Mercurial > repos > davidvanzessen > prisca
view RScript.r @ 7:7ce82833977c draft
Uploaded
| author | davidvanzessen |
|---|---|
| date | Tue, 12 Dec 2017 04:53:20 -0500 |
| parents | a9d2ed661541 |
| children | eb2aa7cffca3 |
line wrap: on
line source
args <- commandArgs(trailingOnly = TRUE) options(scipen=999) inFile = args[1] outDir = args[2] logfile = args[3] min_freq = as.numeric(args[4]) min_cells = as.numeric(args[5]) mergeOn = args[6] cat("<html><table><tr><td>Starting analysis</td></tr>", file=logfile, append=F) library(ggplot2) library(reshape2) library(data.table) library(grid) library(parallel) #require(xtable) cat("<tr><td>Reading input</td></tr>", file=logfile, append=T) dat = read.table(inFile, header=T, sep="\t", dec=".", fill=T, stringsAsFactors=F) needed_cols = c("Patient", "Receptor", "Sample", "Cell_Count", "Clone_Molecule_Count_From_Spikes", "Log10_Frequency", "J_Segment_Major_Gene", "V_Segment_Major_Gene", "CDR3_Sense_Sequence", "Clone_Sequence") if(!all(needed_cols %in% names(dat))){ cat("Missing column(s):<br />", file=logfile, append=F) missing_cols = needed_cols[!(needed_cols %in% names(dat))] for(col in missing_cols){ cat(paste(col, "<br />"), file=logfile, append=T) } stop("Not all columns are present") } if(!("Total_Read_Count" %in% names(dat))){ dat$Total_Read_Count = 0 } dat = dat[,c("Patient", "Receptor", "Sample", "Cell_Count", "Clone_Molecule_Count_From_Spikes", "Log10_Frequency", "Total_Read_Count", "J_Segment_Major_Gene", "V_Segment_Major_Gene", "CDR3_Sense_Sequence", "Clone_Sequence")] dat = dat[!(nchar(dat$Clone_Sequence) < 2),] dat$dsPerM = 0 dat = dat[!is.na(dat$Patient),] dat$Related_to_leukemia_clone = F setwd(outDir) cat("<tr><td>Selecting first V/J Genes</td></tr>", file=logfile, append=T) dat$V_Segment_Major_Gene = as.factor(as.character(lapply(strsplit(as.character(dat$V_Segment_Major_Gene), "; "), "[[", 1))) dat$J_Segment_Major_Gene = as.factor(as.character(lapply(strsplit(as.character(dat$J_Segment_Major_Gene), "; "), "[[", 1))) cat("<tr><td>Calculating Frequency</td></tr>", file=logfile, append=T) dat$Frequency = ((10^dat$Log10_Frequency)*100) dat = dat[dat$Frequency <= 100,] #just in case? dat = dat[dat$Frequency >= min_freq,] patient.sample.counts = data.frame(data.table(dat)[, list(count=.N), by=c("Patient", "Sample")]) patient.sample.counts = data.frame(data.table(patient.sample.counts)[, list(count=.N), by=c("Patient")]) print("Found the following patients with number of samples:") print(patient.sample.counts) patient.sample.counts.pairs = patient.sample.counts[patient.sample.counts$count %in% 1:2,] patient.sample.counts.triplets = patient.sample.counts[patient.sample.counts$count == 3,] triplets = dat[dat$Patient %in% patient.sample.counts.triplets$Patient,] dat = dat[dat$Patient %in% patient.sample.counts.pairs$Patient,] cat("<tr><td>Normalizing to lowest cell count within locus</td></tr>", file=logfile, append=T) dat$locus_V = substring(dat$V_Segment_Major_Gene, 0, 4) dat$locus_J = substring(dat$J_Segment_Major_Gene, 0, 4) min_cell_count = data.frame(data.table(dat)[, list(min_cell_count=min(.SD$Cell_Count)), by=c("Patient", "locus_V", "locus_J")]) dat$min_cell_paste = paste(dat$Patient, dat$locus_V, dat$locus_J) min_cell_count$min_cell_paste = paste(min_cell_count$Patient, min_cell_count$locus_V, min_cell_count$locus_J) min_cell_count = min_cell_count[,c("min_cell_paste", "min_cell_count")] print(paste("rows:", nrow(dat))) dat = merge(dat, min_cell_count, by="min_cell_paste") print(paste("rows:", nrow(dat))) dat$normalized_read_count = round(dat$Clone_Molecule_Count_From_Spikes / dat$Cell_Count * dat$min_cell_count / 2, digits=2) #??????????????????????????????????? wel of geen / 2 dat = dat[dat$normalized_read_count >= min_cells,] dat$paste = paste(dat$Sample, dat$Clone_Sequence) patients = split(dat, dat$Patient, drop=T) intervalReads = rev(c(0,10,25,50,100,250,500,750,1000,10000)) intervalFreq = rev(c(0,0.01,0.05,0.1,0.5,1,5)) V_Segments = c(".*", "IGHV", "IGHD", "IGKV", "IGKV", "IgKINTR", "TRGV", "TRDV", "TRDD" , "TRBV") J_Segments = c(".*", ".*", ".*", "IGKJ", "KDE", ".*", ".*", ".*", ".*", ".*") Titles = c("Total", "IGH-Vh-Jh", "IGH-Dh-Jh", "Vk-Jk", "Vk-Kde" , "Intron-Kde", "TCRG", "TCRD-Vd-Dd", "TCRD-Dd-Dd", "TCRB-Vb-Jb") Titles = factor(Titles, levels=Titles) TitlesOrder = data.frame("Title"=Titles, "TitlesOrder"=1:length(Titles)) single_patients = dat[NULL,] patient.merge.list = list() #cache the 'both' table, 2x speedup for more memory... patient.merge.list.second = list() scatter_locus_data_list = list() cat(paste("<table border='0' style='font-family:courier;'>", sep=""), file="multiple_matches.html", append=T) cat(paste("<table border='0' style='font-family:courier;'>", sep=""), file="single_matches.html", append=T) patientCountOnColumn <- function(x, product, interval, on, appendtxt=F){ if (!is.data.frame(x) & is.list(x)){ x = x[[1]] } #x$Sample = factor(x$Sample, levels=unique(x$Sample)) x = data.frame(x,stringsAsFactors=F) onShort = "reads" if(on == "Frequency"){ onShort = "freq" } onx = paste(on, ".x", sep="") ony = paste(on, ".y", sep="") splt = split(x, x$Sample, drop=T) type="pair" if(length(splt) == 1){ print(paste(paste(x[1,which(colnames(x) == "Patient")]), "has one sample")) splt[[2]] = splt[[1]][NULL,] type="single" } patient1 = splt[[1]] patient2 = splt[[2]] oneSample = patient1[1,"Sample"] twoSample = patient2[1,"Sample"] patient = x[1,"Patient"] switched = F if(length(grep(".*_Right$", twoSample)) == 1 || length(grep(".*_Dx_BM$", twoSample)) == 1 || length(grep(".*_Dx$", twoSample)) == 1 ){ tmp = twoSample twoSample = oneSample oneSample = tmp tmp = patient1 patient1 = patient2 patient2 = tmp switched = T } if(appendtxt){ cat(paste(patient, oneSample, twoSample, type, sep="\t"), file="patients.txt", append=T, sep="", fill=3) } cat(paste("<tr><td>", patient, "</td>", sep=""), file=logfile, append=T) if(mergeOn == "Clone_Sequence"){ patient1$merge = paste(patient1$Clone_Sequence) patient2$merge = paste(patient2$Clone_Sequence) } else { patient1$merge = paste(patient1$V_Segment_Major_Gene, patient1$J_Segment_Major_Gene, patient1$CDR3_Sense_Sequence) patient2$merge = paste(patient2$V_Segment_Major_Gene, patient2$J_Segment_Major_Gene, patient2$CDR3_Sense_Sequence) } scatterplot_data_columns = c("Patient", "Sample", "Frequency", "normalized_read_count", "V_Segment_Major_Gene", "J_Segment_Major_Gene", "merge") scatterplot_data = patient1[NULL,scatterplot_data_columns] scatterplot.data.type.factor = c(oneSample, twoSample, paste(c(oneSample, twoSample), "In Both")) scatterplot_data$type = character(0) scatterplot_data$link = numeric(0) scatterplot_data$on = character(0) patientMerge = merge(patient1, patient2, by.x="merge", by.y="merge")[NULL,] #blegh cs.exact.matches = patient1[patient1$Clone_Sequence %in% patient2$Clone_Sequence,]$Clone_Sequence start.time = proc.time() merge.list = c() if(patient %in% names(patient.merge.list)){ patientMerge = patient.merge.list[[patient]] merge.list[["second"]] = patient.merge.list.second[[patient]] scatterplot_data = scatter_locus_data_list[[patient]] cat(paste("<td>", nrow(patient1), " in ", oneSample, " and ", nrow(patient2), " in ", twoSample, ", ", nrow(patientMerge), " in both (fetched from cache)</td></tr>", sep=""), file=logfile, append=T) #print(names(patient.merge.list)) } else { #fuzzy matching here... patient1.fuzzy = patient1 patient2.fuzzy = patient2 patient1.fuzzy$merge = paste(patient1.fuzzy$locus_V, patient1.fuzzy$locus_J) patient2.fuzzy$merge = paste(patient2.fuzzy$locus_V, patient2.fuzzy$locus_J) patient.fuzzy = rbind(patient1.fuzzy, patient2.fuzzy) patient.fuzzy = patient.fuzzy[order(nchar(patient.fuzzy$Clone_Sequence)),] merge.list = list() merge.list[["second"]] = vector() link.count = 1 while(nrow(patient.fuzzy) > 1){ first.merge = patient.fuzzy[1,"merge"] first.clone.sequence = patient.fuzzy[1,"Clone_Sequence"] first.sample = patient.fuzzy[1,"Sample"] merge.filter = first.merge == patient.fuzzy$merge #length.filter = nchar(patient.fuzzy$Clone_Sequence) - nchar(first.clone.sequence) <= 9 first.sample.filter = first.sample == patient.fuzzy$Sample second.sample.filter = first.sample != patient.fuzzy$Sample #first match same sample, sum to a single row, same for other sample #then merge rows like 'normal' sequence.filter = grepl(paste("^", first.clone.sequence, sep=""), patient.fuzzy$Clone_Sequence) #match.filter = merge.filter & grepl(first.clone.sequence, patient.fuzzy$Clone_Sequence) & length.filter & sample.filter first.match.filter = merge.filter & sequence.filter & first.sample.filter second.match.filter = merge.filter & sequence.filter & second.sample.filter first.rows = patient.fuzzy[first.match.filter,] second.rows = patient.fuzzy[second.match.filter,] first.rows.v = table(first.rows$V_Segment_Major_Gene) first.rows.v = names(first.rows.v[which.max(first.rows.v)]) first.rows.j = table(first.rows$J_Segment_Major_Gene) first.rows.j = names(first.rows.j[which.max(first.rows.j)]) first.sum = data.frame(merge = first.clone.sequence, Patient = patient, Receptor = first.rows[1,"Receptor"], Sample = first.rows[1,"Sample"], Cell_Count = first.rows[1,"Cell_Count"], Clone_Molecule_Count_From_Spikes = sum(first.rows$Clone_Molecule_Count_From_Spikes), Log10_Frequency = log10(sum(first.rows$Frequency)), Total_Read_Count = sum(first.rows$Total_Read_Count), dsPerM = sum(first.rows$dsPerM), J_Segment_Major_Gene = first.rows.j, V_Segment_Major_Gene = first.rows.v, Clone_Sequence = first.clone.sequence, CDR3_Sense_Sequence = first.rows[1,"CDR3_Sense_Sequence"], Related_to_leukemia_clone = F, Frequency = sum(first.rows$Frequency), locus_V = first.rows[1,"locus_V"], locus_J = first.rows[1,"locus_J"], min_cell_count = first.rows[1,"min_cell_count"], normalized_read_count = sum(first.rows$normalized_read_count), paste = first.rows[1,"paste"], min_cell_paste = first.rows[1,"min_cell_paste"]) if(nrow(second.rows) > 0){ second.rows.v = table(second.rows$V_Segment_Major_Gene) second.rows.v = names(second.rows.v[which.max(second.rows.v)]) second.rows.j = table(second.rows$J_Segment_Major_Gene) second.rows.j = names(second.rows.j[which.max(second.rows.j)]) second.sum = data.frame(merge = first.clone.sequence, Patient = patient, Receptor = second.rows[1,"Receptor"], Sample = second.rows[1,"Sample"], Cell_Count = second.rows[1,"Cell_Count"], Clone_Molecule_Count_From_Spikes = sum(second.rows$Clone_Molecule_Count_From_Spikes), Log10_Frequency = log10(sum(second.rows$Frequency)), Total_Read_Count = sum(second.rows$Total_Read_Count), dsPerM = sum(second.rows$dsPerM), J_Segment_Major_Gene = second.rows.j, V_Segment_Major_Gene = second.rows.v, Clone_Sequence = first.clone.sequence, CDR3_Sense_Sequence = second.rows[1,"CDR3_Sense_Sequence"], Related_to_leukemia_clone = F, Frequency = sum(second.rows$Frequency), locus_V = second.rows[1,"locus_V"], locus_J = second.rows[1,"locus_J"], min_cell_count = second.rows[1,"min_cell_count"], normalized_read_count = sum(second.rows$normalized_read_count), paste = second.rows[1,"paste"], min_cell_paste = second.rows[1,"min_cell_paste"]) #print(names(patientMerge)) #print(merge(first.sum, second.sum, by="merge")) patientMerge = rbind(patientMerge, merge(first.sum, second.sum, by="merge")) #print("test2") patient.fuzzy = patient.fuzzy[!(first.match.filter | second.match.filter),] hidden.clone.sequences = c(first.rows[-1,"Clone_Sequence"], second.rows[second.rows$Clone_Sequence != first.clone.sequence,"Clone_Sequence"]) merge.list[["second"]] = append(merge.list[["second"]], hidden.clone.sequences) tmp.rows = rbind(first.rows, second.rows) #print("test3") tmp.rows = tmp.rows[order(nchar(tmp.rows$Clone_Sequence)),] #add to the scatterplot data scatterplot.row = first.sum[,scatterplot_data_columns] scatterplot.row$type = paste(first.sum[,"Sample"], "In Both") scatterplot.row$link = link.count scatterplot.row$on = onShort scatterplot_data = rbind(scatterplot_data, scatterplot.row) scatterplot.row = second.sum[,scatterplot_data_columns] scatterplot.row$type = paste(second.sum[,"Sample"], "In Both") scatterplot.row$link = link.count scatterplot.row$on = onShort scatterplot_data = rbind(scatterplot_data, scatterplot.row) #write some information about the match to a log file if (nrow(first.rows) > 1 | nrow(second.rows) > 1) { cat(paste("<tr><td>", patient, " row ", 1:nrow(tmp.rows), "</td><td>", tmp.rows$Sample, ":</td><td>", tmp.rows$Clone_Sequence, "</td><td>", tmp.rows$normalized_read_count, "</td></tr>", sep=""), file="multiple_matches.html", append=T) } else { second.clone.sequence = second.rows[1,"Clone_Sequence"] if(nchar(first.clone.sequence) != nchar(second.clone.sequence)){ cat(paste("<tr bgcolor='#DDD'><td>", patient, " row ", 1:nrow(tmp.rows), "</td><td>", tmp.rows$Sample, ":</td><td>", tmp.rows$Clone_Sequence, "</td><td>", tmp.rows$normalized_read_count, "</td></tr>", sep=""), file="single_matches.html", append=T) } else { #cat(paste("<tr><td>", patient, " row ", 1:nrow(tmp.rows), "</td><td>", tmp.rows$Sample, ":</td><td>", tmp.rows$Clone_Sequence, "</td><td>", tmp.rows$normalized_read_count, "</td></tr>", sep=""), file="single_matches.html", append=T) } } } else if(nrow(first.rows) > 1) { if(patient1[1,"Sample"] == first.sample){ patient1 = patient1[!(patient1$Clone_Sequence %in% first.rows$Clone_Sequence),] patient1 = rbind(patient1, first.sum) } else { patient2 = patient2[!(patient2$Clone_Sequence %in% first.rows$Clone_Sequence),] patient2 = rbind(patient2, first.sum) } hidden.clone.sequences = c(first.rows[-1,"Clone_Sequence"]) merge.list[["second"]] = append(merge.list[["second"]], hidden.clone.sequences) patient.fuzzy = patient.fuzzy[-first.match.filter,] #add to the scatterplot data scatterplot.row = first.sum[,scatterplot_data_columns] scatterplot.row$type = first.sum[,"Sample"] scatterplot.row$link = link.count scatterplot.row$on = onShort scatterplot_data = rbind(scatterplot_data, scatterplot.row) cat(paste("<tr bgcolor='#DDF'><td>", patient, " row ", 1:nrow(first.rows), "</td><td>", first.rows$Sample, ":</td><td>", first.rows$Clone_Sequence, "</td><td>", first.rows$normalized_read_count, "</td></tr>", sep=""), file="single_matches.html", append=T) } else { patient.fuzzy = patient.fuzzy[-1,] #add to the scatterplot data scatterplot.row = first.sum[,scatterplot_data_columns] scatterplot.row$type = first.sum[,"Sample"] scatterplot.row$link = link.count scatterplot.row$on = onShort scatterplot_data = rbind(scatterplot_data, scatterplot.row) } link.count = link.count + 1 } patient.merge.list[[patient]] <<- patientMerge patient.merge.list.second[[patient]] <<- merge.list[["second"]] sample.order = data.frame(type = c(oneSample, twoSample, paste(c(oneSample, twoSample), "In Both")),type.order = 1:4) scatterplot_data = merge(scatterplot_data, sample.order, by="type") scatter_locus_data_list[[patient]] <<- scatterplot_data cat(paste("<td>", nrow(patient1), " in ", oneSample, " and ", nrow(patient2), " in ", twoSample, ", ", nrow(patientMerge), " in both (finding both took ", (proc.time() - start.time)[[3]], "s)</td></tr>", sep=""), file=logfile, append=T) } patient1 = patient1[!(patient1$Clone_Sequence %in% patient.merge.list.second[[patient]]),] patient2 = patient2[!(patient2$Clone_Sequence %in% patient.merge.list.second[[patient]]),] patientMerge$thresholdValue = pmax(patientMerge[,onx], patientMerge[,ony]) #patientMerge$thresholdValue = pmin(patientMerge[,onx], patientMerge[,ony]) res1 = vector() res2 = vector() resBoth = vector() read1Count = vector() read2Count = vector() locussum1 = vector() locussum2 = vector() #for(iter in 1){ for(iter in 1:length(product[,1])){ threshhold = product[iter,"interval"] V_Segment = paste(".*", as.character(product[iter,"V_Segments"]), ".*", sep="") J_Segment = paste(".*", as.character(product[iter,"J_Segments"]), ".*", sep="") #both = (grepl(V_Segment, patientMerge$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge$J_Segment_Major_Gene.x) & patientMerge[,onx] > threshhold & patientMerge[,ony] > threshhold) #both higher than threshold both = (grepl(V_Segment, patientMerge$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge$J_Segment_Major_Gene.x) & patientMerge$thresholdValue > threshhold) #highest of both is higher than threshold one = (grepl(V_Segment, patient1$V_Segment_Major_Gene) & grepl(J_Segment, patient1$J_Segment_Major_Gene) & patient1[,on] > threshhold & !(patient1$merge %in% patientMerge[both,]$merge)) two = (grepl(V_Segment, patient2$V_Segment_Major_Gene) & grepl(J_Segment, patient2$J_Segment_Major_Gene) & patient2[,on] > threshhold & !(patient2$merge %in% patientMerge[both,]$merge)) read1Count = append(read1Count, sum(patient1[one,]$normalized_read_count)) read2Count = append(read2Count, sum(patient2[two,]$normalized_read_count)) res1 = append(res1, sum(one)) res2 = append(res2, sum(two)) resBoth = append(resBoth, sum(both)) locussum1 = append(locussum1, sum(patient1[(grepl(V_Segment, patient1$V_Segment_Major_Gene) & grepl(J_Segment, patient1$J_Segment_Major_Gene)),]$normalized_read_count)) locussum2 = append(locussum2, sum(patient2[(grepl(V_Segment, patient2$V_Segment_Major_Gene) & grepl(J_Segment, patient2$J_Segment_Major_Gene)),]$normalized_read_count)) #threshhold = 0 if(threshhold != 0 | T){ if(sum(one) > 0){ dfOne = patient1[one,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone")] colnames(dfOne) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Clone Sequence", "Related_to_leukemia_clone") filenameOne = paste(oneSample, "_", product[iter, "Titles"], "_", threshhold, sep="") write.table(dfOne, file=paste(filenameOne, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } if(sum(two) > 0){ dfTwo = patient2[two,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone")] colnames(dfTwo) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Clone Sequence", "Related_to_leukemia_clone") filenameTwo = paste(twoSample, "_", product[iter, "Titles"], "_", threshhold, sep="") write.table(dfTwo, file=paste(filenameTwo, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } } scatterplot_locus_data = scatterplot_data[grepl(V_Segment, scatterplot_data$V_Segment_Major_Gene) & grepl(J_Segment, scatterplot_data$J_Segment_Major_Gene),] if(nrow(scatterplot_locus_data) > 0){ scatterplot_locus_data$Rearrangement = product[iter, "Titles"] } p = NULL #print(paste("nrow scatterplot_locus_data", nrow(scatterplot_locus_data))) if(nrow(scatterplot_locus_data) != 0){ if(on == "normalized_read_count"){ write.table(scatterplot_locus_data, file=paste(oneSample, twoSample, product[iter, "Titles"], "scatterplot_locus_data.txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) scales = 10^(0:6) #(0:ceiling(log10(max(scatterplot_locus_data$normalized_read_count)))) p = ggplot(scatterplot_locus_data, aes(factor(reorder(type, type.order)), normalized_read_count, group=link)) + geom_line() + scale_y_log10(breaks=scales,labels=scales, limits=c(1,1e6)) + scale_x_discrete(breaks=levels(scatterplot_data$type), labels=levels(scatterplot_data$type), drop=FALSE) } else { p = ggplot(scatterplot_locus_data, aes(factor(reorder(type, type.order)), Frequency, group=link)) + geom_line() + scale_y_log10(limits=c(0.0001,100), breaks=c(0.0001, 0.001, 0.01, 0.1, 1, 10, 100), labels=c("0.0001", "0.001", "0.01", "0.1", "1", "10", "100")) + scale_x_discrete(breaks=levels(scatterplot_data$type), labels=levels(scatterplot_data$type), drop=FALSE) } p = p + geom_point(aes(colour=type), position="dodge") p = p + xlab("In one or both samples") + ylab(onShort) + ggtitle(paste(patient1[1,"Patient"], patient1[1,"Sample"], patient2[1,"Sample"], onShort, product[iter, "Titles"])) } else { p = ggplot(NULL, aes(x=c("In one", "In Both"),y=0)) + geom_blank(NULL) + xlab("In one or both of the samples") + ylab(onShort) + ggtitle(paste(patient1[1,"Patient"], patient1[1,"Sample"], patient2[1,"Sample"], onShort, product[iter, "Titles"])) } png(paste(patient1[1,"Patient"], "_", patient1[1,"Sample"], "_", patient2[1,"Sample"], "_", onShort, "_", product[iter, "Titles"],"_scatter.png", sep="")) print(p) dev.off() if(sum(both) > 0){ dfBoth = patientMerge[both,c("V_Segment_Major_Gene.x", "J_Segment_Major_Gene.x", "normalized_read_count.x", "Frequency.x", "Related_to_leukemia_clone.x", "Clone_Sequence.x", "V_Segment_Major_Gene.y", "J_Segment_Major_Gene.y", "normalized_read_count.y", "Frequency.y", "Related_to_leukemia_clone.y")] colnames(dfBoth) = c(paste("Proximal segment", oneSample), paste("Distal segment", oneSample), paste("Normalized_Read_Count", oneSample), paste("Frequency", oneSample), paste("Related_to_leukemia_clone", oneSample),"Clone Sequence", paste("Proximal segment", twoSample), paste("Distal segment", twoSample), paste("Normalized_Read_Count", twoSample), paste("Frequency", twoSample), paste("Related_to_leukemia_clone", twoSample)) filenameBoth = paste(oneSample, "_", twoSample, "_", product[iter, "Titles"], "_", threshhold, sep="") write.table(dfBoth, file=paste(filenameBoth, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } } patientResult = data.frame("Locus"=product$Titles, "J_Segment"=product$J_Segments, "V_Segment"=product$V_Segments, "cut_off_value"=paste(">", product$interval, sep=""), "Both"=resBoth, "tmp1"=res1, "read_count1" = round(read1Count), "tmp2"=res2, "read_count2"= round(read2Count), "Sum"=res1 + res2 + resBoth, "percentage" = round((resBoth/(res1 + res2 + resBoth)) * 100, digits=2), "Locus_sum1"=locussum1, "Locus_sum2"=locussum2) if(sum(is.na(patientResult$percentage)) > 0){ patientResult[is.na(patientResult$percentage),]$percentage = 0 } colnames(patientResult)[6] = oneSample colnames(patientResult)[8] = twoSample colnamesBak = colnames(patientResult) colnames(patientResult) = c("Ig/TCR gene rearrangement type", "Distal Gene segment", "Proximal gene segment", "cut_off_value", paste("Number of sequences ", patient, "_Both", sep=""), paste("Number of sequences", oneSample, sep=""), paste("Normalized Read Count", oneSample), paste("Number of sequences", twoSample, sep=""), paste("Normalized Read Count", twoSample), paste("Sum number of sequences", patient), paste("Percentage of sequences ", patient, "_Both", sep=""), paste("Locus Sum", oneSample), paste("Locus Sum", twoSample)) write.table(patientResult, file=paste(patient, "_", onShort, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) colnames(patientResult) = colnamesBak patientResult$Locus = factor(patientResult$Locus, Titles) patientResult$cut_off_value = factor(patientResult$cut_off_value, paste(">", interval, sep="")) plt = ggplot(patientResult[,c("Locus", "cut_off_value", "Both")]) plt = plt + geom_bar( aes( x=factor(cut_off_value), y=Both), stat='identity', position="dodge", fill="#79c36a") plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1)) plt = plt + geom_text(aes(ymax=max(Both), x=cut_off_value,y=Both,label=Both), angle=90, hjust=0) plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("Number of clones in both") plt = plt + theme(plot.margin = unit(c(1,8.8,0.5,1.5), "lines")) png(paste(patient, "_", onShort, ".png", sep=""), width=1920, height=1080) print(plt) dev.off() #(t,r,b,l) plt = ggplot(patientResult[,c("Locus", "cut_off_value", "percentage")]) plt = plt + geom_bar( aes( x=factor(cut_off_value), y=percentage), stat='identity', position="dodge", fill="#79c36a") plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1)) plt = plt + geom_text(aes(ymax=max(percentage), x=cut_off_value,y=percentage,label=percentage), angle=90, hjust=0) plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("% clones in both left and right") plt = plt + theme(plot.margin = unit(c(1,8.8,0.5,1.5), "lines")) png(paste(patient, "_percent_", onShort, ".png", sep=""), width=1920, height=1080) print(plt) dev.off() patientResult = melt(patientResult[,c('Locus','cut_off_value', oneSample, twoSample)] ,id.vars=1:2) patientResult$relativeValue = patientResult$value * 10 patientResult[patientResult$relativeValue == 0,]$relativeValue = 1 plt = ggplot(patientResult) plt = plt + geom_bar( aes( x=factor(cut_off_value), y=relativeValue, fill=variable), stat='identity', position="dodge") plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1)) plt = plt + scale_y_continuous(trans="log", breaks=10^c(0:10), labels=c(0, 10^c(0:9))) plt = plt + geom_text(data=patientResult[patientResult$variable == oneSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=-0.2) plt = plt + geom_text(data=patientResult[patientResult$variable == twoSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=0.8) plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle(paste("Number of clones in only ", oneSample, " and only ", twoSample, sep="")) png(paste(patient, "_", onShort, "_both.png", sep=""), width=1920, height=1080) print(plt) dev.off() } if(length(patients) > 0){ cat("<tr><td>Starting Frequency analysis</td></tr>", file=logfile, append=T) interval = intervalFreq intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval)) product = data.frame("Titles"=rep(Titles, each=length(interval)), "interval"=rep(interval, times=10), "V_Segments"=rep(V_Segments, each=length(interval)), "J_Segments"=rep(J_Segments, each=length(interval))) for (current_patient in patients){ print(paste("Started working", unique(current_patient$Patient), "Frequency analysis")) patientCountOnColumn(current_patient, product=product, interval=interval, on="Frequency", appendtxt=T) } cat("<tr><td>Starting Cell Count analysis</td></tr>", file=logfile, append=T) interval = intervalReads intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval)) product = data.frame("Titles"=rep(Titles, each=length(interval)), "interval"=rep(interval, times=10), "V_Segments"=rep(V_Segments, each=length(interval)), "J_Segments"=rep(J_Segments, each=length(interval))) for (current_patient in patients){ print(paste("Started working on ", unique(current_patient$Patient), "Read Count analysis")) patientCountOnColumn(current_patient, product=product, interval=interval, on="normalized_read_count") } } if(nrow(single_patients) > 0){ scales = 10^(0:6) #(0:ceiling(log10(max(scatterplot_locus_data$normalized_read_count)))) p = ggplot(single_patients, aes(Rearrangement, normalized_read_count)) + scale_y_log10(breaks=scales,labels=as.character(scales)) + expand_limits(y=c(0,1000000)) p = p + geom_point(aes(colour=type), position="jitter") p = p + xlab("In one or both samples") + ylab("Reads") p = p + facet_grid(.~Patient) + ggtitle("Scatterplot of the reads of the patients with a single sample") png("singles_reads_scatterplot.png", width=640 * length(unique(single_patients$Patient)) + 100, height=1080) print(p) dev.off() #p = ggplot(single_patients, aes(Rearrangement, Frequency)) + scale_y_continuous(limits = c(0, 100)) + expand_limits(y=c(0,100)) p = ggplot(single_patients, aes(Rearrangement, Frequency)) + scale_y_log10(limits=c(0.0001,100), breaks=c(0.0001, 0.001, 0.01, 0.1, 1, 10, 100), labels=c("0.0001", "0.001", "0.01", "0.1", "1", "10", "100")) + expand_limits(y=c(0,100)) p = p + geom_point(aes(colour=type), position="jitter") p = p + xlab("In one or both samples") + ylab("Frequency") p = p + facet_grid(.~Patient) + ggtitle("Scatterplot of the frequency of the patients with a single sample") png("singles_freq_scatterplot.png", width=640 * length(unique(single_patients$Patient)) + 100, height=1080) print(p) dev.off() } else { empty <- data.frame() p = ggplot(empty) + geom_point() + xlim(0, 10) + ylim(0, 100) + xlab("In one or both samples") + ylab("Frequency") + ggtitle("Scatterplot of the frequency of the patients with a single sample") png("singles_reads_scatterplot.png", width=400, height=300) print(p) dev.off() png("singles_freq_scatterplot.png", width=400, height=300) print(p) dev.off() } patient.merge.list = list() #cache the 'both' table, 2x speedup for more memory... patient.merge.list.second = list() tripletAnalysis <- function(patient1, label1, patient2, label2, patient3, label3, product, interval, on, appendTriplets= FALSE){ onShort = "reads" if(on == "Frequency"){ onShort = "freq" } onx = paste(on, ".x", sep="") ony = paste(on, ".y", sep="") onz = paste(on, ".z", sep="") type="triplet" threshholdIndex = which(colnames(product) == "interval") V_SegmentIndex = which(colnames(product) == "V_Segments") J_SegmentIndex = which(colnames(product) == "J_Segments") titleIndex = which(colnames(product) == "Titles") sampleIndex = which(colnames(patient1) == "Sample") patientIndex = which(colnames(patient1) == "Patient") oneSample = paste(patient1[1,sampleIndex], sep="") twoSample = paste(patient2[1,sampleIndex], sep="") threeSample = paste(patient3[1,sampleIndex], sep="") if(mergeOn == "Clone_Sequence"){ patient1$merge = paste(patient1$Clone_Sequence) patient2$merge = paste(patient2$Clone_Sequence) patient3$merge = paste(patient3$Clone_Sequence) } else { patient1$merge = paste(patient1$V_Segment_Major_Gene, patient1$J_Segment_Major_Gene, patient1$CDR3_Sense_Sequence) patient2$merge = paste(patient2$V_Segment_Major_Gene, patient2$J_Segment_Major_Gene, patient2$CDR3_Sense_Sequence) patient3$merge = paste(patient3$V_Segment_Major_Gene, patient3$J_Segment_Major_Gene, patient3$CDR3_Sense_Sequence) } #patientMerge = merge(patient1, patient2, by="merge")[NULL,] patient1.fuzzy = patient1 patient2.fuzzy = patient2 patient3.fuzzy = patient3 cat(paste("<tr><td>", label1, "</td>", sep=""), file=logfile, append=T) patient1.fuzzy$merge = paste(patient1.fuzzy$locus_V, patient1.fuzzy$locus_J) patient2.fuzzy$merge = paste(patient2.fuzzy$locus_V, patient2.fuzzy$locus_J) patient3.fuzzy$merge = paste(patient3.fuzzy$locus_V, patient3.fuzzy$locus_J) patient.fuzzy = rbind(patient1.fuzzy ,patient2.fuzzy, patient3.fuzzy) patient.fuzzy = patient.fuzzy[order(nchar(patient.fuzzy$Clone_Sequence)),] other.sample.list = list() other.sample.list[[oneSample]] = c(twoSample, threeSample) other.sample.list[[twoSample]] = c(oneSample, threeSample) other.sample.list[[threeSample]] = c(oneSample, twoSample) patientMerge = merge(patient1, patient2, by="merge") patientMerge = merge(patientMerge, patient3, by="merge") colnames(patientMerge)[which(!grepl("(\\.x$)|(\\.y$)|(merge)", names(patientMerge)))] = paste(colnames(patientMerge)[which(!grepl("(\\.x$)|(\\.y$)|(merge)", names(patientMerge), perl=T))], ".z", sep="") #patientMerge$thresholdValue = pmax(patientMerge[,onx], patientMerge[,ony], patientMerge[,onz]) patientMerge = patientMerge[NULL,] duo.merge.list = list() patientMerge12 = merge(patient1, patient2, by="merge") #patientMerge12$thresholdValue = pmax(patientMerge12[,onx], patientMerge12[,ony]) patientMerge12 = patientMerge12[NULL,] duo.merge.list[[paste(oneSample, twoSample)]] = patientMerge12 duo.merge.list[[paste(twoSample, oneSample)]] = patientMerge12 patientMerge13 = merge(patient1, patient3, by="merge") #patientMerge13$thresholdValue = pmax(patientMerge13[,onx], patientMerge13[,ony]) patientMerge13 = patientMerge13[NULL,] duo.merge.list[[paste(oneSample, threeSample)]] = patientMerge13 duo.merge.list[[paste(threeSample, oneSample)]] = patientMerge13 patientMerge23 = merge(patient2, patient3, by="merge") #patientMerge23$thresholdValue = pmax(patientMerge23[,onx], patientMerge23[,ony]) patientMerge23 = patientMerge23[NULL,] duo.merge.list[[paste(twoSample, threeSample)]] = patientMerge23 duo.merge.list[[paste(threeSample, twoSample)]] = patientMerge23 merge.list = list() merge.list[["second"]] = vector() #print(paste(nrow(patient1), nrow(patient2), nrow(patient3), label1, label2, label3)) start.time = proc.time() if(paste(label1, "123") %in% names(patient.merge.list)){ patientMerge = patient.merge.list[[paste(label1, "123")]] patientMerge12 = patient.merge.list[[paste(label1, "12")]] patientMerge13 = patient.merge.list[[paste(label1, "13")]] patientMerge23 = patient.merge.list[[paste(label1, "23")]] #merge.list[["second"]] = patient.merge.list.second[[label1]] cat(paste("<td>", nrow(patient1), " in ", label1, " and ", nrow(patient2), " in ", label2, nrow(patient3), " in ", label3, ", ", nrow(patientMerge), " in both (fetched from cache)</td></tr>", sep=""), file=logfile, append=T) } else { while(nrow(patient.fuzzy) > 0){ first.merge = patient.fuzzy[1,"merge"] first.clone.sequence = patient.fuzzy[1,"Clone_Sequence"] first.sample = paste(patient.fuzzy[1,"Sample"], sep="") merge.filter = first.merge == patient.fuzzy$merge second.sample = other.sample.list[[first.sample]][1] third.sample = other.sample.list[[first.sample]][2] sample.filter.1 = first.sample == patient.fuzzy$Sample sample.filter.2 = second.sample == patient.fuzzy$Sample sample.filter.3 = third.sample == patient.fuzzy$Sample sequence.filter = grepl(paste("^", first.clone.sequence, sep=""), patient.fuzzy$Clone_Sequence) match.filter.1 = sample.filter.1 & sequence.filter & merge.filter match.filter.2 = sample.filter.2 & sequence.filter & merge.filter match.filter.3 = sample.filter.3 & sequence.filter & merge.filter matches.in.1 = any(match.filter.1) matches.in.2 = any(match.filter.2) matches.in.3 = any(match.filter.3) rows.1 = patient.fuzzy[match.filter.1,] sum.1 = data.frame(merge = first.clone.sequence, Patient = label1, Receptor = rows.1[1,"Receptor"], Sample = rows.1[1,"Sample"], Cell_Count = rows.1[1,"Cell_Count"], Clone_Molecule_Count_From_Spikes = sum(rows.1$Clone_Molecule_Count_From_Spikes), Log10_Frequency = log10(sum(rows.1$Frequency)), Total_Read_Count = sum(rows.1$Total_Read_Count), dsPerM = sum(rows.1$dsPerM), J_Segment_Major_Gene = rows.1[1,"J_Segment_Major_Gene"], V_Segment_Major_Gene = rows.1[1,"V_Segment_Major_Gene"], Clone_Sequence = first.clone.sequence, CDR3_Sense_Sequence = rows.1[1,"CDR3_Sense_Sequence"], Related_to_leukemia_clone = F, Frequency = sum(rows.1$Frequency), locus_V = rows.1[1,"locus_V"], locus_J = rows.1[1,"locus_J"], uniqueID = rows.1[1,"uniqueID"], normalized_read_count = sum(rows.1$normalized_read_count)) sum.2 = sum.1[NULL,] rows.2 = patient.fuzzy[match.filter.2,] if(matches.in.2){ sum.2 = data.frame(merge = first.clone.sequence, Patient = label1, Receptor = rows.2[1,"Receptor"], Sample = rows.2[1,"Sample"], Cell_Count = rows.2[1,"Cell_Count"], Clone_Molecule_Count_From_Spikes = sum(rows.2$Clone_Molecule_Count_From_Spikes), Log10_Frequency = log10(sum(rows.2$Frequency)), Total_Read_Count = sum(rows.2$Total_Read_Count), dsPerM = sum(rows.2$dsPerM), J_Segment_Major_Gene = rows.2[1,"J_Segment_Major_Gene"], V_Segment_Major_Gene = rows.2[1,"V_Segment_Major_Gene"], Clone_Sequence = first.clone.sequence, CDR3_Sense_Sequence = rows.2[1,"CDR3_Sense_Sequence"], Related_to_leukemia_clone = F, Frequency = sum(rows.2$Frequency), locus_V = rows.2[1,"locus_V"], locus_J = rows.2[1,"locus_J"], uniqueID = rows.2[1,"uniqueID"], normalized_read_count = sum(rows.2$normalized_read_count)) } sum.3 = sum.1[NULL,] rows.3 = patient.fuzzy[match.filter.3,] if(matches.in.3){ sum.3 = data.frame(merge = first.clone.sequence, Patient = label1, Receptor = rows.3[1,"Receptor"], Sample = rows.3[1,"Sample"], Cell_Count = rows.3[1,"Cell_Count"], Clone_Molecule_Count_From_Spikes = sum(rows.3$Clone_Molecule_Count_From_Spikes), Log10_Frequency = log10(sum(rows.3$Frequency)), Total_Read_Count = sum(rows.3$Total_Read_Count), dsPerM = sum(rows.3$dsPerM), J_Segment_Major_Gene = rows.3[1,"J_Segment_Major_Gene"], V_Segment_Major_Gene = rows.3[1,"V_Segment_Major_Gene"], Clone_Sequence = first.clone.sequence, CDR3_Sense_Sequence = rows.3[1,"CDR3_Sense_Sequence"], Related_to_leukemia_clone = F, Frequency = sum(rows.3$Frequency), locus_V = rows.3[1,"locus_V"], locus_J = rows.3[1,"locus_J"], uniqueID = rows.3[1,"uniqueID"], normalized_read_count = sum(rows.3$normalized_read_count)) } if(matches.in.2 & matches.in.3){ merge.123 = merge(sum.1, sum.2, by="merge") merge.123 = merge(merge.123, sum.3, by="merge") colnames(merge.123)[which(!grepl("(\\.x$)|(\\.y$)|(merge)", names(merge.123)))] = paste(colnames(merge.123)[which(!grepl("(\\.x$)|(\\.y$)|(merge)", names(merge.123), perl=T))], ".z", sep="") #merge.123$thresholdValue = pmax(merge.123[,onx], merge.123[,ony], merge.123[,onz]) patientMerge = rbind(patientMerge, merge.123) patient.fuzzy = patient.fuzzy[!(match.filter.1 | match.filter.2 | match.filter.3),] hidden.clone.sequences = c(rows.1[-1,"Clone_Sequence"], rows.2[rows.2$Clone_Sequence != first.clone.sequence,"Clone_Sequence"], rows.3[rows.3$Clone_Sequence != first.clone.sequence,"Clone_Sequence"]) merge.list[["second"]] = append(merge.list[["second"]], hidden.clone.sequences) } else if (matches.in.2) { #other.sample1 = other.sample.list[[first.sample]][1] #other.sample2 = other.sample.list[[first.sample]][2] second.sample = sum.2[,"Sample"] current.merge.list = duo.merge.list[[paste(first.sample, second.sample)]] merge.12 = merge(sum.1, sum.2, by="merge") current.merge.list = rbind(current.merge.list, merge.12) duo.merge.list[[paste(first.sample, second.sample)]] = current.merge.list patient.fuzzy = patient.fuzzy[!(match.filter.1 | match.filter.2),] hidden.clone.sequences = c(rows.1[-1,"Clone_Sequence"], rows.2[rows.2$Clone_Sequence != first.clone.sequence,"Clone_Sequence"]) merge.list[["second"]] = append(merge.list[["second"]], hidden.clone.sequences) } else if (matches.in.3) { #other.sample1 = other.sample.list[[first.sample]][1] #other.sample2 = other.sample.list[[first.sample]][2] second.sample = sum.3[,"Sample"] current.merge.list = duo.merge.list[[paste(first.sample, second.sample)]] merge.13 = merge(sum.1, sum.3, by="merge") current.merge.list = rbind(current.merge.list, merge.13) duo.merge.list[[paste(first.sample, second.sample)]] = current.merge.list patient.fuzzy = patient.fuzzy[!(match.filter.1 | match.filter.3),] hidden.clone.sequences = c(rows.1[-1,"Clone_Sequence"], rows.3[rows.3$Clone_Sequence != first.clone.sequence,"Clone_Sequence"]) merge.list[["second"]] = append(merge.list[["second"]], hidden.clone.sequences) } else if(nrow(rows.1) > 1){ patient1 = patient1[!(patient1$Clone_Sequence %in% rows.1$Clone_Sequence),] #print(names(patient1)[names(patient1) %in% sum.1]) #print(names(patient1)[!(names(patient1) %in% sum.1)]) #print(names(patient1)) #print(names(sum.1)) #print(summary(sum.1)) #print(summary(patient1)) #print(dim(sum.1)) #print(dim(patient1)) #print(head(sum.1[,names(patient1)])) patient1 = rbind(patient1, sum.1[,names(patient1)]) patient.fuzzy = patient.fuzzy[-match.filter.1,] } else { patient.fuzzy = patient.fuzzy[-1,] } tmp.rows = rbind(rows.1, rows.2, rows.3) tmp.rows = tmp.rows[order(nchar(tmp.rows$Clone_Sequence)),] if (sum(match.filter.1) > 1 | sum(match.filter.2) > 1 | sum(match.filter.1) > 1) { cat(paste("<tr><td>", label1, " row ", 1:nrow(tmp.rows), "</td><td>", tmp.rows$Sample, ":</td><td>", tmp.rows$Clone_Sequence, "</td><td>", tmp.rows$normalized_read_count, "</td></tr>", sep=""), file="multiple_matches.html", append=T) } else { } } patient.merge.list[[paste(label1, "123")]] = patientMerge patientMerge12 = duo.merge.list[[paste(oneSample, twoSample)]] patientMerge13 = duo.merge.list[[paste(oneSample, threeSample)]] patientMerge23 = duo.merge.list[[paste(twoSample, threeSample)]] patient.merge.list[[paste(label1, "12")]] = patientMerge12 patient.merge.list[[paste(label1, "13")]] = patientMerge13 patient.merge.list[[paste(label1, "23")]] = patientMerge23 #patient.merge.list.second[[label1]] = merge.list[["second"]] } cat(paste("<td>", nrow(patient1), " in ", label1, " and ", nrow(patient2), " in ", label2, nrow(patient3), " in ", label3, ", ", nrow(patientMerge), " in both (finding both took ", (proc.time() - start.time)[[3]], "s)</td></tr>", sep=""), file=logfile, append=T) patientMerge$thresholdValue = pmax(patientMerge[,onx], patientMerge[,ony], patientMerge[,onz]) patientMerge12$thresholdValue = pmax(patientMerge12[,onx], patientMerge12[,ony]) patientMerge13$thresholdValue = pmax(patientMerge13[,onx], patientMerge13[,ony]) patientMerge23$thresholdValue = pmax(patientMerge23[,onx], patientMerge23[,ony]) #patientMerge$thresholdValue = pmin(patientMerge[,onx], patientMerge[,ony], patientMerge[,onz]) #patientMerge12$thresholdValue = pmin(patientMerge12[,onx], patientMerge12[,ony]) #patientMerge13$thresholdValue = pmin(patientMerge13[,onx], patientMerge13[,ony]) #patientMerge23$thresholdValue = pmin(patientMerge23[,onx], patientMerge23[,ony]) patient1 = patient1[!(patient1$Clone_Sequence %in% merge.list[["second"]]),] patient2 = patient2[!(patient2$Clone_Sequence %in% merge.list[["second"]]),] patient3 = patient3[!(patient3$Clone_Sequence %in% merge.list[["second"]]),] if(F){ patientMerge = merge(patient1, patient2, by="merge") patientMerge = merge(patientMerge, patient3, by="merge") colnames(patientMerge)[which(!grepl("(\\.x$)|(\\.y$)|(merge)", names(patientMerge)))] = paste(colnames(patientMerge)[which(!grepl("(\\.x$)|(\\.y$)|(merge)", names(patientMerge), perl=T))], ".z", sep="") patientMerge$thresholdValue = pmax(patientMerge[,onx], patientMerge[,ony], patientMerge[,onz]) patientMerge12 = merge(patient1, patient2, by="merge") patientMerge12$thresholdValue = pmax(patientMerge12[,onx], patientMerge12[,ony]) patientMerge13 = merge(patient1, patient3, by="merge") patientMerge13$thresholdValue = pmax(patientMerge13[,onx], patientMerge13[,ony]) patientMerge23 = merge(patient2, patient3, by="merge") patientMerge23$thresholdValue = pmax(patientMerge23[,onx], patientMerge23[,ony]) } scatterplot_data_columns = c("Clone_Sequence", "Frequency", "normalized_read_count", "V_Segment_Major_Gene", "J_Segment_Major_Gene", "merge") scatterplot_data = rbind(patient1[,scatterplot_data_columns], patient2[,scatterplot_data_columns], patient3[,scatterplot_data_columns]) scatterplot_data = scatterplot_data[!duplicated(scatterplot_data$merge),] scatterplot_data$type = factor(x="In one", levels=c("In one", "In two", "In three", "In multiple")) res1 = vector() res2 = vector() res3 = vector() res12 = vector() res13 = vector() res23 = vector() resAll = vector() read1Count = vector() read2Count = vector() read3Count = vector() if(appendTriplets){ cat(paste(label1, label2, label3, sep="\t"), file="triplets.txt", append=T, sep="", fill=3) } for(iter in 1:length(product[,1])){ threshhold = product[iter,threshholdIndex] V_Segment = paste(".*", as.character(product[iter,V_SegmentIndex]), ".*", sep="") J_Segment = paste(".*", as.character(product[iter,J_SegmentIndex]), ".*", sep="") #all = (grepl(V_Segment, patientMerge$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge$J_Segment_Major_Gene.x) & patientMerge[,onx] > threshhold & patientMerge[,ony] > threshhold & patientMerge[,onz] > threshhold) all = (grepl(V_Segment, patientMerge$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge$J_Segment_Major_Gene.x) & patientMerge$thresholdValue > threshhold) one_two = (grepl(V_Segment, patientMerge12$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge12$J_Segment_Major_Gene.x) & patientMerge12$thresholdValue > threshhold & !(patientMerge12$merge %in% patientMerge[all,]$merge)) one_three = (grepl(V_Segment, patientMerge13$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge13$J_Segment_Major_Gene.x) & patientMerge13$thresholdValue > threshhold & !(patientMerge13$merge %in% patientMerge[all,]$merge)) two_three = (grepl(V_Segment, patientMerge23$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge23$J_Segment_Major_Gene.x) & patientMerge23$thresholdValue > threshhold & !(patientMerge23$merge %in% patientMerge[all,]$merge)) one = (grepl(V_Segment, patient1$V_Segment_Major_Gene) & grepl(J_Segment, patient1$J_Segment_Major_Gene) & patient1[,on] > threshhold & !(patient1$merge %in% patientMerge[all,]$merge) & !(patient1$merge %in% patientMerge12[one_two,]$merge) & !(patient1$merge %in% patientMerge13[one_three,]$merge)) two = (grepl(V_Segment, patient2$V_Segment_Major_Gene) & grepl(J_Segment, patient2$J_Segment_Major_Gene) & patient2[,on] > threshhold & !(patient2$merge %in% patientMerge[all,]$merge) & !(patient2$merge %in% patientMerge12[one_two,]$merge) & !(patient2$merge %in% patientMerge23[two_three,]$merge)) three = (grepl(V_Segment, patient3$V_Segment_Major_Gene) & grepl(J_Segment, patient3$J_Segment_Major_Gene) & patient3[,on] > threshhold & !(patient3$merge %in% patientMerge[all,]$merge) & !(patient3$merge %in% patientMerge13[one_three,]$merge) & !(patient3$merge %in% patientMerge23[two_three,]$merge)) read1Count = append(read1Count, sum(patient1[one,]$normalized_read_count) + sum(patientMerge[all,]$normalized_read_count.x)) read2Count = append(read2Count, sum(patient2[two,]$normalized_read_count) + sum(patientMerge[all,]$normalized_read_count.y)) read3Count = append(read3Count, sum(patient3[three,]$normalized_read_count) + sum(patientMerge[all,]$normalized_read_count.z)) res1 = append(res1, sum(one)) res2 = append(res2, sum(two)) res3 = append(res3, sum(three)) resAll = append(resAll, sum(all)) res12 = append(res12, sum(one_two)) res13 = append(res13, sum(one_three)) res23 = append(res23, sum(two_three)) #threshhold = 0 if(threshhold != 0){ if(sum(one) > 0){ dfOne = patient1[one,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone")] colnames(dfOne) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone") filenameOne = paste(label1, "_", product[iter, titleIndex], "_", threshhold, sep="") write.table(dfOne, file=paste(filenameOne, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } if(sum(two) > 0){ dfTwo = patient2[two,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone")] colnames(dfTwo) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone") filenameTwo = paste(label2, "_", product[iter, titleIndex], "_", threshhold, sep="") write.table(dfTwo, file=paste(filenameTwo, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } if(sum(three) > 0){ dfThree = patient3[three,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone")] colnames(dfThree) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Clone_Sequence", "Related_to_leukemia_clone") filenameThree = paste(label3, "_", product[iter, titleIndex], "_", threshhold, sep="") write.table(dfThree, file=paste(filenameThree, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } if(sum(one_two) > 0){ dfOne_two = patientMerge12[one_two,c("V_Segment_Major_Gene.x", "J_Segment_Major_Gene.x", "normalized_read_count.x", "Frequency.x", "Related_to_leukemia_clone.x", "Clone_Sequence.x", "V_Segment_Major_Gene.y", "J_Segment_Major_Gene.y", "normalized_read_count.y", "Frequency.y", "Related_to_leukemia_clone.y")] colnames(dfOne_two) = c(paste("Proximal segment", oneSample), paste("Distal segment", oneSample), paste("Normalized_Read_Count", oneSample), paste("Frequency", oneSample), paste("Related_to_leukemia_clone", oneSample),"Clone_Sequence", paste("Proximal segment", twoSample), paste("Distal segment", twoSample), paste("Normalized_Read_Count", twoSample), paste("Frequency", twoSample), paste("Related_to_leukemia_clone", twoSample)) filenameOne_two = paste(label1, "_", label2, "_", product[iter, titleIndex], "_", threshhold, onShort, sep="") write.table(dfOne_two, file=paste(filenameOne_two, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } if(sum(one_three) > 0){ dfOne_three = patientMerge13[one_three,c("V_Segment_Major_Gene.x", "J_Segment_Major_Gene.x", "normalized_read_count.x", "Frequency.x", "Related_to_leukemia_clone.x", "Clone_Sequence.x", "V_Segment_Major_Gene.y", "J_Segment_Major_Gene.y", "normalized_read_count.y", "Frequency.y", "Related_to_leukemia_clone.y")] colnames(dfOne_three) = c(paste("Proximal segment", oneSample), paste("Distal segment", oneSample), paste("Normalized_Read_Count", oneSample), paste("Frequency", oneSample), paste("Related_to_leukemia_clone", oneSample),"Clone_Sequence", paste("Proximal segment", threeSample), paste("Distal segment", threeSample), paste("Normalized_Read_Count", threeSample), paste("Frequency", threeSample), paste("Related_to_leukemia_clone", threeSample)) filenameOne_three = paste(label1, "_", label3, "_", product[iter, titleIndex], "_", threshhold, onShort, sep="") write.table(dfOne_three, file=paste(filenameOne_three, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } if(sum(two_three) > 0){ dfTwo_three = patientMerge23[two_three,c("V_Segment_Major_Gene.x", "J_Segment_Major_Gene.x", "normalized_read_count.x", "Frequency.x", "Related_to_leukemia_clone.x", "Clone_Sequence.x", "V_Segment_Major_Gene.y", "J_Segment_Major_Gene.y", "normalized_read_count.y", "Frequency.y", "Related_to_leukemia_clone.y")] colnames(dfTwo_three) = c(paste("Proximal segment", twoSample), paste("Distal segment", twoSample), paste("Normalized_Read_Count", twoSample), paste("Frequency", twoSample), paste("Related_to_leukemia_clone", twoSample),"Clone_Sequence", paste("Proximal segment", threeSample), paste("Distal segment", threeSample), paste("Normalized_Read_Count", threeSample), paste("Frequency", threeSample), paste("Related_to_leukemia_clone", threeSample)) filenameTwo_three = paste(label2, "_", label3, "_", product[iter, titleIndex], "_", threshhold, onShort, sep="") write.table(dfTwo_three, file=paste(filenameTwo_three, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } } else { #scatterplot data scatterplot_locus_data = scatterplot_data[grepl(V_Segment, scatterplot_data$V_Segment_Major_Gene) & grepl(J_Segment, scatterplot_data$J_Segment_Major_Gene),] scatterplot_locus_data = scatterplot_locus_data[!(scatterplot_locus_data$merge %in% merge.list[["second"]]),] in_two = (scatterplot_locus_data$merge %in% patientMerge12[one_two,]$merge) | (scatterplot_locus_data$merge %in% patientMerge13[one_three,]$merge) | (scatterplot_locus_data$merge %in% patientMerge23[two_three,]$merge) if(sum(in_two) > 0){ scatterplot_locus_data[in_two,]$type = "In two" } in_three = (scatterplot_locus_data$merge %in% patientMerge[all,]$merge) if(sum(in_three)> 0){ scatterplot_locus_data[in_three,]$type = "In three" } not_in_one = scatterplot_locus_data$type != "In one" if(sum(not_in_one) > 0){ #scatterplot_locus_data[not_in_one,]$type = "In multiple" } p = NULL if(nrow(scatterplot_locus_data) != 0){ filename.scatter = paste(label1, "_", label2, "_", label3, "_", product[iter, titleIndex], "_scatter_", threshhold, sep="") write.table(scatterplot_locus_data, file=paste(filename.scatter, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) if(on == "normalized_read_count"){ scales = 10^(0:6) #(0:ceiling(log10(max(scatterplot_locus_data$normalized_read_count)))) p = ggplot(scatterplot_locus_data, aes(type, normalized_read_count)) + scale_y_log10(breaks=scales,labels=scales, limits=c(1, 1e6)) } else { p = ggplot(scatterplot_locus_data, aes(type, Frequency)) + scale_y_log10(limits=c(0.0001,100), breaks=c(0.0001, 0.001, 0.01, 0.1, 1, 10, 100), labels=c("0.0001", "0.001", "0.01", "0.1", "1", "10", "100")) + expand_limits(y=c(0,100)) #p = ggplot(scatterplot_locus_data, aes(type, Frequency)) + scale_y_continuous(limits = c(0, 100)) + expand_limits(y=c(0,100)) } p = p + geom_point(aes(colour=type), position="jitter") p = p + xlab("In one or in multiple samples") + ylab(onShort) + ggtitle(paste(label1, label2, label3, onShort, product[iter, titleIndex])) } else { p = ggplot(NULL, aes(x=c("In one", "In multiple"),y=0)) + geom_blank(NULL) + xlab("In two or in three of the samples") + ylab(onShort) + ggtitle(paste(label1, label2, label3, onShort, product[iter, titleIndex])) } png(paste(label1, "_", label2, "_", label3, "_", onShort, "_", product[iter, titleIndex],"_scatter.png", sep="")) print(p) dev.off() } if(sum(all) > 0){ dfAll = patientMerge[all,c("V_Segment_Major_Gene.x", "J_Segment_Major_Gene.x", "normalized_read_count.x", "Frequency.x", "Related_to_leukemia_clone.x", "Clone_Sequence.x", "V_Segment_Major_Gene.y", "J_Segment_Major_Gene.y", "normalized_read_count.y", "Frequency.y", "Related_to_leukemia_clone.y", "V_Segment_Major_Gene.z", "J_Segment_Major_Gene.z", "normalized_read_count.z", "Frequency.z", "Related_to_leukemia_clone.z")] colnames(dfAll) = c(paste("Proximal segment", oneSample), paste("Distal segment", oneSample), paste("Normalized_Read_Count", oneSample), paste("Frequency", oneSample), paste("Related_to_leukemia_clone", oneSample),"Clone_Sequence", paste("Proximal segment", twoSample), paste("Distal segment", twoSample), paste("Normalized_Read_Count", twoSample), paste("Frequency", twoSample), paste("Related_to_leukemia_clone", twoSample), paste("Proximal segment", threeSample), paste("Distal segment", threeSample), paste("Normalized_Read_Count", threeSample), paste("Frequency", threeSample), paste("Related_to_leukemia_clone", threeSample)) filenameAll = paste(label1, "_", label2, "_", label3, "_", product[iter, titleIndex], "_", threshhold, sep="") write.table(dfAll, file=paste(filenameAll, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) } } #patientResult = data.frame("Locus"=product$Titles, "J_Segment"=product$J_Segments, "V_Segment"=product$V_Segments, "cut_off_value"=paste(">", product$interval, sep=""), "All"=resAll, "tmp1"=res1, "read_count1" = round(read1Count), "tmp2"=res2, "read_count2"= round(read2Count), "tmp3"=res3, "read_count3"=round(read3Count)) patientResult = data.frame("Locus"=product$Titles, "J_Segment"=product$J_Segments, "V_Segment"=product$V_Segments, "cut_off_value"=paste(">", product$interval, sep=""), "All"=resAll, "tmp1"=res1, "tmp2"=res2, "tmp3"=res3, "tmp12"=res12, "tmp13"=res13, "tmp23"=res23) colnames(patientResult)[6] = oneSample colnames(patientResult)[7] = twoSample colnames(patientResult)[8] = threeSample colnames(patientResult)[9] = paste(oneSample, twoSample, sep="_") colnames(patientResult)[10] = paste(oneSample, twoSample, sep="_") colnames(patientResult)[11] = paste(oneSample, twoSample, sep="_") colnamesBak = colnames(patientResult) colnames(patientResult) = c("Ig/TCR gene rearrangement type", "Distal Gene segment", "Proximal gene segment", "cut_off_value", "Number of sequences All", paste("Number of sequences", oneSample), paste("Number of sequences", twoSample), paste("Number of sequences", threeSample), paste("Number of sequences", oneSample, twoSample), paste("Number of sequences", oneSample, threeSample), paste("Number of sequences", twoSample, threeSample)) write.table(patientResult, file=paste(label1, "_", label2, "_", label3, "_", onShort, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T) colnames(patientResult) = colnamesBak patientResult$Locus = factor(patientResult$Locus, Titles) patientResult$cut_off_value = factor(patientResult$cut_off_value, paste(">", interval, sep="")) plt = ggplot(patientResult[,c("Locus", "cut_off_value", "All")]) plt = plt + geom_bar( aes( x=factor(cut_off_value), y=All), stat='identity', position="dodge", fill="#79c36a") plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1)) plt = plt + geom_text(aes(ymax=max(All), x=cut_off_value,y=All,label=All), angle=90, hjust=0) plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("Number of clones in All") plt = plt + theme(plot.margin = unit(c(1,8.8,0.5,1.5), "lines")) png(paste(label1, "_", label2, "_", label3, "_", onShort, "_total_all.png", sep=""), width=1920, height=1080) print(plt) dev.off() fontSize = 4 bak = patientResult patientResult = melt(patientResult[,c('Locus','cut_off_value', oneSample, twoSample, threeSample)] ,id.vars=1:2) patientResult$relativeValue = patientResult$value * 10 patientResult[patientResult$relativeValue == 0,]$relativeValue = 1 plt = ggplot(patientResult) plt = plt + geom_bar( aes( x=factor(cut_off_value), y=relativeValue, fill=variable), stat='identity', position="dodge") plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1)) plt = plt + scale_y_continuous(trans="log", breaks=10^c(0:10), labels=c(0, 10^c(0:9))) plt = plt + geom_text(data=patientResult[patientResult$variable == oneSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=-0.7, size=fontSize) plt = plt + geom_text(data=patientResult[patientResult$variable == twoSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=0.4, size=fontSize) plt = plt + geom_text(data=patientResult[patientResult$variable == threeSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=1.5, size=fontSize) plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("Number of clones in only one sample") png(paste(label1, "_", label2, "_", label3, "_", onShort, "_indiv_all.png", sep=""), width=1920, height=1080) print(plt) dev.off() } if(nrow(triplets) != 0){ cat("<tr><td>Starting triplet analysis</td></tr>", file=logfile, append=T) triplets$uniqueID = paste(triplets$Patient, triplets$Sample, sep="_") cat("<tr><td>Normalizing to lowest cell count within locus</td></tr>", file=logfile, append=T) triplets$locus_V = substring(triplets$V_Segment_Major_Gene, 0, 4) triplets$locus_J = substring(triplets$J_Segment_Major_Gene, 0, 4) min_cell_count = data.frame(data.table(triplets)[, list(min_cell_count=min(.SD$Cell_Count)), by=c("uniqueID", "locus_V", "locus_J")]) triplets$min_cell_paste = paste(triplets$uniqueID, triplets$locus_V, triplets$locus_J) min_cell_count$min_cell_paste = paste(min_cell_count$uniqueID, min_cell_count$locus_V, min_cell_count$locus_J) min_cell_count = min_cell_count[,c("min_cell_paste", "min_cell_count")] triplets = merge(triplets, min_cell_count, by="min_cell_paste") triplets$normalized_read_count = round(triplets$Clone_Molecule_Count_From_Spikes / triplets$Cell_Count * triplets$min_cell_count / 2, digits=2) triplets = triplets[triplets$normalized_read_count >= min_cells,] column_drops = c("min_cell_count", "min_cell_paste") triplets = triplets[,!(colnames(triplets) %in% column_drops)] cat("<tr><td>Starting Cell Count analysis</td></tr>", file=logfile, append=T) interval = intervalReads intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval)) product = data.frame("Titles"=rep(Titles, each=length(interval)), "interval"=rep(interval, times=10), "V_Segments"=rep(V_Segments, each=length(interval)), "J_Segments"=rep(J_Segments, each=length(interval))) triplets = split(triplets, triplets$Patient, drop=T) #print(nrow(triplets)) for(triplet in triplets){ samples = unique(triplet$Sample) one = triplet[triplet$Sample == samples[1],] two = triplet[triplet$Sample == samples[2],] three = triplet[triplet$Sample == samples[3],] print(paste(nrow(triplet), nrow(one), nrow(two), nrow(three))) tripletAnalysis(one, one[1,"uniqueID"], two, two[1,"uniqueID"], three, three[1,"uniqueID"], product=product, interval=interval, on="normalized_read_count", T) } cat("<tr><td>Starting Frequency analysis</td></tr>", file=logfile, append=T) interval = intervalFreq intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval)) product = data.frame("Titles"=rep(Titles, each=length(interval)), "interval"=rep(interval, times=10), "V_Segments"=rep(V_Segments, each=length(interval)), "J_Segments"=rep(J_Segments, each=length(interval))) for(triplet in triplets){ samples = unique(triplet$Sample) one = triplet[triplet$Sample == samples[1],] two = triplet[triplet$Sample == samples[2],] three = triplet[triplet$Sample == samples[3],] tripletAnalysis(one, one[1,"uniqueID"], two, two[1,"uniqueID"], three, three[1,"uniqueID"], product=product, interval=interval, on="Frequency", F) } } else { cat("", file="triplets.txt") } cat("</table></html>", file=logfile, append=T)