Mercurial > repos > davidvanzessen > argalaxy_tools

comparison report_clonality/RScript.r.old @ 20:9185c3dfc679 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Fri, 27 Jan 2017 03:44:18 -0500
parents
children
comparison
equal deleted inserted replaced
19:3ef457aa5df6 20:9185c3dfc679
1 # ---------------------- load/install packages ----------------------
2
3 if (!("gridExtra" %in% rownames(installed.packages()))) {
4 install.packages("gridExtra", repos="http://cran.xl-mirror.nl/")
5 }
6 library(gridExtra)
7 if (!("ggplot2" %in% rownames(installed.packages()))) {
8 install.packages("ggplot2", repos="http://cran.xl-mirror.nl/")
9 }
10 library(ggplot2)
11 if (!("plyr" %in% rownames(installed.packages()))) {
12 install.packages("plyr", repos="http://cran.xl-mirror.nl/")
13 }
14 library(plyr)
15
16 if (!("data.table" %in% rownames(installed.packages()))) {
17 install.packages("data.table", repos="http://cran.xl-mirror.nl/")
18 }
19 library(data.table)
20
21 if (!("reshape2" %in% rownames(installed.packages()))) {
22 install.packages("reshape2", repos="http://cran.xl-mirror.nl/")
23 }
24 library(reshape2)
25
26 if (!("lymphclon" %in% rownames(installed.packages()))) {
27 install.packages("lymphclon", repos="http://cran.xl-mirror.nl/")
28 }
29 library(lymphclon)
30
31 # ---------------------- parameters ----------------------
32
33 args <- commandArgs(trailingOnly = TRUE)
34
35 infile = args[1] #path to input file
36 outfile = args[2] #path to output file
37 outdir = args[3] #path to output folder (html/images/data)
38 clonaltype = args[4] #clonaltype definition, or 'none' for no unique filtering
39 ct = unlist(strsplit(clonaltype, ","))
40 species = args[5] #human or mouse
41 locus = args[6] # IGH, IGK, IGL, TRB, TRA, TRG or TRD
42 filterproductive = ifelse(args[7] == "yes", T, F) #should unproductive sequences be filtered out? (yes/no)
43 clonality_method = args[8]
44
45
46 # ---------------------- Data preperation ----------------------
47
48 print("Report Clonality - Data preperation")
49
50 inputdata = read.table(infile, sep="\t", header=TRUE, fill=T, comment.char="", stringsAsFactors=F)
51
52 print(paste("nrows: ", nrow(inputdata)))
53
54 setwd(outdir)
55
56 # remove weird rows
57 inputdata = inputdata[inputdata$Sample != "",]
58
59 print(paste("nrows: ", nrow(inputdata)))
60
61 #remove the allele from the V,D and J genes
62 inputdata$Top.V.Gene = gsub("[*]([0-9]+)", "", inputdata$Top.V.Gene)
63 inputdata$Top.D.Gene = gsub("[*]([0-9]+)", "", inputdata$Top.D.Gene)
64 inputdata$Top.J.Gene = gsub("[*]([0-9]+)", "", inputdata$Top.J.Gene)
65
66 print(paste("nrows: ", nrow(inputdata)))
67
68 #filter uniques
69 inputdata.removed = inputdata[NULL,]
70
71 print(paste("nrows: ", nrow(inputdata)))
72
73 inputdata$clonaltype = 1:nrow(inputdata)
74
75 #keep track of the count of sequences in samples or samples/replicates for the front page overview
76 input.sample.count = data.frame(data.table(inputdata)[, list(All=.N), by=c("Sample")])
77 input.rep.count = data.frame(data.table(inputdata)[, list(All=.N), by=c("Sample", "Replicate")])
78
79 PRODF = inputdata
80 UNPROD = inputdata
81 if(filterproductive){
82 if("Functionality" %in% colnames(inputdata)) { # "Functionality" is an IMGT column
83 #PRODF = inputdata[inputdata$Functionality == "productive" | inputdata$Functionality == "productive (see comment)", ]
84 PRODF = inputdata[inputdata$Functionality %in% c("productive (see comment)","productive"),]
85
86 PRODF.count = data.frame(data.table(PRODF)[, list(count=.N), by=c("Sample")])
87
88 UNPROD = inputdata[inputdata$Functionality %in% c("unproductive (see comment)","unproductive"), ]
89 } else {
90 PRODF = inputdata[inputdata$VDJ.Frame != "In-frame with stop codon" & inputdata$VDJ.Frame != "Out-of-frame" & inputdata$CDR3.Found.How != "NOT_FOUND" , ]
91 UNPROD = inputdata[!(inputdata$VDJ.Frame != "In-frame with stop codon" & inputdata$VDJ.Frame != "Out-of-frame" & inputdata$CDR3.Found.How != "NOT_FOUND" ), ]
92 }
93 }
94
95 for(i in 1:nrow(UNPROD)){
96 if(!is.numeric(UNPROD[i,"CDR3.Length"])){
97 UNPROD[i,"CDR3.Length"] = 0
98 }
99 }
100
101 prod.sample.count = data.frame(data.table(PRODF)[, list(Productive=.N), by=c("Sample")])
102 prod.rep.count = data.frame(data.table(PRODF)[, list(Productive=.N), by=c("Sample", "Replicate")])
103
104 unprod.sample.count = data.frame(data.table(UNPROD)[, list(Unproductive=.N), by=c("Sample")])
105 unprod.rep.count = data.frame(data.table(UNPROD)[, list(Unproductive=.N), by=c("Sample", "Replicate")])
106
107 clonalityFrame = PRODF
108
109 #remove duplicates based on the clonaltype
110 if(clonaltype != "none"){
111 clonaltype = paste(clonaltype, ",Sample", sep="") #add sample column to clonaltype, unique within samples
112 PRODF$clonaltype = do.call(paste, c(PRODF[unlist(strsplit(clonaltype, ","))], sep = ":"))
113 PRODF = PRODF[!duplicated(PRODF$clonaltype), ]
114
115 UNPROD$clonaltype = do.call(paste, c(UNPROD[unlist(strsplit(clonaltype, ","))], sep = ":"))
116 UNPROD = UNPROD[!duplicated(UNPROD$clonaltype), ]
117
118 #again for clonalityFrame but with sample+replicate
119 clonalityFrame$clonaltype = do.call(paste, c(clonalityFrame[unlist(strsplit(clonaltype, ","))], sep = ":"))
120 clonalityFrame$clonality_clonaltype = do.call(paste, c(clonalityFrame[unlist(strsplit(paste(clonaltype, ",Replicate", sep=""), ","))], sep = ":"))
121 clonalityFrame = clonalityFrame[!duplicated(clonalityFrame$clonality_clonaltype), ]
122 }
123
124 print("SAMPLE TABLE:")
125 print(table(PRODF$Sample))
126
127 prod.unique.sample.count = data.frame(data.table(PRODF)[, list(Productive_unique=.N), by=c("Sample")])
128 prod.unique.rep.count = data.frame(data.table(PRODF)[, list(Productive_unique=.N), by=c("Sample", "Replicate")])
129
130 unprod.unique.sample.count = data.frame(data.table(UNPROD)[, list(Unproductive_unique=.N), by=c("Sample")])
131 unprod.unique.rep.count = data.frame(data.table(UNPROD)[, list(Unproductive_unique=.N), by=c("Sample", "Replicate")])
132
133 PRODF$freq = 1
134
135 if(any(grepl(pattern="_", x=PRODF$ID))){ #the frequency can be stored in the ID with the pattern ".*_freq_.*"
136 PRODF$freq = gsub("^[0-9]+_", "", PRODF$ID)
137 PRODF$freq = gsub("_.*", "", PRODF$freq)
138 PRODF$freq = as.numeric(PRODF$freq)
139 if(any(is.na(PRODF$freq))){ #if there was an "_" in the ID, but not the frequency, go back to frequency of 1 for every sequence
140 PRODF$freq = 1
141 }
142 }
143
144 #make a names list with sample -> color
145 naive.colors = c('blue4', 'darkred', 'olivedrab3', 'red', 'gray74', 'darkviolet', 'lightblue1', 'gold', 'chartreuse2', 'pink', 'Paleturquoise3', 'Chocolate1', 'Yellow', 'Deeppink3', 'Mediumorchid1', 'Darkgreen', 'Blue', 'Gray36', 'Hotpink', 'Yellow4')
146 unique.samples = unique(PRODF$Sample)
147
148 if(length(unique.samples) <= length(naive.colors)){
149 sample.colors = naive.colors[1:length(unique.samples)]
150 } else {
151 sample.colors = rainbow(length(unique.samples))
152 }
153
154 names(sample.colors) = unique.samples
155
156 print("Sample.colors")
157 print(sample.colors)
158
159
160 #write the complete dataset that is left over, will be the input if 'none' for clonaltype and 'no' for filterproductive
161 write.table(PRODF, "allUnique.txt", sep="\t",quote=F,row.names=F,col.names=T)
162 write.table(PRODF, "allUnique.csv", sep=",",quote=F,row.names=F,col.names=T)
163 write.table(UNPROD, "allUnproductive.csv", sep=",",quote=F,row.names=F,col.names=T)
164
165 #write the samples to a file
166 sampleFile <- file("samples.txt")
167 un = unique(inputdata$Sample)
168 un = paste(un, sep="\n")
169 writeLines(un, sampleFile)
170 close(sampleFile)
171
172 # ---------------------- Counting the productive/unproductive and unique sequences ----------------------
173
174 print("Report Clonality - counting productive/unproductive/unique")
175
176 #create the table on the overview page with the productive/unique counts per sample/replicate
177 #first for sample
178 sample.count = merge(input.sample.count, prod.sample.count, by="Sample", all.x=T)
179 sample.count$perc_prod = round(sample.count$Productive / sample.count$All * 100)
180 sample.count = merge(sample.count, prod.unique.sample.count, by="Sample", all.x=T)
181 sample.count$perc_prod_un = round(sample.count$Productive_unique / sample.count$All * 100)
182
183 sample.count = merge(sample.count , unprod.sample.count, by="Sample", all.x=T)
184 sample.count$perc_unprod = round(sample.count$Unproductive / sample.count$All * 100)
185 sample.count = merge(sample.count, unprod.unique.sample.count, by="Sample", all.x=T)
186 sample.count$perc_unprod_un = round(sample.count$Unproductive_unique / sample.count$All * 100)
187
188 #then sample/replicate
189 rep.count = merge(input.rep.count, prod.rep.count, by=c("Sample", "Replicate"), all.x=T)
190 rep.count$perc_prod = round(rep.count$Productive / rep.count$All * 100)
191 rep.count = merge(rep.count, prod.unique.rep.count, by=c("Sample", "Replicate"), all.x=T)
192 rep.count$perc_prod_un = round(rep.count$Productive_unique / rep.count$All * 100)
193
194 rep.count = merge(rep.count, unprod.rep.count, by=c("Sample", "Replicate"), all.x=T)
195 rep.count$perc_unprod = round(rep.count$Unproductive / rep.count$All * 100)
196 rep.count = merge(rep.count, unprod.unique.rep.count, by=c("Sample", "Replicate"), all.x=T)
197 rep.count$perc_unprod_un = round(rep.count$Unproductive_unique / rep.count$All * 100)
198
199 rep.count$Sample = paste(rep.count$Sample, rep.count$Replicate, sep="_")
200 rep.count = rep.count[,names(rep.count) != "Replicate"]
201
202 count = rbind(sample.count, rep.count)
203
204
205
206 write.table(x=count, file="productive_counting.txt", sep=",",quote=F,row.names=F,col.names=F)
207
208 # ---------------------- V+J+CDR3 sequence count ----------------------
209
210 VJCDR3.count = data.frame(table(clonalityFrame$Top.V.Gene, clonalityFrame$Top.J.Gene, clonalityFrame$CDR3.Seq.DNA))
211 names(VJCDR3.count) = c("Top.V.Gene", "Top.J.Gene", "CDR3.Seq.DNA", "Count")
212
213 VJCDR3.count = VJCDR3.count[VJCDR3.count$Count > 0,]
214 VJCDR3.count = VJCDR3.count[order(-VJCDR3.count$Count),]
215
216 write.table(x=VJCDR3.count, file="VJCDR3_count.txt", sep="\t",quote=F,row.names=F,col.names=T)
217
218 # ---------------------- Frequency calculation for V, D and J ----------------------
219
220 print("Report Clonality - frequency calculation V, D and J")
221
222 PRODFV = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.V.Gene")])
223 Total = ddply(PRODFV, .(Sample), function(x) data.frame(Total = sum(x$Length)))
224 PRODFV = merge(PRODFV, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
225 PRODFV = ddply(PRODFV, c("Sample", "Top.V.Gene"), summarise, relFreq= (Length*100 / Total))
226
227 PRODFD = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.D.Gene")])
228 Total = ddply(PRODFD, .(Sample), function(x) data.frame(Total = sum(x$Length)))
229 PRODFD = merge(PRODFD, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
230 PRODFD = ddply(PRODFD, c("Sample", "Top.D.Gene"), summarise, relFreq= (Length*100 / Total))
231
232 PRODFJ = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.J.Gene")])
233 Total = ddply(PRODFJ, .(Sample), function(x) data.frame(Total = sum(x$Length)))
234 PRODFJ = merge(PRODFJ, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
235 PRODFJ = ddply(PRODFJ, c("Sample", "Top.J.Gene"), summarise, relFreq= (Length*100 / Total))
236
237 # ---------------------- Setting up the gene names for the different species/loci ----------------------
238
239 print("Report Clonality - getting genes for species/loci")
240
241 Vchain = ""
242 Dchain = ""
243 Jchain = ""
244
245 if(species == "custom"){
246 print("Custom genes: ")
247 splt = unlist(strsplit(locus, ";"))
248 print(paste("V:", splt[1]))
249 print(paste("D:", splt[2]))
250 print(paste("J:", splt[3]))
251
252 Vchain = unlist(strsplit(splt[1], ","))
253 Vchain = data.frame(v.name = Vchain, chr.orderV = 1:length(Vchain))
254
255 Dchain = unlist(strsplit(splt[2], ","))
256 if(length(Dchain) > 0){
257 Dchain = data.frame(v.name = Dchain, chr.orderD = 1:length(Dchain))
258 } else {
259 Dchain = data.frame(v.name = character(0), chr.orderD = numeric(0))
260 }
261
262 Jchain = unlist(strsplit(splt[3], ","))
263 Jchain = data.frame(v.name = Jchain, chr.orderJ = 1:length(Jchain))
264
265 } else {
266 genes = read.table("genes.txt", sep="\t", header=TRUE, fill=T, comment.char="")
267
268 Vchain = genes[grepl(species, genes$Species) & genes$locus == locus & genes$region == "V",c("IMGT.GENE.DB", "chr.order")]
269 colnames(Vchain) = c("v.name", "chr.orderV")
270 Dchain = genes[grepl(species, genes$Species) & genes$locus == locus & genes$region == "D",c("IMGT.GENE.DB", "chr.order")]
271 colnames(Dchain) = c("v.name", "chr.orderD")
272 Jchain = genes[grepl(species, genes$Species) & genes$locus == locus & genes$region == "J",c("IMGT.GENE.DB", "chr.order")]
273 colnames(Jchain) = c("v.name", "chr.orderJ")
274 }
275 useD = TRUE
276 if(nrow(Dchain) == 0){
277 useD = FALSE
278 cat("No D Genes in this species/locus")
279 }
280 print(paste(nrow(Vchain), "genes in V"))
281 print(paste(nrow(Dchain), "genes in D"))
282 print(paste(nrow(Jchain), "genes in J"))
283
284 # ---------------------- merge with the frequency count ----------------------
285
286 PRODFV = merge(PRODFV, Vchain, by.x='Top.V.Gene', by.y='v.name', all.x=TRUE)
287
288 PRODFD = merge(PRODFD, Dchain, by.x='Top.D.Gene', by.y='v.name', all.x=TRUE)
289
290 PRODFJ = merge(PRODFJ, Jchain, by.x='Top.J.Gene', by.y='v.name', all.x=TRUE)
291
292 # ---------------------- Create the V, D and J frequency plots and write the data.frame for every plot to a file ----------------------
293
294 print("Report Clonality - V, D and J frequency plots")
295
296 pV = ggplot(PRODFV)
297 pV = pV + geom_bar( aes( x=factor(reorder(Top.V.Gene, chr.orderV)), y=relFreq, fill=Sample), stat='identity', position="dodge") + theme(axis.text.x = element_text(angle = 90, hjust = 1))
298 pV = pV + xlab("Summary of V gene") + ylab("Frequency") + ggtitle("Relative frequency of V gene usage") + scale_fill_manual(values=sample.colors)
299 pV = pV + theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
300 write.table(x=PRODFV, file="VFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
301
302 png("VPlot.png",width = 1280, height = 720)
303 pV
304 dev.off();
305
306 if(useD){
307 pD = ggplot(PRODFD)
308 pD = pD + geom_bar( aes( x=factor(reorder(Top.D.Gene, chr.orderD)), y=relFreq, fill=Sample), stat='identity', position="dodge") + theme(axis.text.x = element_text(angle = 90, hjust = 1))
309 pD = pD + xlab("Summary of D gene") + ylab("Frequency") + ggtitle("Relative frequency of D gene usage") + scale_fill_manual(values=sample.colors)
310 pD = pD + theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
311 write.table(x=PRODFD, file="DFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
312
313 png("DPlot.png",width = 800, height = 600)
314 print(pD)
315 dev.off();
316 }
317
318 pJ = ggplot(PRODFJ)
319 pJ = pJ + geom_bar( aes( x=factor(reorder(Top.J.Gene, chr.orderJ)), y=relFreq, fill=Sample), stat='identity', position="dodge") + theme(axis.text.x = element_text(angle = 90, hjust = 1))
320 pJ = pJ + xlab("Summary of J gene") + ylab("Frequency") + ggtitle("Relative frequency of J gene usage") + scale_fill_manual(values=sample.colors)
321 pJ = pJ + theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
322 write.table(x=PRODFJ, file="JFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
323
324 png("JPlot.png",width = 800, height = 600)
325 pJ
326 dev.off();
327
328 # ---------------------- Now the frequency plots of the V, D and J families ----------------------
329
330 print("Report Clonality - V, D and J family plots")
331
332 VGenes = PRODF[,c("Sample", "Top.V.Gene")]
333 VGenes$Top.V.Gene = gsub("-.*", "", VGenes$Top.V.Gene)
334 VGenes = data.frame(data.table(VGenes)[, list(Count=.N), by=c("Sample", "Top.V.Gene")])
335 TotalPerSample = data.frame(data.table(VGenes)[, list(total=sum(.SD$Count)), by=Sample])
336 VGenes = merge(VGenes, TotalPerSample, by="Sample")
337 VGenes$Frequency = VGenes$Count * 100 / VGenes$total
338 VPlot = ggplot(VGenes)
339 VPlot = VPlot + geom_bar(aes( x = Top.V.Gene, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
340 ggtitle("Distribution of V gene families") +
341 ylab("Percentage of sequences") +
342 scale_fill_manual(values=sample.colors) +
343 theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
344 png("VFPlot.png")
345 VPlot
346 dev.off();
347 write.table(x=VGenes, file="VFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
348
349 if(useD){
350 DGenes = PRODF[,c("Sample", "Top.D.Gene")]
351 DGenes$Top.D.Gene = gsub("-.*", "", DGenes$Top.D.Gene)
352 DGenes = data.frame(data.table(DGenes)[, list(Count=.N), by=c("Sample", "Top.D.Gene")])
353 TotalPerSample = data.frame(data.table(DGenes)[, list(total=sum(.SD$Count)), by=Sample])
354 DGenes = merge(DGenes, TotalPerSample, by="Sample")
355 DGenes$Frequency = DGenes$Count * 100 / DGenes$total
356 DPlot = ggplot(DGenes)
357 DPlot = DPlot + geom_bar(aes( x = Top.D.Gene, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
358 ggtitle("Distribution of D gene families") +
359 ylab("Percentage of sequences") +
360 scale_fill_manual(values=sample.colors) +
361 theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
362 png("DFPlot.png")
363 print(DPlot)
364 dev.off();
365 write.table(x=DGenes, file="DFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
366 }
367
368 # ---------------------- Plotting the cdr3 length ----------------------
369
370 print("Report Clonality - CDR3 length plot")
371
372 CDR3Length = data.frame(data.table(PRODF)[, list(Count=.N), by=c("Sample", "CDR3.Length")])
373 TotalPerSample = data.frame(data.table(CDR3Length)[, list(total=sum(.SD$Count)), by=Sample])
374 CDR3Length = merge(CDR3Length, TotalPerSample, by="Sample")
375 CDR3Length$Frequency = CDR3Length$Count * 100 / CDR3Length$total
376 CDR3LengthPlot = ggplot(CDR3Length)
377 CDR3LengthPlot = CDR3LengthPlot + geom_bar(aes( x = factor(reorder(CDR3.Length, as.numeric(CDR3.Length))), y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
378 ggtitle("Length distribution of CDR3") +
379 xlab("CDR3 Length") +
380 ylab("Percentage of sequences") +
381 scale_fill_manual(values=sample.colors) +
382 theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
383 png("CDR3LengthPlot.png",width = 1280, height = 720)
384 CDR3LengthPlot
385 dev.off()
386 write.table(x=CDR3Length, file="CDR3LengthPlot.csv", sep=",",quote=F,row.names=F,col.names=T)
387
388 # ---------------------- Plot the heatmaps ----------------------
389
390 #get the reverse order for the V and D genes
391 revVchain = Vchain
392 revDchain = Dchain
393 revVchain$chr.orderV = rev(revVchain$chr.orderV)
394 revDchain$chr.orderD = rev(revDchain$chr.orderD)
395
396 if(useD){
397 print("Report Clonality - Heatmaps VD")
398 plotVD <- function(dat){
399 if(length(dat[,1]) == 0){
400 return()
401 }
402
403 img = ggplot() +
404 geom_tile(data=dat, aes(x=factor(reorder(Top.D.Gene, chr.orderD)), y=factor(reorder(Top.V.Gene, chr.orderV)), fill=relLength)) +
405 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
406 scale_fill_gradient(low="gold", high="blue", na.value="white") +
407 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
408 xlab("D genes") +
409 ylab("V Genes") +
410 theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), panel.grid.major = element_line(colour = "gainsboro"))
411
412 png(paste("HeatmapVD_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Dchain$v.name)), height=100+(15*length(Vchain$v.name)))
413 print(img)
414 dev.off()
415 write.table(x=acast(dat, Top.V.Gene~Top.D.Gene, value.var="Length"), file=paste("HeatmapVD_", unique(dat[3])[1,1], ".csv", sep=""), sep=",",quote=F,row.names=T,col.names=NA)
416 }
417
418 VandDCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.V.Gene", "Top.D.Gene", "Sample")])
419
420 VandDCount$l = log(VandDCount$Length)
421 maxVD = data.frame(data.table(VandDCount)[, list(max=max(l)), by=c("Sample")])
422 VandDCount = merge(VandDCount, maxVD, by.x="Sample", by.y="Sample", all.x=T)
423 VandDCount$relLength = VandDCount$l / VandDCount$max
424 check = is.nan(VandDCount$relLength)
425 if(any(check)){
426 VandDCount[check,"relLength"] = 0
427 }
428
429 cartegianProductVD = expand.grid(Top.V.Gene = Vchain$v.name, Top.D.Gene = Dchain$v.name)
430
431 completeVD = merge(VandDCount, cartegianProductVD, by.x=c("Top.V.Gene", "Top.D.Gene"), by.y=c("Top.V.Gene", "Top.D.Gene"), all=TRUE)
432
433 completeVD = merge(completeVD, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE)
434
435 completeVD = merge(completeVD, Dchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE)
436
437 fltr = is.nan(completeVD$relLength)
438 if(all(fltr)){
439 completeVD[fltr,"relLength"] = 0
440 }
441
442 VDList = split(completeVD, f=completeVD[,"Sample"])
443 lapply(VDList, FUN=plotVD)
444 }
445
446 print("Report Clonality - Heatmaps VJ")
447
448 plotVJ <- function(dat){
449 if(length(dat[,1]) == 0){
450 return()
451 }
452 cat(paste(unique(dat[3])[1,1]))
453 img = ggplot() +
454 geom_tile(data=dat, aes(x=factor(reorder(Top.J.Gene, chr.orderJ)), y=factor(reorder(Top.V.Gene, chr.orderV)), fill=relLength)) +
455 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
456 scale_fill_gradient(low="gold", high="blue", na.value="white") +
457 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
458 xlab("J genes") +
459 ylab("V Genes") +
460 theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), panel.grid.major = element_line(colour = "gainsboro"))
461
462 png(paste("HeatmapVJ_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Jchain$v.name)), height=100+(15*length(Vchain$v.name)))
463 print(img)
464 dev.off()
465 write.table(x=acast(dat, Top.V.Gene~Top.J.Gene, value.var="Length"), file=paste("HeatmapVJ_", unique(dat[3])[1,1], ".csv", sep=""), sep=",",quote=F,row.names=T,col.names=NA)
466 }
467
468 VandJCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.V.Gene", "Top.J.Gene", "Sample")])
469
470 VandJCount$l = log(VandJCount$Length)
471 maxVJ = data.frame(data.table(VandJCount)[, list(max=max(l)), by=c("Sample")])
472 VandJCount = merge(VandJCount, maxVJ, by.x="Sample", by.y="Sample", all.x=T)
473 VandJCount$relLength = VandJCount$l / VandJCount$max
474
475 check = is.nan(VandJCount$relLength)
476 if(any(check)){
477 VandJCount[check,"relLength"] = 0
478 }
479
480 cartegianProductVJ = expand.grid(Top.V.Gene = Vchain$v.name, Top.J.Gene = Jchain$v.name)
481
482 completeVJ = merge(VandJCount, cartegianProductVJ, all.y=TRUE)
483 completeVJ = merge(completeVJ, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE)
484 completeVJ = merge(completeVJ, Jchain, by.x="Top.J.Gene", by.y="v.name", all.x=TRUE)
485
486 fltr = is.nan(completeVJ$relLength)
487 if(any(fltr)){
488 completeVJ[fltr,"relLength"] = 1
489 }
490
491 VJList = split(completeVJ, f=completeVJ[,"Sample"])
492 lapply(VJList, FUN=plotVJ)
493
494
495
496 if(useD){
497 print("Report Clonality - Heatmaps DJ")
498 plotDJ <- function(dat){
499 if(length(dat[,1]) == 0){
500 return()
501 }
502 img = ggplot() +
503 geom_tile(data=dat, aes(x=factor(reorder(Top.J.Gene, chr.orderJ)), y=factor(reorder(Top.D.Gene, chr.orderD)), fill=relLength)) +
504 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
505 scale_fill_gradient(low="gold", high="blue", na.value="white") +
506 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
507 xlab("J genes") +
508 ylab("D Genes") +
509 theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), panel.grid.major = element_line(colour = "gainsboro"))
510
511 png(paste("HeatmapDJ_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Jchain$v.name)), height=100+(15*length(Dchain$v.name)))
512 print(img)
513 dev.off()
514 write.table(x=acast(dat, Top.D.Gene~Top.J.Gene, value.var="Length"), file=paste("HeatmapDJ_", unique(dat[3])[1,1], ".csv", sep=""), sep=",",quote=F,row.names=T,col.names=NA)
515 }
516
517
518 DandJCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.D.Gene", "Top.J.Gene", "Sample")])
519
520 DandJCount$l = log(DandJCount$Length)
521 maxDJ = data.frame(data.table(DandJCount)[, list(max=max(l)), by=c("Sample")])
522 DandJCount = merge(DandJCount, maxDJ, by.x="Sample", by.y="Sample", all.x=T)
523 DandJCount$relLength = DandJCount$l / DandJCount$max
524
525 check = is.nan(DandJCount$relLength)
526 if(any(check)){
527 DandJCount[check,"relLength"] = 0
528 }
529
530 cartegianProductDJ = expand.grid(Top.D.Gene = Dchain$v.name, Top.J.Gene = Jchain$v.name)
531
532 completeDJ = merge(DandJCount, cartegianProductDJ, all.y=TRUE)
533 completeDJ = merge(completeDJ, revDchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE)
534 completeDJ = merge(completeDJ, Jchain, by.x="Top.J.Gene", by.y="v.name", all.x=TRUE)
535
536 fltr = is.nan(completeDJ$relLength)
537 if(any(fltr)){
538 completeDJ[fltr, "relLength"] = 1
539 }
540
541 DJList = split(completeDJ, f=completeDJ[,"Sample"])
542 lapply(DJList, FUN=plotDJ)
543 }
544
545
546 # ---------------------- output tables for the circos plots ----------------------
547
548 print("Report Clonality - Circos data")
549
550 for(smpl in unique(PRODF$Sample)){
551 PRODF.sample = PRODF[PRODF$Sample == smpl,]
552
553 fltr = PRODF.sample$Top.V.Gene == ""
554 if(any(fltr, na.rm=T)){
555 PRODF.sample[fltr, "Top.V.Gene"] = "NA"
556 }
557
558 fltr = PRODF.sample$Top.D.Gene == ""
559 if(any(fltr, na.rm=T)){
560 PRODF.sample[fltr, "Top.D.Gene"] = "NA"
561 }
562
563 fltr = PRODF.sample$Top.J.Gene == ""
564 if(any(fltr, na.rm=T)){
565 PRODF.sample[fltr, "Top.J.Gene"] = "NA"
566 }
567
568 v.d = table(PRODF.sample$Top.V.Gene, PRODF.sample$Top.D.Gene)
569 v.j = table(PRODF.sample$Top.V.Gene, PRODF.sample$Top.J.Gene)
570 d.j = table(PRODF.sample$Top.D.Gene, PRODF.sample$Top.J.Gene)
571
572 write.table(v.d, file=paste(smpl, "_VD_circos.txt", sep=""), sep="\t", quote=F, row.names=T, col.names=NA)
573 write.table(v.j, file=paste(smpl, "_VJ_circos.txt", sep=""), sep="\t", quote=F, row.names=T, col.names=NA)
574 write.table(d.j, file=paste(smpl, "_DJ_circos.txt", sep=""), sep="\t", quote=F, row.names=T, col.names=NA)
575 }
576
577 # ---------------------- calculating the clonality score ----------------------
578
579 if("Replicate" %in% colnames(inputdata)) #can only calculate clonality score when replicate information is available
580 {
581 print("Report Clonality - Clonality")
582 write.table(clonalityFrame, "clonalityComplete.csv", sep=",",quote=F,row.names=F,col.names=T)
583 if(clonality_method == "boyd"){
584 samples = split(clonalityFrame, clonalityFrame$Sample, drop=T)
585
586 for (sample in samples){
587 res = data.frame(paste=character(0))
588 sample_id = unique(sample$Sample)[[1]]
589 for(replicate in unique(sample$Replicate)){
590 tmp = sample[sample$Replicate == replicate,]
591 clone_table = data.frame(table(tmp$clonaltype))
592 clone_col_name = paste("V", replicate, sep="")
593 colnames(clone_table) = c("paste", clone_col_name)
594 res = merge(res, clone_table, by="paste", all=T)
595 }
596
597 res[is.na(res)] = 0
598 infer.result = infer.clonality(as.matrix(res[,2:ncol(res)]))
599
600 #print(infer.result)
601
602 write.table(data.table(infer.result[[12]]), file=paste("lymphclon_clonality_", sample_id, ".csv", sep=""), sep=",",quote=F,row.names=F,col.names=F)
603
604 res$type = rowSums(res[,2:ncol(res)])
605
606 coincidence.table = data.frame(table(res$type))
607 colnames(coincidence.table) = c("Coincidence Type", "Raw Coincidence Freq")
608 write.table(coincidence.table, file=paste("lymphclon_coincidences_", sample_id, ".csv", sep=""), sep=",",quote=F,row.names=F,col.names=T)
609 }
610 } else {
611 clonalFreq = data.frame(data.table(clonalityFrame)[, list(Type=.N), by=c("Sample", "clonaltype")])
612
613 #write files for every coincidence group of >1
614 samples = unique(clonalFreq$Sample)
615 for(sample in samples){
616 clonalFreqSample = clonalFreq[clonalFreq$Sample == sample,]
617 if(max(clonalFreqSample$Type) > 1){
618 for(i in 2:max(clonalFreqSample$Type)){
619 clonalFreqSampleType = clonalFreqSample[clonalFreqSample$Type == i,]
620 clonalityFrame.sub = clonalityFrame[clonalityFrame$clonaltype %in% clonalFreqSampleType$clonaltype,]
621 clonalityFrame.sub = clonalityFrame.sub[order(clonalityFrame.sub$clonaltype),]
622 write.table(clonalityFrame.sub, file=paste("coincidences_", sample, "_", i, ".txt", sep=""), sep="\t",quote=F,row.names=F,col.names=T)
623 }
624 }
625 }
626
627 clonalFreqCount = data.frame(data.table(clonalFreq)[, list(Count=.N), by=c("Sample", "Type")])
628 clonalFreqCount$realCount = clonalFreqCount$Type * clonalFreqCount$Count
629 clonalSum = data.frame(data.table(clonalFreqCount)[, list(Reads=sum(realCount)), by=c("Sample")])
630 clonalFreqCount = merge(clonalFreqCount, clonalSum, by.x="Sample", by.y="Sample")
631
632 ct = c('Type\tWeight\n2\t1\n3\t3\n4\t6\n5\t10\n6\t15')
633 tcct = textConnection(ct)
634 CT = read.table(tcct, sep="\t", header=TRUE)
635 close(tcct)
636 clonalFreqCount = merge(clonalFreqCount, CT, by.x="Type", by.y="Type", all.x=T)
637 clonalFreqCount$WeightedCount = clonalFreqCount$Count * clonalFreqCount$Weight
638
639 ReplicateReads = data.frame(data.table(clonalityFrame)[, list(Type=.N), by=c("Sample", "Replicate", "clonaltype")])
640 ReplicateReads = data.frame(data.table(ReplicateReads)[, list(Reads=.N), by=c("Sample", "Replicate")])
641 clonalFreqCount$Reads = as.numeric(clonalFreqCount$Reads)
642 ReplicateReads$Reads = as.numeric(ReplicateReads$Reads)
643 ReplicateReads$squared = as.numeric(ReplicateReads$Reads * ReplicateReads$Reads)
644
645 ReplicatePrint <- function(dat){
646 write.table(dat[-1], paste("ReplicateReads_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
647 }
648
649 ReplicateSplit = split(ReplicateReads, f=ReplicateReads[,"Sample"])
650 lapply(ReplicateSplit, FUN=ReplicatePrint)
651
652 ReplicateReads = data.frame(data.table(ReplicateReads)[, list(ReadsSum=sum(as.numeric(Reads)), ReadsSquaredSum=sum(as.numeric(squared))), by=c("Sample")])
653 clonalFreqCount = merge(clonalFreqCount, ReplicateReads, by.x="Sample", by.y="Sample", all.x=T)
654
655 ReplicateSumPrint <- function(dat){
656 write.table(dat[-1], paste("ReplicateSumReads_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
657 }
658
659 ReplicateSumSplit = split(ReplicateReads, f=ReplicateReads[,"Sample"])
660 lapply(ReplicateSumSplit, FUN=ReplicateSumPrint)
661
662 clonalFreqCountSum = data.frame(data.table(clonalFreqCount)[, list(Numerator=sum(WeightedCount, na.rm=T)), by=c("Sample")])
663 clonalFreqCount = merge(clonalFreqCount, clonalFreqCountSum, by.x="Sample", by.y="Sample", all.x=T)
664 clonalFreqCount$ReadsSum = as.numeric(clonalFreqCount$ReadsSum) #prevent integer overflow
665 clonalFreqCount$Denominator = (((clonalFreqCount$ReadsSum * clonalFreqCount$ReadsSum) - clonalFreqCount$ReadsSquaredSum) / 2)
666 clonalFreqCount$Result = (clonalFreqCount$Numerator + 1) / (clonalFreqCount$Denominator + 1)
667
668 ClonalityScorePrint <- function(dat){
669 write.table(dat$Result, paste("ClonalityScore_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
670 }
671
672 clonalityScore = clonalFreqCount[c("Sample", "Result")]
673 clonalityScore = unique(clonalityScore)
674
675 clonalityScoreSplit = split(clonalityScore, f=clonalityScore[,"Sample"])
676 lapply(clonalityScoreSplit, FUN=ClonalityScorePrint)
677
678 clonalityOverview = clonalFreqCount[c("Sample", "Type", "Count", "Weight", "WeightedCount")]
679
680
681
682 ClonalityOverviewPrint <- function(dat){
683 dat = dat[order(dat[,2]),]
684 write.table(dat[-1], paste("ClonalityOverView_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
685 }
686
687 clonalityOverviewSplit = split(clonalityOverview, f=clonalityOverview$Sample)
688 lapply(clonalityOverviewSplit, FUN=ClonalityOverviewPrint)
689 }
690 }
691
692 bak = PRODF
693
694 imgtcolumns = c("X3V.REGION.trimmed.nt.nb","P3V.nt.nb", "N1.REGION.nt.nb", "P5D.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "P3D.nt.nb", "N2.REGION.nt.nb", "P5J.nt.nb", "X5J.REGION.trimmed.nt.nb", "X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb")
695 if(all(imgtcolumns %in% colnames(inputdata)))
696 {
697 print("found IMGT columns, running junction analysis")
698
699 if(locus %in% c("IGK","IGL", "TRA", "TRG")){
700 print("VJ recombination, no filtering on absent D")
701 } else {
702 print("VDJ recombination, using N column for junction analysis")
703 fltr = nchar(PRODF$Top.D.Gene) < 4
704 print(paste("Removing", sum(fltr), "sequences without a identified D"))
705 PRODF = PRODF[!fltr,]
706 }
707
708
709 #ensure certain columns are in the data (files generated with older versions of IMGT Loader)
710 col.checks = c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb")
711 for(col.check in col.checks){
712 if(!(col.check %in% names(PRODF))){
713 print(paste(col.check, "not found adding new column"))
714 if(nrow(PRODF) > 0){ #because R is anoying...
715 PRODF[,col.check] = 0
716 } else {
717 PRODF = cbind(PRODF, data.frame(N3.REGION.nt.nb=numeric(0), N4.REGION.nt.nb=numeric(0)))
718 }
719 if(nrow(UNPROD) > 0){
720 UNPROD[,col.check] = 0
721 } else {
722 UNPROD = cbind(UNPROD, data.frame(N3.REGION.nt.nb=numeric(0), N4.REGION.nt.nb=numeric(0)))
723 }
724 }
725 }
726
727 num_median = function(x, na.rm=T) { as.numeric(median(x, na.rm=na.rm)) }
728
729 newData = data.frame(data.table(PRODF)[,list(unique=.N,
730 VH.DEL=mean(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
731 P1=mean(.SD$P3V.nt.nb, na.rm=T),
732 N1=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
733 P2=mean(.SD$P5D.nt.nb, na.rm=T),
734 DEL.DH=mean(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
735 DH.DEL=mean(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
736 P3=mean(.SD$P3D.nt.nb, na.rm=T),
737 N2=mean(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
738 P4=mean(.SD$P5J.nt.nb, na.rm=T),
739 DEL.JH=mean(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
740 Total.Del=mean(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
741 Total.N=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
742 Total.P=mean(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T)),
743 Median.CDR3.l=as.double(median(.SD$CDR3.Length))),
744 by=c("Sample")])
745 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
746 write.table(newData, "junctionAnalysisProd_mean.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
747
748 newData = data.frame(data.table(PRODF)[,list(unique=.N,
749 VH.DEL=num_median(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
750 P1=num_median(.SD$P3V.nt.nb, na.rm=T),
751 N1=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
752 P2=num_median(.SD$P5D.nt.nb, na.rm=T),
753 DEL.DH=num_median(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
754 DH.DEL=num_median(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
755 P3=num_median(.SD$P3D.nt.nb, na.rm=T),
756 N2=num_median(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
757 P4=num_median(.SD$P5J.nt.nb, na.rm=T),
758 DEL.JH=num_median(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
759 Total.Del=num_median(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
760 Total.N=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
761 Total.P=num_median(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T)),
762 Median.CDR3.l=as.double(median(.SD$CDR3.Length))),
763 by=c("Sample")])
764 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
765 write.table(newData, "junctionAnalysisProd_median.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
766
767 newData = data.frame(data.table(UNPROD)[,list(unique=.N,
768 VH.DEL=mean(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
769 P1=mean(.SD$P3V.nt.nb, na.rm=T),
770 N1=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
771 P2=mean(.SD$P5D.nt.nb, na.rm=T),
772 DEL.DH=mean(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
773 DH.DEL=mean(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
774 P3=mean(.SD$P3D.nt.nb, na.rm=T),
775 N2=mean(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
776 P4=mean(.SD$P5J.nt.nb, na.rm=T),
777 DEL.JH=mean(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
778 Total.Del=mean(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
779 Total.N=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
780 Total.P=mean(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T)),
781 Median.CDR3.l=as.double(median(.SD$CDR3.Length))),
782 by=c("Sample")])
783 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
784 write.table(newData, "junctionAnalysisUnProd_mean.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
785
786 newData = data.frame(data.table(UNPROD)[,list(unique=.N,
787 VH.DEL=num_median(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
788 P1=num_median(.SD$P3V.nt.nb, na.rm=T),
789 N1=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
790 P2=num_median(.SD$P5D.nt.nb, na.rm=T),
791 DEL.DH=num_median(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
792 DH.DEL=num_median(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
793 P3=num_median(.SD$P3D.nt.nb, na.rm=T),
794 N2=num_median(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
795 P4=num_median(.SD$P5J.nt.nb, na.rm=T),
796 DEL.JH=num_median(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
797 Total.Del=num_median(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
798 Total.N=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
799 Total.P=num_median(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T)),
800 Median.CDR3.l=as.double(median(.SD$CDR3.Length))),
801 by=c("Sample")])
802
803 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
804 write.table(newData, "junctionAnalysisUnProd_median.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
805 }
806
807 PRODF = bak
808
809
810 # ---------------------- D reading frame ----------------------
811
812 D.REGION.reading.frame = PRODF[,c("Sample", "D.REGION.reading.frame")]
813
814 chck = is.na(D.REGION.reading.frame$D.REGION.reading.frame)
815 if(any(chck)){
816 D.REGION.reading.frame[chck,"D.REGION.reading.frame"] = "No D"
817 }
818
819 D.REGION.reading.frame = data.frame(data.table(D.REGION.reading.frame)[, list(Freq=.N), by=c("Sample", "D.REGION.reading.frame")])
820
821 write.table(D.REGION.reading.frame, "DReadingFrame.csv" , sep="\t",quote=F,row.names=F,col.names=T)
822
823 D.REGION.reading.frame = ggplot(D.REGION.reading.frame)
824 D.REGION.reading.frame = D.REGION.reading.frame + geom_bar(aes( x = D.REGION.reading.frame, y = Freq, fill=Sample), stat='identity', position='dodge' ) + ggtitle("D reading frame") + xlab("Frequency") + ylab("Frame")
825 D.REGION.reading.frame = D.REGION.reading.frame + scale_fill_manual(values=sample.colors)
826 D.REGION.reading.frame = D.REGION.reading.frame + theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
827
828 png("DReadingFrame.png")
829 D.REGION.reading.frame
830 dev.off()
831
832
833
834
835 # ---------------------- AA composition in CDR3 ----------------------
836
837 AACDR3 = PRODF[,c("Sample", "CDR3.Seq")]
838
839 TotalPerSample = data.frame(data.table(AACDR3)[, list(total=sum(nchar(as.character(.SD$CDR3.Seq)))), by=Sample])
840
841 AAfreq = list()
842
843 for(i in 1:nrow(TotalPerSample)){
844 sample = TotalPerSample$Sample[i]
845 AAfreq[[i]] = data.frame(table(unlist(strsplit(as.character(AACDR3[AACDR3$Sample == sample,c("CDR3.Seq")]), ""))))
846 AAfreq[[i]]$Sample = sample
847 }
848
849 AAfreq = ldply(AAfreq, data.frame)
850 AAfreq = merge(AAfreq, TotalPerSample, by="Sample", all.x = T)
851 AAfreq$freq_perc = as.numeric(AAfreq$Freq / AAfreq$total * 100)
852
853
854 AAorder = read.table(sep="\t", header=TRUE, text="order.aa\tAA\n1\tR\n2\tK\n3\tN\n4\tD\n5\tQ\n6\tE\n7\tH\n8\tP\n9\tY\n10\tW\n11\tS\n12\tT\n13\tG\n14\tA\n15\tM\n16\tC\n17\tF\n18\tL\n19\tV\n20\tI")
855 AAfreq = merge(AAfreq, AAorder, by.x='Var1', by.y='AA', all.x=TRUE)
856
857 AAfreq = AAfreq[!is.na(AAfreq$order.aa),]
858
859 AAfreqplot = ggplot(AAfreq)
860 AAfreqplot = AAfreqplot + geom_bar(aes( x=factor(reorder(Var1, order.aa)), y = freq_perc, fill = Sample), stat='identity', position='dodge' )
861 AAfreqplot = AAfreqplot + annotate("rect", xmin = 0.5, xmax = 2.5, ymin = 0, ymax = Inf, fill = "red", alpha = 0.2)
862 AAfreqplot = AAfreqplot + annotate("rect", xmin = 3.5, xmax = 4.5, ymin = 0, ymax = Inf, fill = "blue", alpha = 0.2)
863 AAfreqplot = AAfreqplot + annotate("rect", xmin = 5.5, xmax = 6.5, ymin = 0, ymax = Inf, fill = "blue", alpha = 0.2)
864 AAfreqplot = AAfreqplot + annotate("rect", xmin = 6.5, xmax = 7.5, ymin = 0, ymax = Inf, fill = "red", alpha = 0.2)
865 AAfreqplot = AAfreqplot + ggtitle("Amino Acid Composition in the CDR3") + xlab("Amino Acid, from Hydrophilic (left) to Hydrophobic (right)") + ylab("Percentage") + scale_fill_manual(values=sample.colors)
866 AAfreqplot = AAfreqplot + theme(panel.background = element_rect(fill = "white", colour="black"),text = element_text(size=15, colour="black"), axis.text.x = element_text(angle = 45, hjust = 1), panel.grid.major.y = element_line(colour = "black"), panel.grid.major.x = element_blank())
867
868 png("AAComposition.png",width = 1280, height = 720)
869 AAfreqplot
870 dev.off()
871 write.table(AAfreq, "AAComposition.csv" , sep=",",quote=F,na="-",row.names=F,col.names=T)
872
873 # ---------------------- AA median CDR3 length ----------------------
874
875 median.aa.l = data.frame(data.table(PRODF)[, list(median=as.double(median(.SD$CDR3.Length))), by=c("Sample")])
876 write.table(median.aa.l, "AAMedianBySample.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
877