.7797840 -0.07590570 -0.2144058 NA ARR10
# 16 AT5G60200 0.000000000 6.627797e-01 -0.3210403 1.5411595 -0.06501800 -0.2618487 NA DOF5.3
# 14 AT5G16560 0.009221995 4.351465e-03 0.4371744 1.5307764 0.04171660 0.5045378 NA KAN
# 6 AT2G22430 0.000000000 8.916344e-01 -0.5026041 1.4761931 -0.11028331 -0.4729412 NA ATHB-6
# 3 AT1G63480 0.012537756 5.041249e-03 0.1390710 1.3466403 0.06012580 0.2322689 NA AT1G63480
# 5 AT1G80840 0.000000000 8.335613e-01 -0.4675158 1.1657589 -0.03904494 -0.3576951 NA WRKY40
# 7 AT2G46830 0.000000000 8.288264e-01 -0.2435563 0.9598848 -0.01694524 -0.1338295 NA CCA1
# 8 AT3G01470 0.000000000 6.096893e-01 0.3443222 0.7905707 0.04480747 0.3185114 NA ATHB-1
# 10 AT3G15270 0.000000000 5.967193e-01 -0.3595525 0.7188257 -0.02229638 -0.1956783 NA SPL5
#----------------------------------------------------------------------------------------
# 24th model: 47 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AGL42")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(24963074,24965075),
end=c(24965074,24967075),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AGL42",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x24 <- build(builder)
lapply(x24, dim)
tbl.model.24 <- x24$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.24$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.24$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT3G54810 0.00000000 0.084785117 -0.2644196 6.188577 -0.146332585 -0.200000000 NA GATA8
# 10 AT3G15270 0.07540106 0.006998546 0.3772227 4.869652 0.069474696 0.198559424 NA SPL5
# 24 AT5G65590 0.00000000 0.518332449 -0.1392564 4.552919 -0.018034682 -0.325138055 NA AT5G65590
# 7 AT2G37590 0.00000000 0.058326562 0.2887062 3.853160 0.223484191 0.307274910 NA DOF2.4
# 1 AT1G47655 0.00000000 0.179587983 -0.2823909 3.703040 -0.053497594 -0.145450180 NA AT1G47655
# 20 AT5G18830 0.00000000 0.122141996 -0.2049080 3.656628 -0.178760199 -0.091764706 NA SPL7
# 8 AT2G46130 0.00000000 0.268403642 0.1314209 3.474132 0.019499434 -0.006386555 NA WRKY43
# 17 AT4G40060 -0.38767399 0.002192146 -0.4233002 3.322573 -0.190071649 -0.348283313 NA ATHB-16
# 12 AT4G14465 0.00000000 0.824461667 -0.2942996 3.096320 -0.032989690 -0.171668667 NA AHL20
# 5 AT1G69780 0.00000000 0.480709469 -0.2050077 2.786546 -0.004923472 0.056902761 NA ATHB13
#----------------------------------------------------------------------------------------
# 25th model: 49 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "SPL7")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(6274114,6276115),
end=c(6276114,6278115),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="SPL7",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x25 <- build(builder)
lapply(x25, dim)
tbl.model.25 <- x25$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.25$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.25$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 18 AT4G34000 0.077602497 0.000558144 0.4708817 0.4809283 0.040375745 0.47716687 NA ABF3
# 10 AT2G46830 0.000000000 0.276421291 0.2673371 0.4729973 0.009602261 0.33474190 NA CCA1
# 30 AT5G65590 0.000000000 0.034660547 -0.2762644 0.4410077 -0.018100695 -0.22487395 NA AT5G65590
# 8 AT2G46590 0.005972524 0.023016935 0.2930937 0.3572919 0.022259839 -0.02722689 NA DAG2
# 21 AT5G03790 0.000000000 0.037914342 0.2938346 0.2258425 0.016411370 0.30919568 NA ATHB-51
# 11 AT3G01220 0.000000000 0.784879006 -0.1079931 0.2041922 -0.001507687 -0.08657863 NA ATHB20
#----------------------------------------------------------------------------------------
# 26th model: 42 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB20")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(71598,73599),
end=c(73598,74099),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB20",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x26 <- build(builder)
lapply(x26, dim)
tbl.model.26 <- x26$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.26$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.26$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 10 AT4G31920 0.41772711 3.406814e-101 0.4580854 381.0669 0.34747429 0.4738247 NA ARR10
# 14 AT5G25830 0.19961820 1.318223e-92 0.4445244 363.0023 0.17044102 0.4215165 NA GATA12
# 8 AT4G16110 0.43145786 7.786821e-97 0.4506042 361.5957 0.35814863 0.4004716 NA ARR2
# 3 AT1G69780 0.38328304 3.096823e-74 0.3917870 241.0333 0.29992065 0.3124582 NA ATHB13
# 4 AT2G22430 -0.33991092 1.593083e-39 -0.3110957 123.9119 -0.27593860 -0.2621734 NA ATHB-6
# 11 AT4G40060 0.16606026 1.371820e-26 0.2986074 121.7611 0.16935473 0.3205553 NA ATHB-16
# 20 AT5G60200 0.06673388 5.472103e-20 0.2569738 118.5619 0.06216835 0.2385958 NA DOF5.3
# 1 AT1G52150 0.00000000 9.359605e-01 0.2043231 112.4625 0.02076009 0.2827836 NA ATHB-15
# 19 AT5G56270 -0.11981506 3.764489e-21 -0.2444904 105.9166 -0.15869759 -0.2278204 NA WRKY2
# 12 AT5G12870 0.01968777 2.114678e-11 0.2610498 104.6406 0.04900232 0.2855275 NA MYB46
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 19 AT5G62000 -0.2862536 0.000191864 -0.5037641 13.112046 -0.11058991 -0.49848739 NA ARF2
# 8 AT2G23320 0.0000000 0.454605160 0.1141251 8.021538 0.09733246 0.06160864 NA WRKY15
# 4 AT1G64620 0.0000000 0.156291095 0.2507090 6.958213 0.17797144 0.17887155 NA AT1G64620
# 9 AT2G37590 0.1879416 0.017642741 0.3330993 6.609022 0.34492495 0.33618247 NA DOF2.4
# 7 AT2G22430 0.0000000 0.670812321 -0.3491920 6.598530 -0.10096907 -0.38477791 NA ATHB-6
# 18 AT5G60200 0.0000000 0.681120586 -0.2310342 4.441324 -0.10450002 -0.31130852 NA DOF5.3
# 13 AT4G37180 0.0000000 0.955245566 -0.1257275 4.421769 -0.01297474 -0.11769508 NA AT4G37180
# 2 AT1G08010 0.0000000 0.711167545 0.2947692 3.809657 0.10327200 0.28384154 NA GATA11
# 5 AT1G69780 0.0000000 0.200130094 0.3239425 3.625423 0.08096407 0.24504202 NA ATHB13
# 3 AT1G52150 0.0000000 0.779012361 -0.2248019 3.554240 -0.10040755 -0.24504202 NA ATHB-15
#----------------------------------------------------------------------------------------
# 27th model: 43 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB-15")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(19407912,19409913),
end=c(19409912,19410413),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB-15",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x27 <- build(builder)
lapply(x27, dim)
tbl.model.27 <- x27$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.27$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.27$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 12 AT5G60200 0.21340936 2.382589e-140 0.5291238 180.25009 0.15742201 0.5104056 NA DOF5.3
# 11 AT5G59780 0.12134591 9.156398e-89 0.4541901 134.24507 0.10896906 0.5001797 NA MYB59
# 9 AT5G12870 0.10519445 1.720585e-64 0.4073818 113.38724 0.09952470 0.3741304 NA MYB46
# 3 AT1G64620 0.17991695 1.974772e-73 0.4309093 104.51558 0.17044489 0.3959355 NA AT1G64620
# 6 AT2G37590 0.05159066 9.584809e-12 0.3314857 79.61398 0.09905095 0.3236278 NA DOF2.4
# 1 AT1G19850 0.05712985 7.918163e-11 0.2524720 57.30713 0.07933547 0.2915014 NA ARF5
# 7 AT3G01470 0.00000000 9.624542e-02 0.1885680 52.84702 -0.02167611 0.3273231 NA ATHB-1
# 15 AT5G65310 0.02036068 2.148224e-03 0.1811693 49.68740 0.07261525 0.2085411 NA ATHB-5
# 13 AT5G62165 0.00000000 9.062988e-01 0.2863012 42.93329 0.01439715 0.3785580 NA AGL42
# 4 AT1G80840 -0.03598329 1.281630e-13 -0.1940822 39.17065 -0.04269959 -0.2079710 NA WRKY40
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 20 AT5G62940 0.10801195 0.0006322369 0.4679888 0.8921942 0.065451480 0.135846339 NA DOF5.6
# 1 AT1G19850 0.00000000 0.0500041441 0.2097378 0.8527553 0.040026434 0.374597839 NA ARF5
# 12 AT4G17460 0.00000000 0.6630970577 -0.1509035 0.7576329 -0.002103981 0.044321729 NA HAT1
# 16 AT5G40350 0.04182587 0.0018847096 0.4204592 0.7557620 0.028697198 0.386506603 NA MYB24
# 21 AT5G65310 0.06294694 0.0005086301 0.4738772 0.6652933 0.042665227 0.151500600 NA ATHB-5
# 4 AT2G22430 0.00000000 0.7015060035 0.2227988 0.4832946 0.016838179 0.122785114 NA ATHB-6
# 8 AT3G01470 0.00000000 0.8829875080 -0.1533313 0.3471583 -0.014339995 0.001776711 NA ATHB-1
# 5 AT2G42200 0.00000000 0.8771819918 0.1964781 0.3297144 0.012000219 0.330804322 NA SPL9
# 10 AT3G15500 0.00000000 0.1752398645 -0.1873589 0.3038921 -0.014458599 -0.170900360 NA NAC055
# 2 AT1G53160 0.00000000 0.6867735006 0.1634598 0.2932131 0.016955481 0.004945978 NA SPL4
#----------------------------------------------------------------------------------------
# 28th model: 72 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB6")
tbl.atac <- data.frame(chrom=rep("Chr2", 2),
start=c(9522469,9526470),
end=c(9526469,9526970),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB6",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x28 <- build(builder)
lapply(x28, dim)
tbl.model.28 <- x28$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.28$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.28$symbol <- tf.symbols
#with bigger matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 26 AT5G25830 -0.09483151 1.140345e-60 -0.3607472 43.16806 -0.06685793 -0.33905727 NA GATA12
# 11 AT3G01470 0.22466705 1.211813e-22 0.1711951 42.22054 0.17356091 0.15181251 NA ATHB-1
# 17 AT3G59060 0.01548082 7.926326e-40 0.3216937 34.07054 0.01428551 0.24764231 NA PIF5
# 19 AT4G14465 -0.01061005 7.291409e-04 -0.1975364 28.95192 -0.01726298 -0.06137936 NA AHL20
# 27 AT5G39660 -0.12605465 1.095966e-23 -0.2464525 24.99268 -0.10654957 -0.21454257 NA CDF2
# 5 AT1G53170 0.05656884 4.104794e-15 0.2205160 24.07504 0.04516757 0.22032783 NA ERF8
# 3 AT1G50640 0.03543838 9.952316e-13 0.1508054 21.46977 0.05594105 0.11563041 NA ERF3
# 1 AT1G25550 -0.09614162 2.697151e-09 -0.1148429 21.07468 -0.10490239 -0.06847917 NA AT1G25550
# 30 AT5G62165 -0.01626665 2.727340e-15 -0.2656049 20.32351 -0.01924904 -0.24787569 NA AGL42
# 24 AT4G37180 0.03308813 3.732311e-08 0.1568264 16.61269 0.04429686 0.16640427 NA AT4G37180
#with smaller matrix
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 25 AT3G59060 0.00000000 6.013292e-03 0.4177917 1.4925109 0.04271919 0.3814166 NA PIF5
# 2 AT1G06160 0.00000000 1.376525e-02 -0.3846738 1.4547389 -0.01850690 -0.5465066 NA ERF094
# 16 AT2G37590 -0.25457628 1.593588e-04 -0.5006159 0.6920075 -0.13581099 -0.3996639 NA DOF2.4
# 32 AT5G16560 0.00000000 4.887221e-01 -0.4302894 0.5912776 -0.01893960 -0.4443217 NA KAN
# 9 AT1G53170 0.03921353 5.176469e-04 0.5012652 0.5717675 0.03725612 0.5334454 NA ERF8
# 5 AT1G28370 0.02500702 5.105083e-05 0.5403776 0.5613470 0.01986683 0.4971429 NA ERF11
# 27 AT4G31800 0.04091388 3.129935e-04 0.5047127 0.5010410 0.02207024 0.4797599 NA WRKY18
# 28 AT4G36540 -0.05821874 3.747159e-04 -0.5026041 0.4860311 -0.04020065 -0.4729412 NA BEE2
# 34 AT5G25830 0.00000000 6.219192e-03 -0.3594084 0.4298235 -0.02886218 -0.4058103 NA GATA12
# 23 AT3G23240 0.00000000 1.194863e-01 0.4357799 0.3651550 0.02055299 0.4164706 NA ERF1B
#----------------------------------------------------------------------------------------
# 29th model: 50 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "GATA8")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20294956,20296957),
end=c(20296956,20298957),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="GATA8",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x29 <- build(builder)
lapply(x29, dim)
tbl.model.29 <- x29$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.29$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.29$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 19 AT4G37260 0.1625935 8.031956e-07 0.6331905 1.4551616 0.07785364 0.5953902 NA MYB73
# 18 AT4G36930 0.1319216 5.388445e-05 0.5527549 1.1165373 0.07551803 0.5863625 NA SPT
# 27 AT5G62165 0.0000000 2.469158e-02 -0.2644196 0.9213861 -0.02436430 -0.2000000 NA AGL42
# 28 AT5G62940 0.0000000 4.853429e-01 -0.2108329 0.8014314 -0.01950336 -0.1296038 NA DOF5.6
# 21 AT5G16560 0.0000000 2.791695e-02 0.2876827 0.5131881 0.02265711 0.2155582 NA KAN
# 15 AT4G14465 0.0000000 2.568635e-01 0.4068371 0.5001389 0.02587057 0.3986074 NA AHL20
#----------------------------------------------------------------------------------------
# 30th model: 27 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "GATA15")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(2124658,2126659),
end=c(2126658,2127159),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="GATA15",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x30 <- build(builder)
lapply(x30, dim)
tbl.model.30 <- x30$model[1:9,]
tf.symbols <- unlist(lapply(tbl.model.30$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.30$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 1 AT1G42990 -0.33508042 1.351745e-54 -0.3428298 136.45933 -0.28399798 -0.33740759 NA BZIP60
# 3 AT4G16110 -0.11672350 3.434959e-07 -0.1189034 135.80445 -0.11646256 -0.11692962 NA ARR2
# 2 AT2G22430 -0.10248342 1.859476e-08 -0.1354756 101.82278 -0.10607455 -0.09941991 NA ATHB-6
# 4 AT4G31920 0.07838795 4.844631e-05 0.1150265 93.92049 0.08577908 0.10381406 NA ARR10
# 5 AT5G39660 -0.04991222 3.873510e-04 -0.1007136 93.64231 -0.06195625 -0.12276830 NA CDF2
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G80840 -0.1237662 1.995854e-05 -0.5639998 1.7139994 -0.05265239 -0.5307690 NA WRKY40
# 7 AT5G49520 0.0000000 8.609518e-01 -0.3637126 1.1599551 -0.02557242 -0.3389277 NA WRKY48
# 5 AT2G22430 0.0000000 8.372288e-01 -0.2135799 0.9022090 -0.02643684 -0.2063435 NA ATHB-6
# 1 AT1G42990 0.0000000 3.054324e-01 -0.3711281 0.8352476 -0.05579421 -0.3108358 NA BZIP60
# 9 AT5G65590 0.0000000 2.991368e-01 -0.1733936 0.6920649 -0.01608158 -0.1782996 NA AT5G65590
# 8 AT5G60200 0.0000000 8.251771e-01 -0.1577859 0.5924579 -0.02278387 -0.1964513 NA DOF5.3
# 2 AT1G49560 0.0000000 7.170586e-01 -0.1130704 0.5691463 -0.02138932 -0.1672069 NA AT1G49560
# 3 AT1G77920 0.0000000 3.590363e-01 -0.1727587 0.5415838 -0.02227929 -0.1709525 NA AT1G77920
# 6 AT5G39660 0.0000000 6.216311e-01 -0.1109419 0.4314407 -0.01836529 -0.1417081 NA CDF2
#----------------------------------------------------------------------------------------
# 31st model: 29 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "DOF5.3")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(24239077,24241078),
end=c(24241077,24241578),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="DOF5.3",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x31 <- build(builder)
lapply(x31, dim)
tbl.model.31 <- x31$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.31$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.31$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G14465 0.214155150 4.916909e-114 0.4832782 239.50481 0.15668213 0.44778587 NA AHL20
# 6 AT2G37590 0.357741795 1.568545e-90 0.4384045 188.61084 0.31449635 0.42497559 NA DOF2.4
# 2 AT1G64620 0.211534579 1.091095e-58 0.3815155 165.40192 0.19713607 0.41092741 NA AT1G64620
# 11 AT5G62165 0.070304535 7.690478e-39 0.3851670 130.60854 0.07760712 0.34801100 NA AGL42
# 4 AT2G18380 0.080615872 2.501859e-21 0.2916731 94.26407 0.09837057 0.28940058 NA GATA20
# 9 AT4G31920 0.000000000 6.329018e-02 0.1446562 78.76389 -0.04578754 0.20417106 NA ARR10
# 12 AT5G62940 0.072017247 3.750608e-13 0.2035832 77.73078 0.08849514 0.17916316 NA DOF5.6
# 10 AT5G25830 0.007394708 1.987848e-03 0.2516686 66.08657 0.04835927 0.26882677 NA GATA12
# 5 AT2G22430 0.000000000 1.158494e-01 -0.1197237 64.66402 0.03705229 -0.09882664 NA ATHB-6
# 3 AT1G80840 -0.001951971 1.292859e-02 -0.1299032 57.67901 -0.02536580 -0.13267635 NA WRKY40
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G80840 0.038016905 0.01034647 0.3594898 0.9927881 0.02919905 0.38986795 NA WRKY40
# 14 AT5G62940 0.051637742 0.01879464 0.3284058 0.9061619 0.05611126 0.25570228 NA DOF5.6
# 11 AT4G31920 0.000000000 0.64088629 -0.1142101 0.7730006 -0.02855464 -0.06228091 NA ARR10
# 2 AT1G64620 0.000000000 0.09435306 -0.2271095 0.6201841 -0.05787497 -0.16600240 NA AT1G64620
# 12 AT5G16560 0.000000000 0.23323122 -0.2672106 0.5990362 -0.01935544 -0.36211285 NA KAN
# 10 AT4G14465 0.000000000 0.87438075 -0.2027642 0.5986355 -0.01076157 -0.18943577 NA AHL20
# 9 AT3G15270 0.000000000 0.15020663 0.2614142 0.5646575 0.01586613 0.24081633 NA SPL5
# 13 AT5G62165 0.003856408 0.05513555 0.2602714 0.4449524 0.02247415 0.28172869 NA AGL42
# 8 AT3G06740 0.000000000 0.69841228 -0.1577859 0.4156874 -0.02077967 -0.19645130 NA GATA15
# 5 AT2G18380 0.000000000 0.35127576 0.1476492 0.4000534 0.01118307 0.02809124 NA GATA20
#----------------------------------------------------------------------------------------
# 32nd model: 47 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "DOF1.8")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(24005287,24007288),
end=c(24007287,24007788),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="DOF1.8",
type="noDNA.tfsSupplied",
matrix=mtx,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x32 <- build(builder)
lapply(x32, dim)
tbl.model.32 <- x32$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.32$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.32$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT5G60200 0.187105963 3.251037e-68 0.3815155 122.88080 0.143461584 0.41092741 NA DOF5.3
# 6 AT2G37590 0.195499776 6.374630e-33 0.3108857 108.78777 0.191921494 0.30756576 NA DOF2.4
# 9 AT4G37790 0.065707743 8.327646e-10 0.1851465 70.06470 0.073489682 0.21960667 NA HAT22
# 12 AT5G62940 0.023270984 3.497178e-04 0.1522472 61.25121 0.041208829 0.15454709 NA DOF5.6
# 4 AT1G75080 0.000000000 5.510751e-01 -0.1180820 59.31977 0.004938775 -0.09912931 NA BZR1
# 2 AT1G50640 0.020920599 4.737249e-03 0.1408610 57.08185 0.051081277 0.18047068 NA ERF3
# 7 AT4G14465 0.009766064 3.472301e-06 0.2410530 55.09684 0.029934282 0.27975820 NA AHL20
# 8 AT4G16110 0.000000000 6.983879e-01 0.1330114 50.03913 0.013573475 0.21742204 NA ARR2
# 5 AT2G23340 0.002045172 5.886952e-02 0.1147439 46.57284 0.024416134 0.12424610 NA DEAR3
# 3 AT1G53170 -0.032093620 1.722291e-05 -0.1726655 44.97815 -0.040581825 -0.17985127 NA ERF8
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT1G75080 0.000000000 0.07045807 0.2569291 0.8972444 0.065996112 0.26530612 NA BZR1
# 18 AT5G65210 0.000000000 0.06127087 0.2385765 0.5332703 0.045748444 0.19788715 NA TGA1
# 15 AT5G43410 0.008428567 0.02544334 0.3191712 0.5219050 0.022891551 0.28931573 NA AT5G43410
# 8 AT2G22430 0.000000000 0.05771801 -0.2717833 0.5065601 -0.038136257 -0.23697479 NA ATHB-6
# 12 AT3G23230 0.007146285 0.02207873 0.3231495 0.4882881 0.015521526 0.32235294 NA AT3G23230
# 2 AT1G19210 0.000000000 0.19981887 -0.1720897 0.3969417 -0.014377444 -0.09147659 NA AT1G19210
# 17 AT5G60200 0.000000000 0.09561391 -0.2271095 0.3778582 -0.047719742 -0.16600240 NA DOF5.3
# 1 AT1G06160 0.000000000 0.67509515 0.1302599 0.3752255 0.003923286 0.06055222 NA ERF094
# 14 AT4G37790 0.000000000 0.09713631 0.2282825 0.3720663 0.066419448 0.26492197 NA HAT22
# 13 AT4G16750 -0.016138806 0.02399706 -0.3204023 0.3581817 -0.026578976 -0.34588235 NA AT4G16750
#----------------------------------------------------------------------------------------
# 33rd model: 35 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AT5G65590")
tbl.atac <- data.frame(chrom=rep("Chr5", 2),
start=c(26210083,26212084),
end=c(26212083,26214084),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AT5G65590",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x33 <- build(builder)
lapply(x33, dim)
tbl.model.33 <- x33$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.33$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.33$symbol <- tf.symbols
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT2G01760 0.4033589 0.0005560655 0.4710025 9.069182 0.19982640 0.387851140 NA ARR14
# 21 AT5G02460 0.0000000 0.1215603958 0.1929624 6.025120 0.06524601 0.006194478 NA AT5G02460
# 14 AT3G50410 0.1970301 0.0043926087 0.3994689 5.597807 0.16865208 0.348955582 NA OBP1
# 16 AT4G16750 0.0000000 0.6450941741 0.2044169 5.379886 0.03300583 0.279231693 NA AT4G16750
# 9 AT2G42200 0.0000000 0.6635019982 0.2290212 4.347205 0.04489868 0.158799520 NA SPL9
# 15 AT4G14465 0.0000000 0.4708751326 0.1463848 4.095072 0.03802711 0.289699880 NA AHL20
#----------------------------------------------------------------------------------------
# 34th model: 28 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "MYC4")
tbl.atac <- data.frame(chrom=rep("Chr4", 2),
start=c(9931701,9933702),
end=c(9933701,9934202),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="MYC4",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x34 <- build(builder)
lapply(x34, dim)
tbl.model.34 <- x34$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.34$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.34$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT5G62000 0.03293216 2.059567e-05 0.1116754 95.07190 0.04064757 0.05220811 NA ARF2
# 3 AT3G47500 0.10271069 8.896262e-15 0.1744021 88.43036 0.09811551 0.15830419 NA CDF3
# 2 AT3G06740 0.10785022 2.374480e-09 0.1340536 88.02321 0.10871368 0.11134859 NA GATA15
# 5 AT5G39660 0.09237757 1.068600e-12 0.1690606 87.22162 0.09327233 0.12990382 NA CDF2
# 4 AT4G14465 0.07248091 3.511876e-15 0.1774471 84.26109 0.06375901 0.20562914 NA AHL20
# 7 AT5G62165 0.00709379 2.461137e-02 0.1026428 83.80349 0.01739119 0.12144408 NA AGL42
# 1 AT1G49560 0.10077876 1.331365e-10 0.1442764 75.67232 0.09751514 0.13397775 NA AT1G49560
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 5 AT3G06740 0.40927216 0.0002450054 0.4965225 3.209359 0.21101236 0.5101203 NA GATA15
# 6 AT3G47500 0.10391120 0.0138938349 0.3601382 2.830953 0.11856463 0.4494118 NA CDF3
# 3 AT2G22430 -0.13576117 0.0131700679 -0.3625533 2.570702 -0.10784324 -0.2595438 NA ATHB-6
# 8 AT5G16560 0.00000000 0.0756487998 0.3501211 2.274969 0.04497062 0.3777671 NA KAN
# 9 AT5G39660 0.09777027 0.0079118978 0.3307048 2.254376 0.07885702 0.1379592 NA CDF2
# 11 AT5G62940 0.00000000 0.1484288502 0.1361369 1.653449 0.05301587 0.1466026 NA DOF5.6
# 2 AT1G64620 0.00000000 0.2869265118 0.2155694 1.575104 0.07660245 0.1059784 NA AT1G64620
# 4 AT2G37590 0.00000000 0.1679122457 0.2146886 1.373392 0.10137605 0.1597599 NA DOF2.4
# 10 AT5G60200 -0.01546326 0.0570073327 -0.2469768 1.316233 -0.11839511 -0.1868427 NA DOF5.3
# 1 AT1G49560 0.00000000 0.7869961954 -0.2202242 1.166599 -0.05085925 -0.2133493 NA AT1G49560
#----------------------------------------------------------------------------------------
# 35th model: 48 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "AT1G25550")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(8974643,8976644),
end=c(8976643,8977143),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="AT1G25550",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x35 <- build(builder)
lapply(x35, dim)
tbl.model.35 <- x35$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.35$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.35$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 8 AT4G37180 0.1931458598 1.021758e-48 0.3242890 94.59979 0.163751975 0.30083819 NA AT4G37180
# 3 AT1G80840 0.1066981489 4.115582e-43 0.3056062 82.41714 0.088806587 0.30201622 NA WRKY40
# 4 AT2G22430 -0.1848360668 3.291719e-12 -0.1148429 69.93445 -0.168051347 -0.06847917 NA ATHB-6
# 7 AT4G36540 0.0522095676 4.995605e-09 0.1523035 60.67394 0.054368338 0.16142347 NA BEE2
# 6 AT3G06740 -0.0842381150 5.731672e-09 -0.1691969 55.23117 -0.089748937 -0.15614525 NA GATA15
# 9 AT4G38620 0.0151246394 2.285397e-05 0.1541935 45.27454 0.024560415 0.11727319 NA MYB4
# 5 AT2G44840 0.0000000000 4.091548e-01 0.1630523 43.35259 0.002397413 0.14819748 NA ERF13
# 10 AT5G49520 0.0161834476 1.974998e-03 0.1682330 42.05875 0.030072592 0.16698822 NA WRKY48
# 1 AT1G08000 -0.0340783357 2.511792e-05 -0.1266205 38.88968 -0.028013797 -0.15010301 NA GATA10
# 2 AT1G08010 -0.0009570583 1.000000e+00 -0.1266205 38.43543 -0.027596918 -0.15010301 NA GATA11
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 9 AT3G06740 -0.299907577 6.439146e-05 -0.5342536 1.8577021 -0.15109740 -0.5212130 NA GATA15
# 4 AT1G80840 0.060545281 4.300083e-04 0.5025548 1.7573561 0.04383603 0.5300840 NA WRKY40
# 7 AT2G44840 0.000000000 8.838719e-01 0.3936641 1.1996919 0.02183806 0.4226170 NA ERF13
# 21 AT5G16560 0.000000000 2.603671e-01 -0.2740614 1.1184633 -0.02773111 -0.3031453 NA KAN
# 11 AT3G54810 0.009909955 2.564185e-02 0.2725065 0.9697879 0.08001210 0.2621369 NA GATA8
# 13 AT4G00050 0.000000000 8.322380e-01 -0.1452447 0.9422956 -0.01770847 -0.1826170 NA UNE10
# 12 AT3G59060 0.000000000 9.752550e-01 0.1444829 0.8469312 0.01177003 0.2466747 NA PIF5
# 14 AT4G14465 0.000000000 1.443136e-01 0.1358985 0.8118220 0.02571154 0.1400720 NA AHL20
# 5 AT2G01760 0.000000000 8.058925e-01 0.1600451 0.7753703 0.03482297 0.2307323 NA ARR14
# 22 AT5G42630 0.013153562 1.765845e-02 0.2956409 0.7692770 0.02849112 0.2895078 NA ATS
#----------------------------------------------------------------------------------------
# 36th model: 47 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "WRKY40")
tbl.atac <- data.frame(chrom=rep("Chr1", 2),
start=c(30381833,30383834),
end=c(30383833,30384334),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="WRKY40",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x36 <- build(builder)
lapply(x36, dim)
tbl.model.36 <- x36$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.36$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.36$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 6 AT4G31800 0.51525632 4.964320e-198 0.6100412 702.2947 0.38318834 0.5895222 NA WRKY18
# 11 AT5G49520 0.20422611 4.091283e-24 0.3789303 312.8799 0.21953091 0.3569045 NA WRKY48
# 5 AT4G14465 -0.16862619 5.477636e-25 -0.2222530 157.1492 -0.12068589 -0.1928905 NA AHL20
# 8 AT4G40060 -0.08719258 2.176891e-11 -0.1802024 142.2125 -0.11461538 -0.1942995 NA ATHB-16
# 1 AT1G52150 -0.08624685 1.354613e-16 -0.1940822 138.6247 -0.10105451 -0.2079710 NA ATHB-15
# 7 AT4G31920 -0.02157285 2.252443e-03 -0.2124314 126.4829 -0.07640374 -0.2480439 NA ARR10
# 4 AT3G54810 0.05375731 6.408844e-07 0.1900307 121.3865 0.11027880 0.1646632 NA GATA8
# 12 AT5G60200 -0.04284976 4.494388e-07 -0.1299032 116.7443 -0.06072735 -0.1326763 NA DOF5.3
# 3 AT3G01470 0.00000000 4.510046e-02 -0.2207950 112.6439 -0.08703087 -0.2035526 NA ATHB-1
# 13 AT5G62165 0.00000000 2.372011e-01 -0.1474643 107.6476 -0.01547067 -0.1469365 NA AGL42
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 17 AT5G49520 0.30946554 2.050576e-08 0.6956844 16.702504 0.15112743 0.6376471 NA WRKY48
# 11 AT4G31800 0.14937448 7.223464e-06 0.6214864 8.312755 0.10574067 0.5374790 NA WRKY18
# 7 AT3G06740 -0.42878420 1.995821e-05 -0.5639998 5.991193 -0.31067054 -0.5307690 NA GATA15
# 9 AT3G50410 0.00000000 1.557519e-02 0.3614604 4.713995 0.12785945 0.4128211 NA OBP1
# 15 AT5G16560 0.00000000 8.691337e-02 -0.4332895 3.753391 -0.05814861 -0.5171188 NA KAN
# 2 AT1G69780 -0.02171626 3.770456e-03 -0.4954645 2.945440 -0.10626775 -0.4257863 NA ATHB13
# 3 AT2G22430 0.00000000 7.659822e-02 0.4055689 2.699469 0.12850483 0.3892917 NA ATHB-6
# 16 AT5G39660 0.00000000 9.408914e-01 0.1232301 2.434100 0.04390803 0.2660744 NA CDF2
# 5 AT2G46830 0.00000000 5.002095e-01 0.1257317 1.782495 0.01610316 0.2545498 NA CCA1
# 19 AT5G60200 0.00000000 1.777865e-01 0.3594898 1.689535 0.13931685 0.3898679 NA DOF5.3
#----------------------------------------------------------------------------------------
# 37th model: 63 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "HDG1")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(22628768,22630769),
end=c(22630768,22631269),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="HDG1",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x37 <- build(builder)
lapply(x37, dim)
tbl.model.37 <- x37$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.37$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.37$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 15 AT4G14465 -0.154765390 2.596098e-115 -0.4856727 237.18354 -0.09040606 -0.4259938 NA AHL20
# 28 AT5G62165 -0.143733663 8.555141e-99 -0.4655235 194.10680 -0.10233955 -0.4206713 NA AGL42
# 4 AT1G53170 0.111083778 4.538903e-39 0.3885351 123.54931 0.10857248 0.3412866 NA ERF8
# 27 AT5G60200 -0.003436034 3.476019e-03 -0.3182049 94.32478 -0.03456264 -0.3386358 NA DOF5.3
# 14 AT3G16770 0.095642215 1.287627e-23 0.2962935 89.88464 0.09575802 0.3110551 NA RAP2-3
# 23 AT5G42630 0.000000000 4.896472e-01 -0.2929949 85.81616 -0.02404363 -0.2472149 NA ATS
# 2 AT1G43160 -0.038001593 1.805203e-35 -0.3635980 72.15085 -0.03897048 -0.3642711 NA RAP2-6
# 19 AT5G16560 0.098398516 1.032615e-20 0.2124910 70.89221 0.10234046 0.1777708 NA KAN
# 1 AT1G22985 -0.168688436 1.072297e-27 -0.2819869 65.52232 -0.13920117 -0.2649606 NA AT1G22985
# 26 AT5G59780 0.000000000 9.785459e-01 -0.3170906 61.38923 -0.02360199 -0.3422669 NA MYB59
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G19850 -0.15160166 0.0001122043 -0.5191075 2.7433970 -0.07520242 -0.6030732 NA ARF5
# 13 AT3G06740 0.04224990 0.0058084445 0.3901419 1.2922685 0.07388667 0.4125909 NA GATA15
# 5 AT1G28370 -0.02436093 0.0025452370 -0.4350029 0.8931289 -0.02253084 -0.4381753 NA ERF11
# 10 AT2G33810 0.00000000 0.2602239950 -0.2489882 0.6723020 -0.01894787 -0.3287875 NA SPL3
# 20 AT4G14465 0.00000000 0.4723964614 0.3278602 0.6388994 0.01115103 0.3110204 NA AHL20
# 25 AT5G16560 0.00530587 0.0085767546 0.3717312 0.5917463 0.02276504 0.2917167 NA KAN
# 19 AT3G54810 0.09148248 0.0025801359 0.4166885 0.4226205 0.08816604 0.4079232 NA GATA8
# 7 AT1G50640 0.00000000 0.0297311200 -0.1685020 0.3406988 -0.04450800 -0.1623529 NA ERF3
# 15 AT3G15270 0.00000000 0.6326714126 -0.2212659 0.3184976 -0.01069675 -0.1128932 NA SPL5
# 16 AT3G16770 0.00000000 0.0438706191 0.2069915 0.3161507 0.02201526 0.1376711 NA RAP2-3
#----------------------------------------------------------------------------------------
# 38th model: 39 TFs with >= 95% motif match within 2000 upstream, 500 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "OBP3")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(20525197,20527198),
end=c(20527197,20527698),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="OBP3",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x38 <- build(builder)
lapply(x38, dim)
tbl.model.38 <- x38$model[1:10,]
tf.symbols <- unlist(lapply(tbl.model.38$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.38$symbol <- tf.symbols
#with bigger mtx
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 2 AT1G52150 0.30032344 7.613967e-139 0.5288124 294.84890 0.21362574 0.52041400 NA ATHB-15
# 20 AT5G60200 0.19500629 1.186395e-142 0.5328210 278.27108 0.15273043 0.49603968 NA DOF5.3
# 13 AT4G14465 0.21858158 9.676539e-138 0.5271657 245.22635 0.12879812 0.49331909 NA AHL20
# 8 AT2G03500 0.13145435 3.818057e-36 0.3335143 98.22692 0.11723378 0.34131853 NA AT2G03500
# 21 AT5G62165 0.06357558 1.869604e-32 0.4095408 85.46927 0.06649715 0.42466574 NA AGL42
# 11 AT2G37590 0.06264759 1.633947e-13 0.3418685 70.43422 0.11443598 0.32495722 NA DOF2.4
# 19 AT5G42630 0.00000000 6.830392e-01 0.2720615 69.88591 0.01116839 0.29793409 NA ATS
# 17 AT5G16560 -0.02219210 2.446977e-05 -0.1551953 69.29002 -0.05366961 -0.08497416 NA KAN
# 4 AT1G64620 0.03786746 1.173939e-09 0.3421884 67.12839 0.08976062 0.33598918 NA AT1G64620
# 12 AT3G01470 0.00000000 6.418641e-01 0.2890155 66.15996 0.04800213 0.34502592 NA ATHB-1
#with smaller mtx3
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT1G70920 0.00000000 0.0549188851 -0.3220010 2.4308873 -0.02119084 -0.29114046 NA ATHB18
# 16 AT5G42630 0.00000000 0.2775032985 -0.2703003 1.5977141 -0.01979709 -0.08235294 NA ATS
# 6 AT2G01930 0.07720251 0.0096555402 0.3393510 1.5553011 0.06560244 0.24206483 NA BPC1
# 3 AT1G63480 -0.08890034 0.0027604541 -0.4203422 1.4083491 -0.08789321 -0.33714286 NA AT1G63480
# 2 AT1G52150 0.00000000 0.5393362014 0.1994989 1.2331230 0.04341727 -0.06151261 NA ATHB-15
# 11 AT4G17460 0.00000000 0.0360625159 -0.3557152 1.2322980 -0.04324252 -0.20364946 NA HAT1
# 14 AT4G37790 0.00000000 0.0431609962 -0.3080342 1.0922844 -0.10320802 -0.21286915 NA HAT22
# 20 AT5G65310 0.12564916 0.0005152703 0.4734606 0.9821801 0.06209034 0.14468187 NA ATHB-5
# 19 AT5G62940 0.00000000 0.1403350982 0.3512926 0.9386974 0.06213477 0.12297719 NA DOF5.6
# 7 AT2G03500 0.00000000 0.9089545184 0.1293064 0.7918994 0.01891004 0.17051621 NA AT2G03500
#----------------------------------------------------------------------------------------
# 39th model: 34 TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "DOF2.4")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(15767292,15769293),
end=c(15769292,15771292),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="DOF2.4",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x39 <- build(builder)
lapply(x39, dim)
tbl.model.39 <- x39$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.39$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.39$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 4 AT2G22430 -0.151795953 0.000213524 -0.5006159 0.7358094 -0.074575421 -0.3996639 NA ATHB-6
# 1 AT1G24260 0.000000000 0.228304076 0.2310974 0.4672936 0.011216871 0.1334454 NA SEP3
# 9 AT3G47500 0.000000000 0.262294024 -0.2275964 0.4037608 -0.020777263 -0.2905642 NA CDF3
# 18 AT5G62165 0.002932115 0.031305813 0.2887062 0.3832981 0.018771658 0.3072749 NA AGL42
# 13 AT5G39610 0.000000000 0.154705994 -0.3126121 0.3506971 -0.019748291 -0.3306122 NA ATNAC2
# 6 AT2G46130 0.000000000 0.396256146 -0.1520034 0.3352320 -0.007884915 -0.1622569 NA WRKY43
#----------------------------------------------------------------------------------------
# 40th model: TFs with >= 95% motif match within 2000 upstream, 2000 downstream of TSS
# use both matrices mtx and mtx3
#----------------------------------------------------------------------------------------
targetGene <- canonicalizeName(tp, "ATHB13")
tbl.atac <- data.frame(chrom=rep("Chr3", 2),
start=c(26257165,26259166),
end=c(26259165,26261165),
stringsAsFactors=FALSE)
tbl.tfs <- findCandidateTranscriptionFactorsByMotifInSequence(tp, tbl.atac, 95L)
candidate.tfs <- sort(unique(tbl.tfs$orf))
length(candidate.tfs)
recipe <- list(title="ATHB13",
type="noDNA.tfsSupplied",
matrix=mtx3,
candidateTFs=candidate.tfs,
tfPool=getAllTranscriptionFactors(tp, "MotifDb"),
tfPrefilterCorrelation=0.1,
annotationDbFile=dbfile(org.At.tair.db),
orderModelByColumn="rfScore",
solverNames=c("lasso", "lassopv", "pearson", "randomForest", "ridge", "spearman"),
quiet=TRUE)
builder <- NoDnaModelBuilder(genome, targetGene, recipe, quiet=TRUE)
x40 <- build(builder)
lapply(x40, dim)
tbl.model.40 <- x40$model[1:6,]
tf.symbols <- unlist(lapply(tbl.model.40$gene, function(tf) getGeneNames(tp, tf)$symbol))
tbl.model.40$symbol <- tf.symbols
# gene betaLasso lassoPValue pearsonCoeff rfScore betaRidge spearmanCoeff bindingSites symbol
# 11 AT4G14465 0.19343639 9.251529e-08 0.6717750 6.812816 0.07842681 0.44105642 NA AHL20
# 12 AT4G31920 -0.02028084 8.951545e-03 -0.4447952 3.867082 -0.11947308 -0.21046819 NA ARR10
# 8 AT3G01470 0.19255800 2.463540e-04 0.5321812 2.562428 0.11928985 0.35750300 NA ATHB-1
# 18 AT5G60200 0.00000000 8.904937e-01 -0.3163195 1.875129 -0.06562144 -0.16686675 NA DOF5.3
# 5 AT2G22430 -0.01002657 1.165891e-02 -0.4936664 1.565940 -0.12507994 -0.62084034 NA ATHB-6
# 17 AT5G59780 0.00000000 5.213319e-01 0.3437988 1.378598 0.03966587 0.06314526 NA MYB59
#------------------------------------------------------------------------------------------------------------------------
if(!interactive())
runTests()
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.