In venelin/MGPMSimulations: Simulation Test for the Mixed Gaussian Phylogenetic Model
knitr::opts_chunk$set(echo = TRUE)
library(ggplot2)
library(PCMBase)
library(PCMFit)
library(MGPMSimulations)
library(data.table)
library(rmarkdown)
library(knitr)
library(ggplot2)
library(scales)
opts_chunk$set(dev='pdf',
#results="hide",
warning=FALSE,
dev.args=list(
family="ArialMT",
pointsize=10,
colormodel='rgb'
),
dpi=600, bg='white')
options(PCMBase.Value.NA = -1e20)
options(PCMBase.Lmr.mode = 11)
options(PCMBase.Threshold.EV = 1e-7)
options(width = 120)
ids <- testData_t5[, list(id = min(.I), .N), keyby=list(treeType, treeSize, numClusters)][, sort(id)]
treeTypesShort <- c(`non-ultrametric` = "Non-ultram.", `ultrametric` = "Ultram.")
plList <- lapply(ids, function(id) {
PCMTreePlot(PCMTree(testData_t5$treeWithRegimes[[id]]), layout="fan", open.angle=2, size=.25) +
ggtitle(paste0(LETTERS[match(id, ids)], " ",
treeTypesShort[testData_t5$treeType[[id]]], " / ",
testData_t5$treeSize[[id]], " / ",
"R=", testData_t5$numClusters[[id]])) +
theme(title = element_text(size = 9))
})
cowplot::plot_grid(plotlist = plList, nrow = 4, ncol = 4)
testData_t5_fitted <- MGPMSimulations::testData_t5[testData_t5_fittedIds]
data <- rbindlist(lapply(1:nrow(testData_t5_fitted), function(i) {
pl <- PCMPlotTraitData2D(
X = testData_t5_fitted$X[[i]][, 1:PCMTreeNumTips(testData_t5_fitted$treeWithRegimes[[i]])],
tree = PCMTree(testData_t5_fitted$treeWithRegimes[[i]]))
cbind(pl$data, testData_t5_fitted[i, list(
rowId = c(IdGlob, rep(as.integer(NA), nrow(pl$data)-1)),
x0 = c(1.0, rep(as.double(NA), nrow(pl$data) - 1)),
y0 = c(-1.0, rep(as.double(NA), nrow(pl$data) - 1)),
treeType, treeSize, numClusters, clusterNodes,
mapping,
logLik = c(logLik[[1]], rep(as.double(NA), nrow(pl$data)-1)),
AIC = c(AIC[[1]], rep(as.double(NA), nrow(pl$data)-1)),
nobs = c(nobs[[1]], rep(as.integer(NA), nrow(pl$data)-1)),
df = c(df[[1]], rep(as.integer(NA), nrow(pl$data)-1)),
IdMappingForClustering, IdParamForMapping, IdSimulationForParam)])
}))
data[, labLogLik:=sapply(logLik, function(ll) if(is.na(ll)) as.character(NA) else paste0("L: ", round(ll, 2)))]
data[, labAIC:=sapply(AIC, function(a) if(is.na(a)) as.character(NA) else paste0("AIC: ", round(a, 2)))]
data[, IdMappingLETTERS:=
paste0(IdMappingForClustering, ". ",
sapply(mapping, function(m) do.call(paste0, as.list(LETTERS[m]))))]
ScatterPlotSimulation <- function(dt) {
treeTypesShort <- c(`non-ultrametric` = "Non-ultram.", `ultrametric` = "Ultram.")
if(nrow(dt) >0 ) {
ggplot(dt, aes(x, y)) +
geom_point(aes(color = regime), size = .1, alpha = .5) +
geom_point(aes(x=x0, y=y0)) +
scale_x_continuous(limits = c(-15, 15)) +
scale_y_continuous(limits = c(-15, 15)) +
geom_label(aes(x=-14, y=12, label = rowId), hjust = "left") +
geom_text(aes(x=-13, y=6, label = labLogLik), size = 2.5, hjust = "left") +
facet_grid(paste0("Parameter ", IdParamForMapping)~paste0("Simulation ", IdSimulationForParam)) +
theme_bw() +
ggtitle(label = paste0(
treeTypesShort[dt$treeType[[1L]]], " tree",
" / ", dt$treeSize[[1L]],
" / R=", dt$numClusters[[1L]],
" / Mapping ", dt$IdMappingLETTERS[[1L]]))
} else {
NULL
}
}
listPlots <- c(
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=80" & numClusters==2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=80" & numClusters > 2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=80" & numClusters==2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=80" & numClusters>2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=159" & numClusters == 2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=159" & numClusters > 2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=159" & numClusters==2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=159" & numClusters>2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=318" & numClusters == 2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=318" & numClusters > 2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=318" & numClusters==2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=318" & numClusters>2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=638" & numClusters == 2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "ultrametric" & treeSize=="N=638" & numClusters > 2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=638" & numClusters==2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}),
lapply(1:4, function(i) {
dt <- data[treeType == "non-ultrametric" & treeSize=="N=638" & numClusters>2 & IdMappingForClustering == i]
ScatterPlotSimulation(dt)
}))
listPlots <- listPlots[!sapply(listPlots, is.null)]
for(p in listPlots) {
print(p)
}
dtSimulationFits <- rbindlist(list(
fits_TrueModels_t5[
, cbind(.SD, data.table(fitType2 = "MGPM A-F TRUE MLE q=n.a." ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = 0))],
fits_MGPM_A_F_all_AIC_t5[
, cbind(.SD, data.table(fitType2 = "MGPM A-F FULL AIC q=20" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]]
})))],
fits_MGPM_A_F_all_AIC2_t5[
, cbind(.SD, data.table(fitType2 = "MGPM A-F FULL AIC2 q=20" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]-length(unique(clusterNodes[[1]]))]
})))],
fits_MGPM_A_F_best_clade_RR_AIC_t5[
, cbind(.SD, data.table(fitType2 = "MGPM A-F RCP B.1 RR AIC q=20" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]]
})))],
fits_MGPM_A_F_best_clade_2_RR_AIC_t5[
, cbind(.SD, data.table(fitType2 = "MGPM A-F RCP B.2 RR AIC q=20" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]]
})))],
fits_MGPM_A_F_best_clade_2_AIC_t5[
, cbind(.SD, data.table(fitType2 = "MGPM A-F RCP B.2 AIC q=20" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]]
})))],
fits_SCALAROU_best_clade_2_AIC_t5[
, cbind(.SD, data.table(fitType2 = "SCALAR OU RCP AIC q=20" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]]
})))],
fits_SURFACE_best_clade_2_AICc_t5[
, cbind(.SD, data.table(fitType2 = "SURFACE RCP AICc q=20" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
p <- testData_t5[idGlob, df]
N <- testData_t5[idGlob, 2*nobs]
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1)
})))],
fits_SURFACE_best_clade_2_AICc_mcs10_t5[
, cbind(.SD, data.table(fitType2 = "SURFACE RCP AICc q=10" ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
p <- testData_t5[idGlob, df]
N <- testData_t5[idGlob, 2*nobs]
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1)
})))],
fits_surface_bwd_t5[
, cbind(.SD, data.table(fitType2 = "SURFACE FWD-BWD AICc q=n.a." ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
p <- testData_t5[idGlob, df]
N <- testData_t5[idGlob, 2*nobs]
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1)
})))],
fits_surface_fwd_t5[
, cbind(.SD, data.table(fitType2 = "SURFACE FWD AICc q=n.a." ),
data.table(
deltaNumRegimes =
sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) -
sapply(IdGlob, function(idGlob) {
testData_t5[idGlob, length(unique(clusterNodes[[1]]))]
}),
deltaScore = score -
sapply(IdGlob, function(idGlob) {
p <- testData_t5[idGlob, df]
N <- testData_t5[idGlob, 2*nobs]
fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1)
})))]
), fill = TRUE)
dtSimulationFits[, fitType:=factor(
fitType2,
levels = c("MGPM A-F TRUE MLE q=n.a.",
"MGPM A-F FULL AIC q=20" ,
"MGPM A-F FULL AIC2 q=20" ,
"MGPM A-F RCP B.2 RR AIC q=20",
"MGPM A-F RCP B.2 AIC q=20" ,
"MGPM A-F RCP B.1 RR AIC q=20" ,
"SCALAR OU RCP AIC q=20" ,
"SURFACE RCP AICc q=20" ,
"SURFACE RCP AICc q=10" ,
"SURFACE FWD-BWD AICc q=n.a." ,
"SURFACE FWD AICc q=n.a."))]
for(i in seq_len(nrow(dtSimulationFits))) {
dtSimulationFits[
i,
c("TreeType", "TreeSize", "Mapping") := {
idx <- list(IdTree[[1]],
IdClusteringForTree[[1]],
IdMappingForClustering[[1]],
IdParamForMapping[[1]])
subTestData <- testData_t5[
idx, list(treeType, treeSize, mapping)]
list(
subTestData$treeType[1],
subTestData$treeSize[1],
sapply(
subTestData$mapping[1],
function(m) {
do.call(paste0, as.list(names(MGPMDefaultModelTypes())[m]))
}))
}]
}
# The original calculation in the CalculateDistanceToTrueModelDistributions_t5.R
# script calculates the squared Mahalanobis distance, so we have to correct it.
dtSimulationFits[, dMahalanobis:=sqrt(dMahalanobis)]
dtSimulationFits[
, TreeType2:=factor(
TreeType, levels = c("ultrametric", "non-ultrametric"),
labels = c("Ultram.", "Non-ultram."))]
setnames(dtSimulationFits, c("TreeType", "TreeType2"), c("TreeType2", "TreeType"))
dtSimulationFits[
, TreeSize2:=factor(TreeSize, levels = paste0("N=", c(80,159,318,638)))]
setnames(dtSimulationFits, c("TreeSize", "TreeSize2"), c("TreeSize2", "TreeSize"))
dtSimulationFits[
, NumRegimes:=stringr::str_length(Mapping)]
dtSimulationFits[, ParameterGroup:=factor(IdParamForMapping <= 4, levels = c(TRUE, FALSE), labels = c("DistinctThetaOU", "SimilarThetaOU"))]
dtSimulationFits[, RegimeGroup:=factor(NumRegimes == 2, levels = c(TRUE, FALSE), labels = c("R=2", "R > 2"))]
usethis::use_data(dtSimulationFits, overwrite = TRUE)
dtFig <- dtSimulationFits[
IdGlob %in% testData_t5_fittedIds,
list(
TreeType = unique(TreeType),
TreeType2 = unique(TreeType2),
TreeSize = unique(TreeSize),
NumRegimes = unique(NumRegimes),
Mapping = unique(Mapping),
ParameterGroup = unique(ParameterGroup),
RegimeGroup = unique(RegimeGroup),
dBhattacharyya = mean(dBhattacharyya, na.rm = TRUE),
dMahalanobis = mean(dMahalanobis, na.rm = TRUE),
perf_Cluster_tpr = mean(perf_Cluster_tpr, na.rm = TRUE),
perf_Cluster_fpr = mean(perf_Cluster_fpr, na.rm = TRUE),
perf_BM_tpr = mean(perf_BM_tpr, na.rm = TRUE),
perf_BM_fpr = mean(perf_BM_fpr, na.rm = TRUE),
perf_OU_tpr = mean(perf_OU_tpr, na.rm = TRUE),
perf_OU_fpr = mean(perf_OU_fpr, na.rm = TRUE),
perf_Uncorrelated_tpr = mean(perf_Uncorrelated_tpr, na.rm = TRUE),
perf_Uncorrelated_fpr = mean(perf_Uncorrelated_fpr, na.rm = TRUE),
perf_Correlated_tpr = mean(perf_Correlated_tpr, na.rm = TRUE),
perf_Correlated_fpr = mean(perf_Correlated_fpr, na.rm = TRUE),
perf_NonDiagonalH_tpr = mean(perf_NonDiagonalH_tpr, na.rm = TRUE),
perf_NonDiagonalH_fpr = mean(perf_NonDiagonalH_fpr, na.rm = TRUE),
perf_AsymmetricH_tpr = mean(perf_AsymmetricH_tpr, na.rm = TRUE),
perf_AsymmetricH_fpr = mean(perf_AsymmetricH_fpr, na.rm = TRUE),
deltaNumRegimes = mean(deltaNumRegimes, na.rm = TRUE),
deltaScore = mean(deltaScore, na.rm = TRUE)
),
keyby = list(IdTree, IdClusteringForTree, IdMappingForClustering, IdParamForMapping, fitType)]
usethis::use_data(dtFig, overwrite = TRUE)
fitTypesBigger80 <- c(
"MGPM A-F TRUE MLE q=n.a.",
"MGPM A-F RCP B.1 RR AIC q=20" ,
"MGPM A-F RCP B.2 RR AIC q=20",
"MGPM A-F RCP B.2 AIC q=20" ,
"SCALAR OU RCP AIC q=20" ,
"SURFACE RCP AICc q=20")
usethis::use_data(fitTypesBigger80, overwrite = TRUE)
rmre <- colorRampPalette(c("red", muted("red")))(25)
yre <- colorRampPalette(c("yellow", "red"))(50)
greye <- colorRampPalette(c("green", "yellow"))(50)
blgre <- colorRampPalette(c("blue", "green"))(50)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(25)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = dMahalanobis))+
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Mahalanobis\ndistance\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25)
yre <- colorRampPalette(c("yellow", "red"))(50)
greye <- colorRampPalette(c("green", "yellow"))(50)
blgre <- colorRampPalette(c("blue", "green"))(50)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(25)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = dMahalanobis)) +
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Mahalanobis\ndistance\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25)
yre <- colorRampPalette(c("yellow", "red"))(50)
greye <- colorRampPalette(c("green", "yellow"))(50)
blgre <- colorRampPalette(c("blue", "green"))(50)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(25)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = dBhattacharyya))+
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Bhattacharyya\ndistance\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25)
yre <- colorRampPalette(c("yellow", "red"))(50)
greye <- colorRampPalette(c("green", "yellow"))(50)
blgre <- colorRampPalette(c("blue", "green"))(50)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(25)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = dBhattacharyya))+
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Bhattacharyya\ndistance\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(2)
yre <- colorRampPalette(c("yellow", "red"))(5)
greye <- colorRampPalette(c("green", "yellow"))(5)
blgre <- colorRampPalette(c("blue", "green"))(5)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(2)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = deltaNumRegimes))+
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-8, 8), space = "Lab", name="Number of regimes\ndifference\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(2)
yre <- colorRampPalette(c("yellow", "red"))(5)
greye <- colorRampPalette(c("green", "yellow"))(5)
blgre <- colorRampPalette(c("blue", "green"))(5)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(2)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = deltaNumRegimes))+
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-8, 8), space = "Lab", name="Number of regimes\ndifference\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25)
yre <- colorRampPalette(c("yellow", "red"))(50)
greye <- colorRampPalette(c("green", "yellow"))(50)
blgre <- colorRampPalette(c("blue", "green"))(50)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(25)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = deltaScore))+
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-200, 200), space = "Lab", name="Score\ndifference\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25)
yre <- colorRampPalette(c("yellow", "red"))(50)
greye <- colorRampPalette(c("green", "yellow"))(50)
blgre <- colorRampPalette(c("blue", "green"))(50)
mbl <- colorRampPalette(c(muted("blue"), "blue"))(25)
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = deltaScore))+
geom_tile(color = "white") +
scale_fill_gradientn(
colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-200, 200), space = "Lab", name="Score\ndifference\n") +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = perf_Cluster_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_Cluster_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = perf_Cluster_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_Cluster_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = perf_OU_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_OU_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = perf_OU_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_OU_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = perf_Correlated_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_Correlated_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = perf_Correlated_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_Correlated_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = perf_NonDiagonalH_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_NonDiagonalH_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = perf_NonDiagonalH_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_NonDiagonalH_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4],
aes(IdParamForMapping, fitType, fill = perf_AsymmetricH_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_AsymmetricH_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638")
fitTypes80 <- c(
fitTypesBigger80[1:4],
c("MGPM (A-F) full AIC q=20",
"MGPM (A-F) full AIC-2 q=20"))
fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3))
plotList <- lapply(seq_along(treeSizes), function(its)
ggplot(
data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4],
aes(IdParamForMapping, fitType, fill = perf_AsymmetricH_tpr))+
geom_tile(color = "white") +
geom_point(aes(color = perf_AsymmetricH_fpr), shape = 19, size = 1.5) +
scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) +
scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) +
theme_grey() +
coord_fixed(ratio = 1) +
xlab("Parameter set") +
theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) +
facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping))))
legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom"))
plTop <- cowplot::plot_grid(
plotlist = lapply(
plotList, function(pl) pl + theme(legend.position = "none")),
ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3)))
cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
options(digits=2)
dtSimulationFits[TreeSize == "N=80"
, summary(lm(deltaScore~fitType))]
dtSimulationFits[TreeSize == "N=80"
, summary(lm(dBhattacharyya~fitType))]
dtSimulationFits[TreeSize == "N=80"
, summary(lm(dMahalanobis~fitType))]
dtSimulationFits[TreeSize == "N=80"
, summary(lm(deltaNumRegimes~ fitType))]
dtSimulationFits[TreeSize == "N=80"
, summary(lm(perf_Cluster_tpr~fitType ))]
dtSimulationFits[TreeSize == "N=80"
, summary(lm(perf_Cluster_fpr~fitType))]
dtSimulationFits[
, summary(lm(dMahalanobis~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
options(digits=2)
dtSimulationFits[
, summary(lm(dBhattacharyya~fitType))]
dtSimulationFits[
, summary(lm(dBhattacharyya~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
options(digits=2)
dtSimulationFits[TreeSize == "N=80"
, summary(lm(deltaNumRegimes~ fitType))]
dtSimulationFits[TreeSize == "N=159"
, summary(lm(deltaNumRegimes~ fitType))]
dtSimulationFits[TreeSize == "N=318"
, summary(lm(deltaNumRegimes~ fitType))]
dtSimulationFits[TreeSize == "N=638" & deltaScore <=0
, summary(lm(deltaNumRegimes~ fitType))]
dtSimulationFits[
, summary(lm(deltaNumRegimes~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
options(digits = 2)
dtSimulationFits[
, summary(lm(perf_Cluster_tpr~fitType ))]
dtSimulationFits[
, summary(lm(perf_Cluster_fpr~fitType))]
dtSimulationFits[
, summary(lm(perf_Cluster_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType ))]
dtSimulationFits[
, summary(lm(perf_Cluster_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[
, summary(lm(perf_OU_tpr~ fitType))]
dtSimulationFits[
, summary(lm(perf_OU_fpr~ fitType))]
dtSimulationFits[
, summary(lm(perf_OU_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType ))]
dtSimulationFits[
, summary(lm(perf_OU_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType ))]
dtSimulationFits[
, summary(lm(perf_Correlated_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[
, summary(lm(perf_Correlated_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[
, summary(lm(perf_NonDiagonalH_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[
, summary(lm(perf_NonDiagonalH_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[
, summary(lm(perf_AsymmetricH_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[
, summary(lm(perf_AsymmetricH_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
venelin/MGPMSimulations documentation built on July 20, 2019, 11:03 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.
knitr::opts_chunk$set(echo = TRUE) library(ggplot2) library(PCMBase) library(PCMFit) library(MGPMSimulations) library(data.table) library(rmarkdown) library(knitr) library(ggplot2) library(scales) opts_chunk$set(dev='pdf', #results="hide", warning=FALSE, dev.args=list( family="ArialMT", pointsize=10, colormodel='rgb' ), dpi=600, bg='white') options(PCMBase.Value.NA = -1e20) options(PCMBase.Lmr.mode = 11) options(PCMBase.Threshold.EV = 1e-7) options(width = 120)
ids <- testData_t5[, list(id = min(.I), .N), keyby=list(treeType, treeSize, numClusters)][, sort(id)] treeTypesShort <- c(`non-ultrametric` = "Non-ultram.", `ultrametric` = "Ultram.") plList <- lapply(ids, function(id) { PCMTreePlot(PCMTree(testData_t5$treeWithRegimes[[id]]), layout="fan", open.angle=2, size=.25) + ggtitle(paste0(LETTERS[match(id, ids)], " ", treeTypesShort[testData_t5$treeType[[id]]], " / ", testData_t5$treeSize[[id]], " / ", "R=", testData_t5$numClusters[[id]])) + theme(title = element_text(size = 9)) }) cowplot::plot_grid(plotlist = plList, nrow = 4, ncol = 4)
testData_t5_fitted <- MGPMSimulations::testData_t5[testData_t5_fittedIds] data <- rbindlist(lapply(1:nrow(testData_t5_fitted), function(i) { pl <- PCMPlotTraitData2D( X = testData_t5_fitted$X[[i]][, 1:PCMTreeNumTips(testData_t5_fitted$treeWithRegimes[[i]])], tree = PCMTree(testData_t5_fitted$treeWithRegimes[[i]])) cbind(pl$data, testData_t5_fitted[i, list( rowId = c(IdGlob, rep(as.integer(NA), nrow(pl$data)-1)), x0 = c(1.0, rep(as.double(NA), nrow(pl$data) - 1)), y0 = c(-1.0, rep(as.double(NA), nrow(pl$data) - 1)), treeType, treeSize, numClusters, clusterNodes, mapping, logLik = c(logLik[[1]], rep(as.double(NA), nrow(pl$data)-1)), AIC = c(AIC[[1]], rep(as.double(NA), nrow(pl$data)-1)), nobs = c(nobs[[1]], rep(as.integer(NA), nrow(pl$data)-1)), df = c(df[[1]], rep(as.integer(NA), nrow(pl$data)-1)), IdMappingForClustering, IdParamForMapping, IdSimulationForParam)]) })) data[, labLogLik:=sapply(logLik, function(ll) if(is.na(ll)) as.character(NA) else paste0("L: ", round(ll, 2)))] data[, labAIC:=sapply(AIC, function(a) if(is.na(a)) as.character(NA) else paste0("AIC: ", round(a, 2)))] data[, IdMappingLETTERS:= paste0(IdMappingForClustering, ". ", sapply(mapping, function(m) do.call(paste0, as.list(LETTERS[m]))))] ScatterPlotSimulation <- function(dt) { treeTypesShort <- c(`non-ultrametric` = "Non-ultram.", `ultrametric` = "Ultram.") if(nrow(dt) >0 ) { ggplot(dt, aes(x, y)) + geom_point(aes(color = regime), size = .1, alpha = .5) + geom_point(aes(x=x0, y=y0)) + scale_x_continuous(limits = c(-15, 15)) + scale_y_continuous(limits = c(-15, 15)) + geom_label(aes(x=-14, y=12, label = rowId), hjust = "left") + geom_text(aes(x=-13, y=6, label = labLogLik), size = 2.5, hjust = "left") + facet_grid(paste0("Parameter ", IdParamForMapping)~paste0("Simulation ", IdSimulationForParam)) + theme_bw() + ggtitle(label = paste0( treeTypesShort[dt$treeType[[1L]]], " tree", " / ", dt$treeSize[[1L]], " / R=", dt$numClusters[[1L]], " / Mapping ", dt$IdMappingLETTERS[[1L]])) } else { NULL } } listPlots <- c( lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=80" & numClusters==2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=80" & numClusters > 2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=80" & numClusters==2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=80" & numClusters>2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=159" & numClusters == 2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=159" & numClusters > 2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=159" & numClusters==2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=159" & numClusters>2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=318" & numClusters == 2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=318" & numClusters > 2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=318" & numClusters==2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=318" & numClusters>2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=638" & numClusters == 2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "ultrametric" & treeSize=="N=638" & numClusters > 2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=638" & numClusters==2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) }), lapply(1:4, function(i) { dt <- data[treeType == "non-ultrametric" & treeSize=="N=638" & numClusters>2 & IdMappingForClustering == i] ScatterPlotSimulation(dt) })) listPlots <- listPlots[!sapply(listPlots, is.null)] for(p in listPlots) { print(p) }
dtSimulationFits <- rbindlist(list( fits_TrueModels_t5[ , cbind(.SD, data.table(fitType2 = "MGPM A-F TRUE MLE q=n.a." ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = 0))], fits_MGPM_A_F_all_AIC_t5[ , cbind(.SD, data.table(fitType2 = "MGPM A-F FULL AIC q=20" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] })))], fits_MGPM_A_F_all_AIC2_t5[ , cbind(.SD, data.table(fitType2 = "MGPM A-F FULL AIC2 q=20" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]-length(unique(clusterNodes[[1]]))] })))], fits_MGPM_A_F_best_clade_RR_AIC_t5[ , cbind(.SD, data.table(fitType2 = "MGPM A-F RCP B.1 RR AIC q=20" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] })))], fits_MGPM_A_F_best_clade_2_RR_AIC_t5[ , cbind(.SD, data.table(fitType2 = "MGPM A-F RCP B.2 RR AIC q=20" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] })))], fits_MGPM_A_F_best_clade_2_AIC_t5[ , cbind(.SD, data.table(fitType2 = "MGPM A-F RCP B.2 AIC q=20" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] })))], fits_SCALAROU_best_clade_2_AIC_t5[ , cbind(.SD, data.table(fitType2 = "SCALAR OU RCP AIC q=20" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] })))], fits_SURFACE_best_clade_2_AICc_t5[ , cbind(.SD, data.table(fitType2 = "SURFACE RCP AICc q=20" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { p <- testData_t5[idGlob, df] N <- testData_t5[idGlob, 2*nobs] fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1) })))], fits_SURFACE_best_clade_2_AICc_mcs10_t5[ , cbind(.SD, data.table(fitType2 = "SURFACE RCP AICc q=10" ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { p <- testData_t5[idGlob, df] N <- testData_t5[idGlob, 2*nobs] fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1) })))], fits_surface_bwd_t5[ , cbind(.SD, data.table(fitType2 = "SURFACE FWD-BWD AICc q=n.a." ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { p <- testData_t5[idGlob, df] N <- testData_t5[idGlob, 2*nobs] fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1) })))], fits_surface_fwd_t5[ , cbind(.SD, data.table(fitType2 = "SURFACE FWD AICc q=n.a." ), data.table( deltaNumRegimes = sapply(clusterNodes, function(cn) if(is.null(cn)) NA_real_ else length(unique(cn))) - sapply(IdGlob, function(idGlob) { testData_t5[idGlob, length(unique(clusterNodes[[1]]))] }), deltaScore = score - sapply(IdGlob, function(idGlob) { p <- testData_t5[idGlob, df] N <- testData_t5[idGlob, 2*nobs] fits_TrueModels_t5[IdGlob == idGlob, AIC[[1]]] + 2*p + (2*p*(p+1)) / (N - p -1) })))] ), fill = TRUE) dtSimulationFits[, fitType:=factor( fitType2, levels = c("MGPM A-F TRUE MLE q=n.a.", "MGPM A-F FULL AIC q=20" , "MGPM A-F FULL AIC2 q=20" , "MGPM A-F RCP B.2 RR AIC q=20", "MGPM A-F RCP B.2 AIC q=20" , "MGPM A-F RCP B.1 RR AIC q=20" , "SCALAR OU RCP AIC q=20" , "SURFACE RCP AICc q=20" , "SURFACE RCP AICc q=10" , "SURFACE FWD-BWD AICc q=n.a." , "SURFACE FWD AICc q=n.a."))] for(i in seq_len(nrow(dtSimulationFits))) { dtSimulationFits[ i, c("TreeType", "TreeSize", "Mapping") := { idx <- list(IdTree[[1]], IdClusteringForTree[[1]], IdMappingForClustering[[1]], IdParamForMapping[[1]]) subTestData <- testData_t5[ idx, list(treeType, treeSize, mapping)] list( subTestData$treeType[1], subTestData$treeSize[1], sapply( subTestData$mapping[1], function(m) { do.call(paste0, as.list(names(MGPMDefaultModelTypes())[m])) })) }] } # The original calculation in the CalculateDistanceToTrueModelDistributions_t5.R # script calculates the squared Mahalanobis distance, so we have to correct it. dtSimulationFits[, dMahalanobis:=sqrt(dMahalanobis)] dtSimulationFits[ , TreeType2:=factor( TreeType, levels = c("ultrametric", "non-ultrametric"), labels = c("Ultram.", "Non-ultram."))] setnames(dtSimulationFits, c("TreeType", "TreeType2"), c("TreeType2", "TreeType")) dtSimulationFits[ , TreeSize2:=factor(TreeSize, levels = paste0("N=", c(80,159,318,638)))] setnames(dtSimulationFits, c("TreeSize", "TreeSize2"), c("TreeSize2", "TreeSize")) dtSimulationFits[ , NumRegimes:=stringr::str_length(Mapping)] dtSimulationFits[, ParameterGroup:=factor(IdParamForMapping <= 4, levels = c(TRUE, FALSE), labels = c("DistinctThetaOU", "SimilarThetaOU"))] dtSimulationFits[, RegimeGroup:=factor(NumRegimes == 2, levels = c(TRUE, FALSE), labels = c("R=2", "R > 2"))] usethis::use_data(dtSimulationFits, overwrite = TRUE)
dtFig <- dtSimulationFits[ IdGlob %in% testData_t5_fittedIds, list( TreeType = unique(TreeType), TreeType2 = unique(TreeType2), TreeSize = unique(TreeSize), NumRegimes = unique(NumRegimes), Mapping = unique(Mapping), ParameterGroup = unique(ParameterGroup), RegimeGroup = unique(RegimeGroup), dBhattacharyya = mean(dBhattacharyya, na.rm = TRUE), dMahalanobis = mean(dMahalanobis, na.rm = TRUE), perf_Cluster_tpr = mean(perf_Cluster_tpr, na.rm = TRUE), perf_Cluster_fpr = mean(perf_Cluster_fpr, na.rm = TRUE), perf_BM_tpr = mean(perf_BM_tpr, na.rm = TRUE), perf_BM_fpr = mean(perf_BM_fpr, na.rm = TRUE), perf_OU_tpr = mean(perf_OU_tpr, na.rm = TRUE), perf_OU_fpr = mean(perf_OU_fpr, na.rm = TRUE), perf_Uncorrelated_tpr = mean(perf_Uncorrelated_tpr, na.rm = TRUE), perf_Uncorrelated_fpr = mean(perf_Uncorrelated_fpr, na.rm = TRUE), perf_Correlated_tpr = mean(perf_Correlated_tpr, na.rm = TRUE), perf_Correlated_fpr = mean(perf_Correlated_fpr, na.rm = TRUE), perf_NonDiagonalH_tpr = mean(perf_NonDiagonalH_tpr, na.rm = TRUE), perf_NonDiagonalH_fpr = mean(perf_NonDiagonalH_fpr, na.rm = TRUE), perf_AsymmetricH_tpr = mean(perf_AsymmetricH_tpr, na.rm = TRUE), perf_AsymmetricH_fpr = mean(perf_AsymmetricH_fpr, na.rm = TRUE), deltaNumRegimes = mean(deltaNumRegimes, na.rm = TRUE), deltaScore = mean(deltaScore, na.rm = TRUE) ), keyby = list(IdTree, IdClusteringForTree, IdMappingForClustering, IdParamForMapping, fitType)] usethis::use_data(dtFig, overwrite = TRUE) fitTypesBigger80 <- c( "MGPM A-F TRUE MLE q=n.a.", "MGPM A-F RCP B.1 RR AIC q=20" , "MGPM A-F RCP B.2 RR AIC q=20", "MGPM A-F RCP B.2 AIC q=20" , "SCALAR OU RCP AIC q=20" , "SURFACE RCP AICc q=20") usethis::use_data(fitTypesBigger80, overwrite = TRUE)
rmre <- colorRampPalette(c("red", muted("red")))(25) yre <- colorRampPalette(c("yellow", "red"))(50) greye <- colorRampPalette(c("green", "yellow"))(50) blgre <- colorRampPalette(c("blue", "green"))(50) mbl <- colorRampPalette(c(muted("blue"), "blue"))(25) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = dMahalanobis))+ geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Mahalanobis\ndistance\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25) yre <- colorRampPalette(c("yellow", "red"))(50) greye <- colorRampPalette(c("green", "yellow"))(50) blgre <- colorRampPalette(c("blue", "green"))(50) mbl <- colorRampPalette(c(muted("blue"), "blue"))(25) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = dMahalanobis)) + geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Mahalanobis\ndistance\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25) yre <- colorRampPalette(c("yellow", "red"))(50) greye <- colorRampPalette(c("green", "yellow"))(50) blgre <- colorRampPalette(c("blue", "green"))(50) mbl <- colorRampPalette(c(muted("blue"), "blue"))(25) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = dBhattacharyya))+ geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Bhattacharyya\ndistance\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25) yre <- colorRampPalette(c("yellow", "red"))(50) greye <- colorRampPalette(c("green", "yellow"))(50) blgre <- colorRampPalette(c("blue", "green"))(50) mbl <- colorRampPalette(c(muted("blue"), "blue"))(25) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = dBhattacharyya))+ geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(0, 200), space = "Lab", name="Bhattacharyya\ndistance\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(2) yre <- colorRampPalette(c("yellow", "red"))(5) greye <- colorRampPalette(c("green", "yellow"))(5) blgre <- colorRampPalette(c("blue", "green"))(5) mbl <- colorRampPalette(c(muted("blue"), "blue"))(2) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = deltaNumRegimes))+ geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-8, 8), space = "Lab", name="Number of regimes\ndifference\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(2) yre <- colorRampPalette(c("yellow", "red"))(5) greye <- colorRampPalette(c("green", "yellow"))(5) blgre <- colorRampPalette(c("blue", "green"))(5) mbl <- colorRampPalette(c(muted("blue"), "blue"))(2) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = deltaNumRegimes))+ geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-8, 8), space = "Lab", name="Number of regimes\ndifference\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25) yre <- colorRampPalette(c("yellow", "red"))(50) greye <- colorRampPalette(c("green", "yellow"))(50) blgre <- colorRampPalette(c("blue", "green"))(50) mbl <- colorRampPalette(c(muted("blue"), "blue"))(25) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = deltaScore))+ geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-200, 200), space = "Lab", name="Score\ndifference\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
rmre <- colorRampPalette(c("red", muted("red")))(25) yre <- colorRampPalette(c("yellow", "red"))(50) greye <- colorRampPalette(c("green", "yellow"))(50) blgre <- colorRampPalette(c("blue", "green"))(50) mbl <- colorRampPalette(c(muted("blue"), "blue"))(25) treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = deltaScore))+ geom_tile(color = "white") + scale_fill_gradientn( colours=c(mbl, blgre, greye, yre, rmre), na.value = "grey98", limits = c(-200, 200), space = "Lab", name="Score\ndifference\n") + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = perf_Cluster_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_Cluster_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes <- c(list(dtFig[, unique(fitType)]), rep(list(fitTypesBigger80), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = perf_Cluster_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_Cluster_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = perf_OU_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_OU_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = perf_OU_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_OU_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = perf_Correlated_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_Correlated_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = perf_Correlated_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_Correlated_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = perf_NonDiagonalH_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_NonDiagonalH_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = perf_NonDiagonalH_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_NonDiagonalH_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping <= 4], aes(IdParamForMapping, fitType, fill = perf_AsymmetricH_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_AsymmetricH_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
treeSizes <- c("N=80", "N=159", "N=318", "N=638") fitTypes80 <- c( fitTypesBigger80[1:4], c("MGPM (A-F) full AIC q=20", "MGPM (A-F) full AIC-2 q=20")) fitTypes <- c(list(fitTypes80), rep(list(fitTypesBigger80[1:4]), 3)) plotList <- lapply(seq_along(treeSizes), function(its) ggplot( data = dtFig[TreeSize == treeSizes[its] & fitType %in% fitTypes[[its]] & IdParamForMapping > 4], aes(IdParamForMapping, fitType, fill = perf_AsymmetricH_tpr))+ geom_tile(color = "white") + geom_point(aes(color = perf_AsymmetricH_fpr), shape = 19, size = 1.5) + scale_fill_gradient(space = "Lab", name="True positive rate \n", limits = c(0, 1)) + scale_color_gradient(space = "Lab", name="False positive rate \n", limits = c(0, 1)) + theme_grey() + coord_fixed(ratio = 1) + xlab("Parameter set") + theme(legend.position = "right", axis.title.y = element_blank(), strip.text = element_text(size = 7)) + facet_grid(TreeSize ~ TreeType + factor(Mapping, levels = unique(Mapping)))) legend <- cowplot::get_legend(plotList[[1]] + theme(legend.position = "bottom")) plTop <- cowplot::plot_grid( plotlist = lapply( plotList, function(pl) pl + theme(legend.position = "none")), ncol = 1, rel_heights = c(rep(0.7, 1), rep(0.52, 3))) cowplot::plot_grid(plTop, legend, ncol = 1, rel_heights = c(6, 1))
options(digits=2)
dtSimulationFits[TreeSize == "N=80" , summary(lm(deltaScore~fitType))] dtSimulationFits[TreeSize == "N=80" , summary(lm(dBhattacharyya~fitType))] dtSimulationFits[TreeSize == "N=80" , summary(lm(dMahalanobis~fitType))] dtSimulationFits[TreeSize == "N=80" , summary(lm(deltaNumRegimes~ fitType))] dtSimulationFits[TreeSize == "N=80" , summary(lm(perf_Cluster_tpr~fitType ))] dtSimulationFits[TreeSize == "N=80" , summary(lm(perf_Cluster_fpr~fitType))]
dtSimulationFits[ , summary(lm(dMahalanobis~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
options(digits=2) dtSimulationFits[ , summary(lm(dBhattacharyya~fitType))] dtSimulationFits[ , summary(lm(dBhattacharyya~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
options(digits=2) dtSimulationFits[TreeSize == "N=80" , summary(lm(deltaNumRegimes~ fitType))] dtSimulationFits[TreeSize == "N=159" , summary(lm(deltaNumRegimes~ fitType))] dtSimulationFits[TreeSize == "N=318" , summary(lm(deltaNumRegimes~ fitType))] dtSimulationFits[TreeSize == "N=638" & deltaScore <=0 , summary(lm(deltaNumRegimes~ fitType))] dtSimulationFits[ , summary(lm(deltaNumRegimes~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
options(digits = 2) dtSimulationFits[ , summary(lm(perf_Cluster_tpr~fitType ))] dtSimulationFits[ , summary(lm(perf_Cluster_fpr~fitType))] dtSimulationFits[ , summary(lm(perf_Cluster_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType ))] dtSimulationFits[ , summary(lm(perf_Cluster_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[ , summary(lm(perf_OU_tpr~ fitType))] dtSimulationFits[ , summary(lm(perf_OU_fpr~ fitType))] dtSimulationFits[ , summary(lm(perf_OU_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType ))] dtSimulationFits[ , summary(lm(perf_OU_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType ))]
dtSimulationFits[ , summary(lm(perf_Correlated_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))] dtSimulationFits[ , summary(lm(perf_Correlated_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[ , summary(lm(perf_NonDiagonalH_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))] dtSimulationFits[ , summary(lm(perf_NonDiagonalH_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
dtSimulationFits[ , summary(lm(perf_AsymmetricH_tpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))] dtSimulationFits[ , summary(lm(perf_AsymmetricH_fpr~TreeSize+TreeType+RegimeGroup+ParameterGroup + fitType))]
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.