|
0
|
1 args <- commandArgs(trailingOnly = TRUE)
|
|
|
2
|
|
|
3 inFile = args[1]
|
|
|
4 outDir = args[2]
|
|
3
|
5 logfile = args[3]
|
|
|
6 min_freq = as.numeric(args[4])
|
|
|
7 min_cells = as.numeric(args[5])
|
|
|
8
|
|
|
9 cat("<html><table><tr><td>Starting analysis</td></tr>", file=logfile, append=F)
|
|
0
|
10
|
|
4
|
11 library(ggplot2)
|
|
|
12 library(reshape2)
|
|
|
13 library(data.table)
|
|
|
14 library(grid)
|
|
|
15 library(parallel)
|
|
0
|
16 #require(xtable)
|
|
3
|
17 cat("<tr><td>Reading input</td></tr>", file=logfile, append=T)
|
|
9
|
18 dat = read.table(inFile, header=T, sep="\t", dec=",", fill=T, stringsAsFactors=F)
|
|
|
19 dat = dat[!is.na(dat$Patient),]
|
|
|
20
|
|
0
|
21 setwd(outDir)
|
|
3
|
22 cat("<tr><td>Selecting first V/J Genes</td></tr>", file=logfile, append=T)
|
|
2
|
23 dat$V_Segment_Major_Gene = as.factor(as.character(lapply(strsplit(as.character(dat$V_Segment_Major_Gene), "; "), "[[", 1)))
|
|
|
24 dat$J_Segment_Major_Gene = as.factor(as.character(lapply(strsplit(as.character(dat$J_Segment_Major_Gene), "; "), "[[", 1)))
|
|
|
25
|
|
9
|
26 str(dat)
|
|
|
27 cat("<tr><td>Deduplication</td></tr>", file=logfile, append=T)
|
|
11
|
28 #dat = data.frame(data.table(dat)[, list(Patient=unique(.SD$Patient), Clone_Molecule_Count_From_Spikes=sum(.SD$Clone_Molecule_Count_From_Spikes), Log10_Frequency=sum(.SD$Log10_Frequency), Total_Read_Count=sum(.SD$Total_Read_Count), Related_to_leukemia_clone=any(.SD$Related_to_leukemia_clone)), by=c("Sample", "Cell_Count", "J_Segment_Major_Gene", "V_Segment_Major_Gene", "CDR3_Sense_Sequence")])
|
|
|
29
|
|
|
30 most.common = function(x){
|
|
|
31 ux = unique(x)
|
|
|
32 if(length(ux) > 1){
|
|
|
33 xtdf = data.frame(table(x))
|
|
|
34 return(xtdf$Var1[which.max(xtdf$Freq)])
|
|
|
35 #print(xtdf)
|
|
|
36 }
|
|
|
37 return(unique(x))
|
|
|
38 }
|
|
|
39
|
|
|
40 dat = data.frame(data.table(dat)[, list(Patient=unique(.SD$Patient), V_Segment_Major_Gene=most.common(.SD$V_Segment_Major_Gene), J_Segment_Major_Gene=most.common(.SD$J_Segment_Major_Gene), Clone_Molecule_Count_From_Spikes=sum(.SD$Clone_Molecule_Count_From_Spikes), Log10_Frequency=sum(.SD$Log10_Frequency), Total_Read_Count=sum(.SD$Total_Read_Count), Related_to_leukemia_clone=any(.SD$Related_to_leukemia_clone)), by=c("Sample", "Cell_Count", "CDR3_Sense_Sequence")])
|
|
9
|
41
|
|
3
|
42 cat("<tr><td>Calculating Frequency</td></tr>", file=logfile, append=T)
|
|
0
|
43 dat$Frequency = ((10^dat$Log10_Frequency)*100)
|
|
2
|
44
|
|
3
|
45 dat = dat[dat$Frequency >= min_freq,]
|
|
|
46
|
|
|
47 cat("<tr><td>Normalizing cell count to 1.000.000</td></tr>", file=logfile, append=T)
|
|
2
|
48 dat$normalized_read_count = round(dat$Clone_Molecule_Count_From_Spikes / dat$Cell_Count * 1000000 / 2)
|
|
3
|
49 dat = dat[dat$normalized_read_count >= min_cells,]
|
|
2
|
50 dat$paste = paste(dat$Sample, dat$V_Segment_Major_Gene, dat$J_Segment_Major_Gene, dat$CDR3_Sense_Sequence)
|
|
9
|
51 triplets = dat[grepl("VanDongen_cALL_14696", dat$Patient) | grepl("(16278)|(26402)|(26759)", dat$Sample),]
|
|
|
52
|
|
0
|
53 patients = split(dat, dat$Patient, drop=T)
|
|
9
|
54 intervalReads = rev(c(0,10,25,50,100,250,500,750,1000,10000))
|
|
6
|
55 intervalFreq = rev(c(0,0.01,0.05,0.1,0.5,1,5))
|
|
0
|
56 V_Segments = c(".*", "IGHV", "IGHD", "IGKV", "IGKV", "IgKINTR", "TRGV", "TRDV", "TRDD" , "TRBV")
|
|
|
57 J_Segments = c(".*", ".*", ".*", "IGKJ", "KDE", ".*", ".*", ".*", ".*", ".*")
|
|
|
58 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")
|
|
|
59 Titles = factor(Titles, levels=Titles)
|
|
|
60 TitlesOrder = data.frame("Title"=Titles, "TitlesOrder"=1:length(Titles))
|
|
|
61
|
|
|
62 patientCountOnColumn <- function(x, product, interval, on, appendtxt=F){
|
|
|
63 x$Sample = factor(x$Sample, levels=unique(x$Sample))
|
|
|
64 onShort = "reads"
|
|
|
65 if(on == "Frequency"){
|
|
|
66 onShort = "freq"
|
|
|
67 }
|
|
|
68 splt = split(x, x$Sample, drop=T)
|
|
4
|
69 type="pair"
|
|
2
|
70 if(length(splt) == 1){
|
|
3
|
71 print(paste(paste(x[1,which(colnames(x) == "Patient")]), "has one sample"))
|
|
4
|
72 splt[[2]] = data.frame("Patient" = character(0), "Receptor" = character(0), "Sample" = character(0), "Cell_Count" = numeric(0), "Clone_Molecule_Count_From_Spikes" = numeric(0), "Log10_Frequency" = numeric(0), "Total_Read_Count" = numeric(0), "dsMol_per_1e6_cells" = numeric(0), "J_Segment_Major_Gene" = character(0), "V_Segment_Major_Gene" = character(0), "Clone_Sequence" = character(0), "CDR3_Sense_Sequence" = character(0), "Related_to_leukemia_clone" = logical(0), "Frequency"= numeric(0), "normalized_read_count" = numeric(0), "paste" = character(0))
|
|
|
73 type="single"
|
|
2
|
74 }
|
|
0
|
75 patient1 = splt[[1]]
|
|
|
76 patient2 = splt[[2]]
|
|
|
77
|
|
|
78 threshholdIndex = which(colnames(product) == "interval")
|
|
|
79 V_SegmentIndex = which(colnames(product) == "V_Segments")
|
|
|
80 J_SegmentIndex = which(colnames(product) == "J_Segments")
|
|
|
81 titleIndex = which(colnames(product) == "Titles")
|
|
|
82 sampleIndex = which(colnames(x) == "Sample")
|
|
|
83 patientIndex = which(colnames(x) == "Patient")
|
|
|
84 oneSample = paste(patient1[1,sampleIndex], sep="")
|
|
|
85 twoSample = paste(patient2[1,sampleIndex], sep="")
|
|
|
86 patient = paste(x[1,patientIndex])
|
|
3
|
87
|
|
0
|
88 switched = F
|
|
|
89 if(length(grep(".*_Right$", twoSample)) == 1 || length(grep(".*_Dx_BM$", twoSample)) == 1 || length(grep(".*_Dx$", twoSample)) == 1 ){
|
|
|
90 tmp = twoSample
|
|
|
91 twoSample = oneSample
|
|
|
92 oneSample = tmp
|
|
|
93 tmp = patient1
|
|
|
94 patient1 = patient2
|
|
|
95 patient2 = tmp
|
|
|
96 switched = T
|
|
|
97 }
|
|
|
98 if(appendtxt){
|
|
4
|
99 cat(paste(patient, oneSample, twoSample, type, sep="\t"), file="patients.txt", append=T, sep="", fill=3)
|
|
0
|
100 }
|
|
3
|
101 cat(paste("<tr><td>", patient, "</td></tr>", sep=""), file=logfile, append=T)
|
|
9
|
102
|
|
|
103 patient1$merge = paste(patient1$V_Segment_Major_Gene, patient1$J_Segment_Major_Gene, patient1$CDR3_Sense_Sequence)
|
|
|
104 patient2$merge = paste(patient2$V_Segment_Major_Gene, patient2$J_Segment_Major_Gene, patient2$CDR3_Sense_Sequence)
|
|
|
105
|
|
|
106 patientMerge = merge(patient1, patient2, by.x="merge", by.y="merge")
|
|
0
|
107 res1 = vector()
|
|
|
108 res2 = vector()
|
|
|
109 resBoth = vector()
|
|
|
110 read1Count = vector()
|
|
|
111 read2Count = vector()
|
|
2
|
112 locussum1 = vector()
|
|
|
113 locussum2 = vector()
|
|
9
|
114
|
|
|
115 print(patient)
|
|
0
|
116 #for(iter in 1){
|
|
|
117 for(iter in 1:length(product[,1])){
|
|
|
118 threshhold = product[iter,threshholdIndex]
|
|
|
119 V_Segment = paste(".*", as.character(product[iter,V_SegmentIndex]), ".*", sep="")
|
|
|
120 J_Segment = paste(".*", as.character(product[iter,J_SegmentIndex]), ".*", sep="")
|
|
|
121 both = (grepl(V_Segment, patientMerge$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge$J_Segment_Major_Gene.x) & patientMerge[,paste(on, ".x", sep="")] > threshhold & patientMerge[,paste(on, ".y", sep="")] > threshhold)
|
|
10
|
122 one = (grepl(V_Segment, patient1$V_Segment_Major_Gene) & grepl(J_Segment, patient1$J_Segment_Major_Gene) & patient1[,on] > threshhold & !(patient1$CDR3_Sense_Sequence %in% patientMerge[both,]$CDR3_Sense_Sequence.x))
|
|
|
123 two = (grepl(V_Segment, patient2$V_Segment_Major_Gene) & grepl(J_Segment, patient2$J_Segment_Major_Gene) & patient2[,on] > threshhold & !(patient2$CDR3_Sense_Sequence %in% patientMerge[both,]$CDR3_Sense_Sequence.x))
|
|
2
|
124 read1Count = append(read1Count, sum(patient1[one,]$normalized_read_count) + sum(patientMerge[both,]$normalized_read_count.x))
|
|
|
125 read2Count = append(read2Count, sum(patient2[two,]$normalized_read_count) + sum(patientMerge[both,]$normalized_read_count.y))
|
|
0
|
126 res1 = append(res1, sum(one))
|
|
2
|
127 res2 = append(res2, sum(two))
|
|
0
|
128 resBoth = append(resBoth, sum(both))
|
|
2
|
129 locussum1 = append(locussum1, sum(patient1[(grepl(V_Segment, patient1$V_Segment_Major_Gene) & grepl(J_Segment, patient1$J_Segment_Major_Gene)),]$normalized_read_count))
|
|
|
130 locussum2 = append(locussum2, sum(patient2[(grepl(V_Segment, patient2$V_Segment_Major_Gene) & grepl(J_Segment, patient2$J_Segment_Major_Gene)),]$normalized_read_count))
|
|
0
|
131 #threshhold = 0
|
|
|
132 if(threshhold != 0){
|
|
|
133 if(sum(one) > 0){
|
|
9
|
134 dfOne = patient1[one,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "CDR3_Sense_Sequence", "Related_to_leukemia_clone")]
|
|
|
135 colnames(dfOne) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "CDR3 Sequence", "Related_to_leukemia_clone")
|
|
0
|
136 filenameOne = paste(oneSample, "_", product[iter, titleIndex], "_", threshhold, sep="")
|
|
|
137 write.table(dfOne, file=paste(filenameOne, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
138 }
|
|
|
139 if(sum(two) > 0){
|
|
9
|
140 dfTwo = patient2[two,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "CDR3_Sense_Sequence", "Related_to_leukemia_clone")]
|
|
|
141 colnames(dfTwo) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "CDR3 Sequence", "Related_to_leukemia_clone")
|
|
0
|
142 filenameTwo = paste(twoSample, "_", product[iter, titleIndex], "_", threshhold, sep="")
|
|
|
143 write.table(dfTwo, file=paste(filenameTwo, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
144 }
|
|
|
145 }
|
|
|
146 if(sum(both) > 0){
|
|
9
|
147 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", "CDR3_Sense_Sequence.x", "V_Segment_Major_Gene.y", "J_Segment_Major_Gene.y", "normalized_read_count.y", "Frequency.y", "Related_to_leukemia_clone.y")]
|
|
|
148 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),"CDR3 Sequence", paste("Proximal segment", twoSample), paste("Distal segment", twoSample), paste("Normalized_Read_Count", twoSample), paste("Frequency", twoSample), paste("Related_to_leukemia_clone", twoSample))
|
|
0
|
149 filenameBoth = paste(oneSample, "_", twoSample, "_", product[iter, titleIndex], "_", threshhold, sep="")
|
|
|
150 write.table(dfBoth, file=paste(filenameBoth, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
151 }
|
|
|
152 }
|
|
2
|
153 patientResult = data.frame("Locus"=product$Titles, "J_Segment"=product$J_Segments, "V_Segment"=product$V_Segments, "cut_off_value"=paste(">", product$interval, sep=""), "Both"=resBoth, "tmp1"=res1, "read_count1" = round(read1Count), "tmp2"=res2, "read_count2"= round(read2Count), "Sum"=res1 + res2 + resBoth, "percentage" = round((resBoth/(res1 + res2 + resBoth)) * 100, digits=2), "Locus_sum1"=locussum1, "Locus_sum2"=locussum2)
|
|
0
|
154 if(sum(is.na(patientResult$percentage)) > 0){
|
|
|
155 patientResult[is.na(patientResult$percentage),]$percentage = 0
|
|
|
156 }
|
|
|
157 colnames(patientResult)[6] = oneSample
|
|
|
158 colnames(patientResult)[8] = twoSample
|
|
|
159 colnamesBak = colnames(patientResult)
|
|
2
|
160 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))
|
|
0
|
161 write.table(patientResult, file=paste(patient, "_", onShort, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
162 colnames(patientResult) = colnamesBak
|
|
|
163
|
|
|
164 patientResult$Locus = factor(patientResult$Locus, Titles)
|
|
|
165 patientResult$cut_off_value = factor(patientResult$cut_off_value, paste(">", interval, sep=""))
|
|
|
166
|
|
|
167 plt = ggplot(patientResult[,c("Locus", "cut_off_value", "Both")])
|
|
|
168 plt = plt + geom_bar( aes( x=factor(cut_off_value), y=Both), stat='identity', position="dodge", fill="#79c36a")
|
|
|
169 plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1))
|
|
|
170 plt = plt + geom_text(aes(ymax=max(Both), x=cut_off_value,y=Both,label=Both), angle=90, hjust=0)
|
|
|
171 plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("Number of clones in both")
|
|
|
172 plt = plt + theme(plot.margin = unit(c(1,8.8,0.5,1.5), "lines"))
|
|
|
173 png(paste(patient, "_", onShort, ".png", sep=""), width=1920, height=1080)
|
|
|
174 print(plt)
|
|
|
175 dev.off()
|
|
|
176 #(t,r,b,l)
|
|
|
177 plt = ggplot(patientResult[,c("Locus", "cut_off_value", "percentage")])
|
|
|
178 plt = plt + geom_bar( aes( x=factor(cut_off_value), y=percentage), stat='identity', position="dodge", fill="#79c36a")
|
|
|
179 plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1))
|
|
|
180 plt = plt + geom_text(aes(ymax=max(percentage), x=cut_off_value,y=percentage,label=percentage), angle=90, hjust=0)
|
|
|
181 plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("% clones in both left and right")
|
|
|
182 plt = plt + theme(plot.margin = unit(c(1,8.8,0.5,1.5), "lines"))
|
|
|
183 png(paste(patient, "_percent_", onShort, ".png", sep=""), width=1920, height=1080)
|
|
|
184 print(plt)
|
|
|
185 dev.off()
|
|
|
186
|
|
|
187 patientResult = melt(patientResult[,c('Locus','cut_off_value', oneSample, twoSample)] ,id.vars=1:2)
|
|
|
188 patientResult$relativeValue = patientResult$value * 10
|
|
|
189 patientResult[patientResult$relativeValue == 0,]$relativeValue = 1
|
|
|
190 plt = ggplot(patientResult)
|
|
|
191 plt = plt + geom_bar( aes( x=factor(cut_off_value), y=relativeValue, fill=variable), stat='identity', position="dodge")
|
|
|
192 plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1))
|
|
|
193 plt = plt + scale_y_continuous(trans="log", breaks=10^c(0:10), labels=c(0, 10^c(0:9)))
|
|
|
194 plt = plt + geom_text(data=patientResult[patientResult$variable == oneSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=-0.2)
|
|
|
195 plt = plt + geom_text(data=patientResult[patientResult$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)
|
|
|
196 plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle(paste("Number of clones in only ", oneSample, " and only ", twoSample, sep=""))
|
|
|
197 png(paste(patient, "_", onShort, "_both.png", sep=""), width=1920, height=1080)
|
|
|
198 print(plt)
|
|
|
199 dev.off()
|
|
|
200 }
|
|
|
201
|
|
3
|
202 cat("<tr><td>Starting Frequency analysis</td></tr>", file=logfile, append=T)
|
|
|
203
|
|
0
|
204 interval = intervalFreq
|
|
|
205 intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval))
|
|
4
|
206 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)))
|
|
|
207 mclapply(patients, FUN=patientCountOnColumn, product = product, interval=interval, on="Frequency", appendtxt=T)
|
|
0
|
208
|
|
3
|
209 cat("<tr><td>Starting Cell Count analysis</td></tr>", file=logfile, append=T)
|
|
|
210
|
|
0
|
211 interval = intervalReads
|
|
|
212 intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval))
|
|
4
|
213 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)))
|
|
9
|
214 mclapply(patients, FUN=patientCountOnColumn, product = product, interval=interval, on="normalized_read_count")
|
|
0
|
215
|
|
3
|
216 cat("</table></html>", file=logfile, append=T)
|
|
|
217
|
|
7
|
218
|
|
|
219 tripletAnalysis <- function(patient1, label1, patient2, label2, patient3, label3, product, interval, on, appendTriplets= FALSE){
|
|
|
220 onShort = "reads"
|
|
|
221 if(on == "Frequency"){
|
|
|
222 onShort = "freq"
|
|
|
223 }
|
|
|
224 type="triplet"
|
|
|
225
|
|
|
226 threshholdIndex = which(colnames(product) == "interval")
|
|
|
227 V_SegmentIndex = which(colnames(product) == "V_Segments")
|
|
|
228 J_SegmentIndex = which(colnames(product) == "J_Segments")
|
|
|
229 titleIndex = which(colnames(product) == "Titles")
|
|
|
230 sampleIndex = which(colnames(patient1) == "Sample")
|
|
|
231 patientIndex = which(colnames(patient1) == "Patient")
|
|
|
232 oneSample = paste(patient1[1,sampleIndex], sep="")
|
|
|
233 twoSample = paste(patient2[1,sampleIndex], sep="")
|
|
|
234 threeSample = paste(patient3[1,sampleIndex], sep="")
|
|
|
235
|
|
9
|
236 patient1$merge = paste(patient1$V_Segment_Major_Gene, patient1$J_Segment_Major_Gene, patient1$CDR3_Sense_Sequence)
|
|
|
237 patient2$merge = paste(patient2$V_Segment_Major_Gene, patient2$J_Segment_Major_Gene, patient2$CDR3_Sense_Sequence)
|
|
|
238 patient3$merge = paste(patient3$V_Segment_Major_Gene, patient3$J_Segment_Major_Gene, patient3$CDR3_Sense_Sequence)
|
|
|
239
|
|
|
240 patientMerge = merge(patient1, patient2, by="merge")
|
|
|
241 patientMerge = merge(patientMerge, patient3, by="merge")
|
|
|
242 colnames(patientMerge)[28:length(colnames(patientMerge))] = paste(colnames(patientMerge)[28:length(colnames(patientMerge))], ".z", sep="")
|
|
7
|
243 res1 = vector()
|
|
|
244 res2 = vector()
|
|
|
245 res3 = vector()
|
|
|
246 resAll = vector()
|
|
|
247 read1Count = vector()
|
|
|
248 read2Count = vector()
|
|
|
249 read3Count = vector()
|
|
|
250
|
|
|
251 if(appendTriplets){
|
|
9
|
252 cat(paste(label1, label2, label3, sep="\t"), file="triplets.txt", append=T, sep="", fill=3)
|
|
7
|
253 }
|
|
|
254 for(iter in 1:length(product[,1])){
|
|
|
255 threshhold = product[iter,threshholdIndex]
|
|
|
256 V_Segment = paste(".*", as.character(product[iter,V_SegmentIndex]), ".*", sep="")
|
|
|
257 J_Segment = paste(".*", as.character(product[iter,J_SegmentIndex]), ".*", sep="")
|
|
|
258 all = (grepl(V_Segment, patientMerge$V_Segment_Major_Gene.x) & grepl(J_Segment, patientMerge$J_Segment_Major_Gene.x) & patientMerge[,paste(on, ".x", sep="")] > threshhold & patientMerge[,paste(on, ".y", sep="")] > threshhold & patientMerge[,paste(on, ".z", sep="")] > threshhold)
|
|
10
|
259 one = (grepl(V_Segment, patient1$V_Segment_Major_Gene) & grepl(J_Segment, patient1$J_Segment_Major_Gene) & patient1[,on] > threshhold & !(patient1$CDR3_Sense_Sequence %in% patientMerge[all,]$CDR3_Sense_Sequence.x))
|
|
|
260 two = (grepl(V_Segment, patient2$V_Segment_Major_Gene) & grepl(J_Segment, patient2$J_Segment_Major_Gene) & patient2[,on] > threshhold & !(patient2$CDR3_Sense_Sequence %in% patientMerge[all,]$CDR3_Sense_Sequence.x))
|
|
|
261 three = (grepl(V_Segment, patient3$V_Segment_Major_Gene) & grepl(J_Segment, patient3$J_Segment_Major_Gene) & patient3[,on] > threshhold & !(patient3$CDR3_Sense_Sequence %in% patientMerge[all,]$CDR3_Sense_Sequence.x))
|
|
7
|
262
|
|
|
263 read1Count = append(read1Count, sum(patient1[one,]$normalized_read_count) + sum(patientMerge[all,]$normalized_read_count.x))
|
|
|
264 read2Count = append(read2Count, sum(patient2[two,]$normalized_read_count) + sum(patientMerge[all,]$normalized_read_count.y))
|
|
|
265 read3Count = append(read3Count, sum(patient3[three,]$normalized_read_count) + sum(patientMerge[all,]$normalized_read_count.z))
|
|
|
266 res1 = append(res1, sum(one))
|
|
|
267 res2 = append(res2, sum(two))
|
|
|
268 res3 = append(res3, sum(three))
|
|
|
269 resAll = append(resAll, sum(all))
|
|
|
270 #threshhold = 0
|
|
|
271 if(threshhold != 0){
|
|
|
272 if(sum(one) > 0){
|
|
9
|
273 dfOne = patient1[one,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "CDR3_Sense_Sequence", "Related_to_leukemia_clone")]
|
|
7
|
274 colnames(dfOne) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Sequence", "Related_to_leukemia_clone")
|
|
|
275 filenameOne = paste(label1, "_", product[iter, titleIndex], "_", threshhold, sep="")
|
|
|
276 write.table(dfOne, file=paste(filenameOne, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
277 }
|
|
|
278 if(sum(two) > 0){
|
|
9
|
279 dfTwo = patient2[two,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "CDR3_Sense_Sequence", "Related_to_leukemia_clone")]
|
|
7
|
280 colnames(dfTwo) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Sequence", "Related_to_leukemia_clone")
|
|
|
281 filenameTwo = paste(label2, "_", product[iter, titleIndex], "_", threshhold, sep="")
|
|
|
282 write.table(dfTwo, file=paste(filenameTwo, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
283 }
|
|
|
284 if(sum(three) > 0){
|
|
9
|
285 dfThree = patient3[three,c("V_Segment_Major_Gene", "J_Segment_Major_Gene", "normalized_read_count", "Frequency", "CDR3_Sense_Sequence", "Related_to_leukemia_clone")]
|
|
7
|
286 colnames(dfThree) = c("Proximal segment", "Distal segment", "normalized_read_count", "Frequency", "Sequence", "Related_to_leukemia_clone")
|
|
|
287 filenameThree = paste(label3, "_", product[iter, titleIndex], "_", threshhold, sep="")
|
|
|
288 write.table(dfThree, file=paste(filenameThree, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
289 }
|
|
|
290 }
|
|
|
291 if(sum(all) > 0){
|
|
9
|
292 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", "CDR3_Sense_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")]
|
|
|
293 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),"CDR3_Sense_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))
|
|
7
|
294 filenameAll = paste(label1, "_", label2, "_", label3, "_", product[iter, titleIndex], "_", threshhold, sep="")
|
|
|
295 write.table(dfAll, file=paste(filenameAll, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
296 }
|
|
|
297 }
|
|
|
298 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))
|
|
|
299 colnames(patientResult)[6] = oneSample
|
|
|
300 colnames(patientResult)[8] = twoSample
|
|
|
301 colnames(patientResult)[10] = threeSample
|
|
|
302
|
|
|
303 colnamesBak = colnames(patientResult)
|
|
|
304 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("Normalized Read Count", oneSample), paste("Number of sequences", twoSample), paste("Normalized Read Count", twoSample), paste("Number of sequences", threeSample), paste("Normalized Read Count", threeSample))
|
|
|
305 write.table(patientResult, file=paste(label1, "_", label2, "_", label3, "_", onShort, ".txt", sep=""), quote=F, sep="\t", dec=",", row.names=F, col.names=T)
|
|
|
306 colnames(patientResult) = colnamesBak
|
|
|
307
|
|
|
308 patientResult$Locus = factor(patientResult$Locus, Titles)
|
|
|
309 patientResult$cut_off_value = factor(patientResult$cut_off_value, paste(">", interval, sep=""))
|
|
|
310
|
|
|
311 plt = ggplot(patientResult[,c("Locus", "cut_off_value", "All")])
|
|
|
312 plt = plt + geom_bar( aes( x=factor(cut_off_value), y=All), stat='identity', position="dodge", fill="#79c36a")
|
|
|
313 plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1))
|
|
|
314 plt = plt + geom_text(aes(ymax=max(All), x=cut_off_value,y=All,label=All), angle=90, hjust=0)
|
|
|
315 plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("Number of clones in All")
|
|
|
316 plt = plt + theme(plot.margin = unit(c(1,8.8,0.5,1.5), "lines"))
|
|
|
317 png(paste(label1, "_", label2, "_", label3, "_", onShort, "_total_all.png", sep=""), width=1920, height=1080)
|
|
|
318 print(plt)
|
|
|
319 dev.off()
|
|
|
320
|
|
|
321 fontSize = 4
|
|
|
322
|
|
|
323 bak = patientResult
|
|
|
324 patientResult = melt(patientResult[,c('Locus','cut_off_value', oneSample, twoSample, threeSample)] ,id.vars=1:2)
|
|
|
325 patientResult$relativeValue = patientResult$value * 10
|
|
|
326 patientResult[patientResult$relativeValue == 0,]$relativeValue = 1
|
|
|
327 plt = ggplot(patientResult)
|
|
|
328 plt = plt + geom_bar( aes( x=factor(cut_off_value), y=relativeValue, fill=variable), stat='identity', position="dodge")
|
|
|
329 plt = plt + facet_grid(.~Locus) + theme(axis.text.x = element_text(angle = 45, hjust = 1))
|
|
|
330 plt = plt + scale_y_continuous(trans="log", breaks=10^c(0:10), labels=c(0, 10^c(0:9)))
|
|
|
331 plt = plt + geom_text(data=patientResult[patientResult$variable == oneSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=-0.7, size=fontSize)
|
|
|
332 plt = plt + geom_text(data=patientResult[patientResult$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)
|
|
|
333 plt = plt + geom_text(data=patientResult[patientResult$variable == threeSample,], aes(ymax=max(value), x=cut_off_value,y=relativeValue,label=value), angle=90, position=position_dodge(width=0.9), hjust=0, vjust=1.5, size=fontSize)
|
|
|
334 plt = plt + xlab("Reads per locus") + ylab("Count") + ggtitle("Number of clones in only one sample")
|
|
|
335 png(paste(label1, "_", label2, "_", label3, "_", onShort, "_indiv_all.png", sep=""), width=1920, height=1080)
|
|
|
336 print(plt)
|
|
|
337 dev.off()
|
|
|
338 }
|
|
|
339
|
|
9
|
340
|
|
|
341 triplets$uniqueID = "ID"
|
|
|
342
|
|
|
343 triplets[grepl("16278_Left", triplets$Sample),]$uniqueID = "16278_26402_26759_Left"
|
|
|
344 triplets[grepl("26402_Left", triplets$Sample),]$uniqueID = "16278_26402_26759_Left"
|
|
|
345 triplets[grepl("26759_Left", triplets$Sample),]$uniqueID = "16278_26402_26759_Left"
|
|
|
346
|
|
|
347 triplets[grepl("16278_Right", triplets$Sample),]$uniqueID = "16278_26402_26759_Right"
|
|
|
348 triplets[grepl("26402_Right", triplets$Sample),]$uniqueID = "16278_26402_26759_Right"
|
|
|
349 triplets[grepl("26759_Right", triplets$Sample),]$uniqueID = "16278_26402_26759_Right"
|
|
|
350
|
|
|
351 triplets[grepl("14696", triplets$Patient),]$uniqueID = "14696"
|
|
|
352
|
|
|
353 triplets = data.frame(data.table(triplets)[, list(Patient=unique(.SD$uniqueID), Clone_Molecule_Count_From_Spikes=sum(.SD$Clone_Molecule_Count_From_Spikes), Log10_Frequency=sum(.SD$Log10_Frequency), Total_Read_Count=sum(.SD$Total_Read_Count), Related_to_leukemia_clone=any(.SD$Related_to_leukemia_clone)), by=c("Sample", "Cell_Count", "J_Segment_Major_Gene", "V_Segment_Major_Gene", "CDR3_Sense_Sequence")])
|
|
|
354
|
|
|
355 triplets$Frequency = (10^as.numeric(triplets$Log10_Frequency))*100
|
|
|
356 triplets$normalized_read_count = round(triplets$Clone_Molecule_Count_From_Spikes / triplets$Cell_Count * 1000000 / 2)
|
|
|
357
|
|
7
|
358 interval = intervalReads
|
|
|
359 intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval))
|
|
|
360 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)))
|
|
|
361
|
|
9
|
362 one = triplets[triplets$Sample == "14696_reg_BM",]
|
|
|
363 two = triplets[triplets$Sample == "24536_reg_BM",]
|
|
|
364 three = triplets[triplets$Sample == "24062_reg_BM",]
|
|
8
|
365 tripletAnalysis(one, "14696_1", two, "14696_2", three, "14696_3", product=product, interval=interval, on="normalized_read_count", T)
|
|
7
|
366
|
|
9
|
367 one = triplets[triplets$Sample == "16278_Left",]
|
|
|
368 two = triplets[triplets$Sample == "26402_Left",]
|
|
|
369 three = triplets[triplets$Sample == "26759_Left",]
|
|
8
|
370 tripletAnalysis(one, "16278_Left", two, "26402_Left", three, "26759_Left", product=product, interval=interval, on="normalized_read_count", T)
|
|
7
|
371
|
|
9
|
372 one = triplets[triplets$Sample == "16278_Right",]
|
|
|
373 two = triplets[triplets$Sample == "26402_Right",]
|
|
|
374 three = triplets[triplets$Sample == "26759_Right",]
|
|
8
|
375 tripletAnalysis(one, "16278_Right", two, "26402_Right", three, "26759_Right", product=product, interval=interval, on="normalized_read_count", T)
|
|
7
|
376
|
|
|
377
|
|
|
378 interval = intervalFreq
|
|
|
379 intervalOrder = data.frame("interval"=paste(">", interval, sep=""), "intervalOrder"=1:length(interval))
|
|
|
380 product = data.frame("Titles"=rep(Titles, each=length(interval)), "interval"=rep(interval, times=10), "V_Segments"=rep(V_Segments, each=length(interval)), "J_Segments"=rep(J_Segments, each=length(interval)))
|
|
|
381
|
|
9
|
382 one = triplets[triplets$Sample == "14696_reg_BM",]
|
|
|
383 two = triplets[triplets$Sample == "24536_reg_BM",]
|
|
|
384 three = triplets[triplets$Sample == "24062_reg_BM",]
|
|
8
|
385 tripletAnalysis(one, "14696_1", two, "14696_2", three, "14696_3", product=product, interval=interval, on="Frequency", F)
|
|
7
|
386
|
|
9
|
387 one = triplets[triplets$Sample == "16278_Left",]
|
|
|
388 two = triplets[triplets$Sample == "26402_Left",]
|
|
|
389 three = triplets[triplets$Sample == "26759_Left",]
|
|
8
|
390 tripletAnalysis(one, "16278_Left", two, "26402_Left", three, "26759_Left", product=product, interval=interval, on="Frequency", F)
|
|
7
|
391
|
|
9
|
392 one = triplets[triplets$Sample == "16278_Right",]
|
|
|
393 two = triplets[triplets$Sample == "26402_Right",]
|
|
|
394 three = triplets[triplets$Sample == "26759_Right",]
|
|
8
|
395 tripletAnalysis(one, "16278_Right", two, "26402_Right", three, "26759_Right", product=product, interval=interval, on="Frequency", F)
|
