juicer_func: Helper function for transforming a GRanges object into matrix...

In preciseTAD: preciseTAD: A machine learning framework for precise TAD boundary prediction

Description

Helper function for transforming a GRanges object into matrix form to be saved as .txt or .BED file and imported into juicer

Usage

juicer_func(grdat)

Arguments

grdat

A GRanges object representing boundary coordinates

Value

A dataframe that can be saved as a BED file to import into juicer

Examples

## Not run: 
# Read in ARROWHEAD-called TADs at 5kb
data(arrowhead_gm12878_5kb)

# Extract unique boundaries
bounds.GR <- extractBoundaries(domains.mat = arrowhead_gm12878_5kb,
                               preprocess = FALSE,
                               CHR = c("CHR21", "CHR22"),
                               resolution = 5000)

# Read in GRangesList of 26 TFBS
data(tfbsList)

tfbsList_filt <- tfbsList[which(names(tfbsList) %in%
                                                   c("Gm12878-Ctcf-Broad",
                                                     "Gm12878-Rad21-Haib",
                                                     "Gm12878-Smc3-Sydh",
                                                     "Gm12878-Znf143-Sydh"))]

# Create the binned data matrix for CHR1 (training) and CHR22 (testing)
# using 5 kb binning, distance-type predictors from 26 different TFBS from
# the GM12878 cell line, and random under-sampling
tadData <- createTADdata(bounds.GR = bounds.GR,
                         resolution = 5000,
                         genomicElements.GR = tfbsList_filt,
                         featureType = "distance",
                         resampling = "rus",
                         trainCHR = "CHR21",
                         predictCHR = "CHR22")

# Perform random forest using TADrandomForest by tuning mtry over 10 values
# using 3-fold CV
tadModel <- TADrandomForest(trainData = tadData[[1]],
                            testData = tadData[[2]],
                            tuneParams = list(mtry = 2,
                                            ntree = 500,
                                            nodesize = 1),
                            cvFolds = 3,
                            cvMetric = "Accuracy",
                            verbose = TRUE,
                            model = TRUE,
                            importances = TRUE,
                            impMeasure = "MDA",
                            performances = TRUE)

# Apply preciseTAD on a specific 2mb section of CHR22:17000000-19000000
pt <- preciseTAD(genomicElements.GR = tfbsList_filt,
                 featureType = "distance",
                 CHR = "CHR22",
                 chromCoords = list(17000000, 19000000),
                 tadModel = tadModel[[1]],
                 threshold = 1.0,
                 verbose = TRUE,
                 parallel = TRUE,
                 cores = 2,
                 splits = 2,
                 DBSCAN_params = list(10000, 3),
                 flank = NULL)

# Transform into juicer format
juicer_func(pt[[2]])

## End(Not run)

preciseTAD documentation built on Nov. 8, 2020, 6:51 p.m.