Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Issue #13: FACS file parsing breaking #14

Merged
merged 3 commits into from
Dec 3, 2018
Merged

Issue #13: FACS file parsing breaking #14

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
9b68e21
Clean up code
robertaboukhalil Dec 1, 2018
80fc8a2
Clean up SoS plot
robertaboukhalil Dec 1, 2018
ae167ec
Fix issue #13
robertaboukhalil Dec 1, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
62 changes: 28 additions & 34 deletions scripts/process.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,6 +9,7 @@
# ------------------------------------------------------------------------------

# Config
minPloidy = 1.5
maxPloidy = 6
minBinWidth = 5
main_dir="/local1/work/ginkgo/scripts"
Expand All @@ -31,7 +32,7 @@ sex = as.numeric(args[[13]])
bb = as.numeric(args[[14]])

# Libraries
library('ctc')
library(ctc)
library(DNAcopy) # Segmentation
library(inline) # Use of c++
library(gplots) # Visual plotting of tables
Expand Down Expand Up @@ -59,8 +60,9 @@ bounds = read.table(paste("bounds_", bm, sep=""), header=FALSE, sep="\t")
setwd(user_dir)
raw = read.table(dat, header=TRUE, sep="\t")
ploidy = rbind(c(0,0), c(0,0))
if (f == 1 | f == 2)
if (f == 1 | f == 2) {
ploidy = read.table(facs, header=FALSE, sep="\t", as.is=TRUE)
}

# Remove bad bins
if (bb)
Expand Down Expand Up @@ -94,15 +96,17 @@ colors[3,] = c('brown2', 'blue4')
# Initialize data structures
l = dim(raw)[1] # Number of bins
w = dim(raw)[2] # Number of cells
n_ploidy = length(seq(1.5, maxPloidy, by=0.05)) # Number of ploidy tests during CN inference
breaks = matrix(0,l,w)
fixed = matrix(0,l,w)
final = matrix(0,l,w)
stats = matrix(0,w,10)
pos = cbind(c(1,bounds[,2]), c(bounds[,2], l))
# Initialize CN inference variables
CNgrid = seq(minPloidy, maxPloidy, by=0.05)
n_ploidy = length(CNgrid) # Number of ploidy tests during CN inference
CNmult = matrix(0,n_ploidy,w)
CNerror = matrix(0,n_ploidy,w)
outerColsums = matrix(0, (20*(maxPloidy-1.5)+1), w)
pos = cbind(c(1,bounds[,2]), c(bounds[,2], l))
outerColsums = matrix(0,n_ploidy,w)

# Normalize cells
normal = sweep(raw+1, 2, colMeans(raw+1), '/')
Expand Down Expand Up @@ -167,10 +171,11 @@ for(k in 1:w)
normal[,k] = lowess.gc( GC[,1], (raw[,k]+1)/mean(raw[,k]+1) )

# Compute log ratio between kth sample and reference
if (stat == 0)
if (stat == 0) {
lr = log2(normal[,k])
else
} else {
lr = log2((normal[,k])/(F))
}

# Determine breakpoints and extract chrom/locations
CNA.object = CNA(genomdat = lr, chrom = loc[,1], maploc = as.numeric(loc[,2]), data.type = 'logratio')
Expand Down Expand Up @@ -200,7 +205,6 @@ for(k in 1:w)
# ----------------------------------------------------------------------------

# Determine Copy Number
CNgrid = seq(1.5, maxPloidy, by=0.05)
outerRaw = fixed[,k] %o% CNgrid
outerRound = round(outerRaw)
outerDiff = (outerRaw - outerRound) ^ 2
Expand All @@ -212,6 +216,8 @@ for(k in 1:w)
CN = CNmult[1,k]
} else if (f == 1) {
CN = ploidy[which(lab[k]==ploidy[,1]),2]
# If user specified FACS file, still calculate CNerror
CNerror_facs = round( sort(colSums((round(fixed[,k] %o% c(CN)) - fixed[,k] %o% c(CN)) ^ 2, na.rm=FALSE, dims=1)), digits=2 )
} else {
estimate = ploidy[which(lab[k]==ploidy[,1]),2]
CN = CNmult[which(abs(CNmult[,k] - estimate)<.4),k][1]
Expand Down Expand Up @@ -390,19 +396,25 @@ for(k in 1:w)
jpeg(filename=paste(lab[k], "_SoS.jpeg", sep=""), width=2500, height=1500)

top = max(outerColsums[,k])
dat = data.frame(x=CNgrid, y=outerColsums[,k])
sosDat = data.frame(x=CNgrid, y=outerColsums[,k])
lim = cbind(c(seq(0,5000,500), 1000000), c(50, 100, 100, 200, 250, 400, 500, 500, 600, 600, 750, 1000))
step = lim[which(top<lim[,1])[1],]
minSoS = data.frame(x=CNmult[1,k], y=CNerror[1,k])
bestSoS = data.frame(x=CN, y=outerColsums[which(CNgrid==CN),k])
# If a FACS file is provided, use CNerror_facs, since CN multiplier could be
# outside the CNgrid range, which would cause "which(CNgrid==CN)" to error out
if(f == 1) {
bestSoS = data.frame(x=CN, y=CNerror_facs)
} else {
bestSoS = data.frame(x=CN, y=outerColsums[which(CNgrid==CN),k])
}

limit = c( min(minPloidy, CN), max(maxPloidy, CN))
plot1 = ggplot() +
geom_line(data=dat, aes(x=x, y=y), size=3) +
geom_point(data=dat, aes(x=x, y=y), shape=21, fill="black", size=5) +
geom_point(data=minSoS, aes(x=x, y=y*1.02, color="Minimum SoS Error"), shape=18, size=15) +
geom_point(data=bestSoS, aes(x=x, y=y*.98, color="Chosen Ploidy"), shape=18, size=15) +
scale_x_continuous(limits=c(1.5, 6), breaks=seq(1.5, 6, .5)) +
scale_y_continuous(limits=c(.5*min(outerColsums[,k]), top), breaks=seq(0, step[1], step[2])) +
geom_line(data=sosDat, aes(x=x, y=y), size=3) +
geom_point(data=sosDat, aes(x=x, y=y), shape=21, fill="black", size=5) +
geom_point(data=minSoS, aes(x=x, y=y, color="Minimum SoS Error"), shape=16, size=15, alpha=0.7) +
geom_point(data=bestSoS, aes(x=x, y=y, color="Chosen Ploidy"), shape=18, size=15, alpha=0.7) +
scale_x_continuous(limits=limit, breaks=seq(limit[1], limit[2], .5)) +
labs(title="Sum of Squares Error Across Potential Copy Number States", x="Copy Number Multiplier", y="Sum of Squares Error") +
theme(plot.title=element_text(size=45, vjust=1.5)) +
theme(axis.title.x=element_text(size=45, vjust=-2.8), axis.title.y=element_text(size=45, vjust=.1)) +
Expand Down Expand Up @@ -680,24 +692,6 @@ jpeg("heatCor.jpeg", width=2000, height=1400)
heatmap.2(t(finalBPs), Colv=FALSE, Rowv=as.dendrogram(clust3), margins=c(5,20), dendrogram="row", trace="none", xlab="Bins", ylab="Samples", cex.main=2, cex.axis=1.5, cex.lab=1.5, cexCol=.001, col=colorRampPalette(c("white","steelblue1","steelblue4","orange","sienna3"))(15), breaks=seq(0,step,step/15))
dev.off()

#jpeg("heatRaw.jpeg", width=2000, height=1400)
#heatmap3(t(rawBPs), distfun = function(x) dist(x), Colv=FALSE, Rowv=as.dendrogram(clust), margins=c(5,20), dendrogram="row", trace="none", xlab="Bins", ylab="Samples", cex.main=2, cex.axis=1.5, cex.lab=1.5, cexCol=.001, col=bluered(2))
#dev.off()
#
#step=quantile(fixedBPs, c(.98))[[1]]
#jpeg("heatNorm.jpeg", width=2000, height=1400)
#heatmap3(t(fixedBPs), distfun = function(x) dist(x), Colv=FALSE, Rowv=as.dendrogram(clust), margins=c(5,20), dendrogram="row", trace="none", xlab="Bins", ylab="Samples", cex.main=2, cex.axis=1.5, cex.lab=1.5, cexCol=.001, col=bluered(15), breaks=seq(0,step,step/15))
#dev.off()
#
#step=min(20, quantile(finalBPs, c(.98))[[1]])
#jpeg("heatCN.jpeg", width=2000, height=1400)
#heatmap3(t(finalBPs), distfun = function(x) dist(x), Colv=FALSE, Rowv=as.dendrogram(clust2), margins=c(5,20), dendrogram="row", trace="none", xlab="Bins", ylab="Samples", cex.main=2, cex.axis=1.5, cex.lab=1.5, cexCol=.001, col=colorRampPalette(c("white","green","green4","violet","purple"))(15), breaks=seq(0,step,step/15))
#dev.off()
#
#jpeg("heatCor.jpeg", width=2000, height=1400)
#heatmap3(t(finalBPs), distfun = function(x) dist(x), Colv=FALSE, Rowv=as.dendrogram(clust3), margins=c(5,20), dendrogram="row", trace="none", xlab="Bins", ylab="Samples", cex.main=2, cex.axis=1.5, cex.lab=1.5, cexCol=.001, col=colorRampPalette(c("white","steelblue1","steelblue4","orange","sienna3"))(15), breaks=seq(0,step,step/15))
#dev.off()

statusFile=file( paste(user_dir, "/", status, sep="") )
writeLines(c("<?xml version='1.0'?>", "<status>", "<step>3</step>", paste("<processingfile>Finished</processingfile>", sep=""), paste("<percentdone>100</percentdone>", sep=""), "<tree>clust.xml</tree>", "</status>"), statusFile)
close(statusFile)
Expand Down