Mercurial > repos > davidvanzessen > argalaxy_tools

comparison report_clonality/RScript.r @ 0:afe85eb6572e draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Mon, 29 Aug 2016 05:41:20 -0400
parents
children 90a05ff900db
comparison
equal deleted inserted replaced
-1:000000000000 0:afe85eb6572e
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="")
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 prod.sample.count = data.frame(data.table(PRODF)[, list(Productive=.N), by=c("Sample")])
96 prod.rep.count = data.frame(data.table(PRODF)[, list(Productive=.N), by=c("Sample", "Replicate")])
97
98 unprod.sample.count = data.frame(data.table(UNPROD)[, list(Unproductive=.N), by=c("Sample")])
99 unprod.rep.count = data.frame(data.table(UNPROD)[, list(Unproductive=.N), by=c("Sample", "Replicate")])
100
101 clonalityFrame = PRODF
102
103 #remove duplicates based on the clonaltype
104 if(clonaltype != "none"){
105 clonaltype = paste(clonaltype, ",Sample", sep="") #add sample column to clonaltype, unique within samples
106 PRODF$clonaltype = do.call(paste, c(PRODF[unlist(strsplit(clonaltype, ","))], sep = ":"))
107 PRODF = PRODF[!duplicated(PRODF$clonaltype), ]
108
109 UNPROD$clonaltype = do.call(paste, c(UNPROD[unlist(strsplit(clonaltype, ","))], sep = ":"))
110 UNPROD = UNPROD[!duplicated(UNPROD$clonaltype), ]
111
112 #again for clonalityFrame but with sample+replicate
113 clonalityFrame$clonaltype = do.call(paste, c(clonalityFrame[unlist(strsplit(clonaltype, ","))], sep = ":"))
114 clonalityFrame$clonality_clonaltype = do.call(paste, c(clonalityFrame[unlist(strsplit(paste(clonaltype, ",Replicate", sep=""), ","))], sep = ":"))
115 clonalityFrame = clonalityFrame[!duplicated(clonalityFrame$clonality_clonaltype), ]
116 }
117
118 prod.unique.sample.count = data.frame(data.table(PRODF)[, list(Productive_unique=.N), by=c("Sample")])
119 prod.unique.rep.count = data.frame(data.table(PRODF)[, list(Productive_unique=.N), by=c("Sample", "Replicate")])
120
121 unprod.unique.sample.count = data.frame(data.table(UNPROD)[, list(Unproductive_unique=.N), by=c("Sample")])
122 unprod.unique.rep.count = data.frame(data.table(UNPROD)[, list(Unproductive_unique=.N), by=c("Sample", "Replicate")])
123
124 PRODF$freq = 1
125
126 if(any(grepl(pattern="_", x=PRODF$ID))){ #the frequency can be stored in the ID with the pattern ".*_freq_.*"
127 PRODF$freq = gsub("^[0-9]+_", "", PRODF$ID)
128 PRODF$freq = gsub("_.*", "", PRODF$freq)
129 PRODF$freq = as.numeric(PRODF$freq)
130 if(any(is.na(PRODF$freq))){ #if there was an "_" in the ID, but not the frequency, go back to frequency of 1 for every sequence
131 PRODF$freq = 1
132 }
133 }
134
135
136
137 #write the complete dataset that is left over, will be the input if 'none' for clonaltype and 'no' for filterproductive
138 write.table(PRODF, "allUnique.txt", sep="\t",quote=F,row.names=F,col.names=T)
139 write.table(PRODF, "allUnique.csv", sep=",",quote=F,row.names=F,col.names=T)
140 write.table(UNPROD, "allUnproductive.csv", sep=",",quote=F,row.names=F,col.names=T)
141
142 #write the samples to a file
143 sampleFile <- file("samples.txt")
144 un = unique(inputdata$Sample)
145 un = paste(un, sep="\n")
146 writeLines(un, sampleFile)
147 close(sampleFile)
148
149 # ---------------------- Counting the productive/unproductive and unique sequences ----------------------
150
151 print("Report Clonality - counting productive/unproductive/unique")
152
153 #create the table on the overview page with the productive/unique counts per sample/replicate
154 #first for sample
155 sample.count = merge(input.sample.count, prod.sample.count, by="Sample", all.x=T)
156 sample.count$perc_prod = round(sample.count$Productive / sample.count$All * 100)
157 sample.count = merge(sample.count, prod.unique.sample.count, by="Sample", all.x=T)
158 sample.count$perc_prod_un = round(sample.count$Productive_unique / sample.count$All * 100)
159
160 sample.count = merge(sample.count , unprod.sample.count, by="Sample", all.x=T)
161 sample.count$perc_unprod = round(sample.count$Unproductive / sample.count$All * 100)
162 sample.count = merge(sample.count, unprod.unique.sample.count, by="Sample", all.x=T)
163 sample.count$perc_unprod_un = round(sample.count$Unproductive_unique / sample.count$All * 100)
164
165 #then sample/replicate
166 rep.count = merge(input.rep.count, prod.rep.count, by=c("Sample", "Replicate"), all.x=T)
167 rep.count$perc_prod = round(rep.count$Productive / rep.count$All * 100)
168 rep.count = merge(rep.count, prod.unique.rep.count, by=c("Sample", "Replicate"), all.x=T)
169 rep.count$perc_prod_un = round(rep.count$Productive_unique / rep.count$All * 100)
170
171 rep.count = merge(rep.count, unprod.rep.count, by=c("Sample", "Replicate"), all.x=T)
172 rep.count$perc_unprod = round(rep.count$Unproductive / rep.count$All * 100)
173 rep.count = merge(rep.count, unprod.unique.rep.count, by=c("Sample", "Replicate"), all.x=T)
174 rep.count$perc_unprod_un = round(rep.count$Unproductive_unique / rep.count$All * 100)
175
176 rep.count$Sample = paste(rep.count$Sample, rep.count$Replicate, sep="_")
177 rep.count = rep.count[,names(rep.count) != "Replicate"]
178
179 count = rbind(sample.count, rep.count)
180
181
182
183 write.table(x=count, file="productive_counting.txt", sep=",",quote=F,row.names=F,col.names=F)
184
185 # ---------------------- V+J+CDR3 sequence count ----------------------
186
187 VJCDR3.count = data.frame(table(clonalityFrame$Top.V.Gene, clonalityFrame$Top.J.Gene, clonalityFrame$CDR3.Seq.DNA))
188 names(VJCDR3.count) = c("Top.V.Gene", "Top.J.Gene", "CDR3.Seq.DNA", "Count")
189
190 VJCDR3.count = VJCDR3.count[VJCDR3.count$Count > 0,]
191 VJCDR3.count = VJCDR3.count[order(-VJCDR3.count$Count),]
192
193 write.table(x=VJCDR3.count, file="VJCDR3_count.txt", sep="\t",quote=F,row.names=F,col.names=T)
194
195 # ---------------------- Frequency calculation for V, D and J ----------------------
196
197 print("Report Clonality - frequency calculation V, D and J")
198
199 PRODFV = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.V.Gene")])
200 Total = ddply(PRODFV, .(Sample), function(x) data.frame(Total = sum(x$Length)))
201 PRODFV = merge(PRODFV, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
202 PRODFV = ddply(PRODFV, c("Sample", "Top.V.Gene"), summarise, relFreq= (Length*100 / Total))
203
204 PRODFD = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.D.Gene")])
205 Total = ddply(PRODFD, .(Sample), function(x) data.frame(Total = sum(x$Length)))
206 PRODFD = merge(PRODFD, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
207 PRODFD = ddply(PRODFD, c("Sample", "Top.D.Gene"), summarise, relFreq= (Length*100 / Total))
208
209 PRODFJ = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.J.Gene")])
210 Total = ddply(PRODFJ, .(Sample), function(x) data.frame(Total = sum(x$Length)))
211 PRODFJ = merge(PRODFJ, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
212 PRODFJ = ddply(PRODFJ, c("Sample", "Top.J.Gene"), summarise, relFreq= (Length*100 / Total))
213
214 # ---------------------- Setting up the gene names for the different species/loci ----------------------
215
216 print("Report Clonality - getting genes for species/loci")
217
218 Vchain = ""
219 Dchain = ""
220 Jchain = ""
221
222 if(species == "custom"){
223 print("Custom genes: ")
224 splt = unlist(strsplit(locus, ";"))
225 print(paste("V:", splt[1]))
226 print(paste("D:", splt[2]))
227 print(paste("J:", splt[3]))
228
229 Vchain = unlist(strsplit(splt[1], ","))
230 Vchain = data.frame(v.name = Vchain, chr.orderV = 1:length(Vchain))
231
232 Dchain = unlist(strsplit(splt[2], ","))
233 if(length(Dchain) > 0){
234 Dchain = data.frame(v.name = Dchain, chr.orderD = 1:length(Dchain))
235 } else {
236 Dchain = data.frame(v.name = character(0), chr.orderD = numeric(0))
237 }
238
239 Jchain = unlist(strsplit(splt[3], ","))
240 Jchain = data.frame(v.name = Jchain, chr.orderJ = 1:length(Jchain))
241
242 } else {
243 genes = read.table("genes.txt", sep="\t", header=TRUE, fill=T, comment.char="")
244
245 Vchain = genes[grepl(species, genes$Species) & genes$locus == locus & genes$region == "V",c("IMGT.GENE.DB", "chr.order")]
246 colnames(Vchain) = c("v.name", "chr.orderV")
247 Dchain = genes[grepl(species, genes$Species) & genes$locus == locus & genes$region == "D",c("IMGT.GENE.DB", "chr.order")]
248 colnames(Dchain) = c("v.name", "chr.orderD")
249 Jchain = genes[grepl(species, genes$Species) & genes$locus == locus & genes$region == "J",c("IMGT.GENE.DB", "chr.order")]
250 colnames(Jchain) = c("v.name", "chr.orderJ")
251 }
252 useD = TRUE
253 if(nrow(Dchain) == 0){
254 useD = FALSE
255 cat("No D Genes in this species/locus")
256 }
257 print(paste(nrow(Vchain), "genes in V"))
258 print(paste(nrow(Dchain), "genes in D"))
259 print(paste(nrow(Jchain), "genes in J"))
260
261 # ---------------------- merge with the frequency count ----------------------
262
263 PRODFV = merge(PRODFV, Vchain, by.x='Top.V.Gene', by.y='v.name', all.x=TRUE)
264
265 PRODFD = merge(PRODFD, Dchain, by.x='Top.D.Gene', by.y='v.name', all.x=TRUE)
266
267 PRODFJ = merge(PRODFJ, Jchain, by.x='Top.J.Gene', by.y='v.name', all.x=TRUE)
268
269 # ---------------------- Create the V, D and J frequency plots and write the data.frame for every plot to a file ----------------------
270
271 print("Report Clonality - V, D and J frequency plots")
272
273 pV = ggplot(PRODFV)
274 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))
275 pV = pV + xlab("Summary of V gene") + ylab("Frequency") + ggtitle("Relative frequency of V gene usage")
276 write.table(x=PRODFV, file="VFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
277
278 png("VPlot.png",width = 1280, height = 720)
279 pV
280 dev.off();
281
282 if(useD){
283 pD = ggplot(PRODFD)
284 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))
285 pD = pD + xlab("Summary of D gene") + ylab("Frequency") + ggtitle("Relative frequency of D gene usage")
286 write.table(x=PRODFD, file="DFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
287
288 png("DPlot.png",width = 800, height = 600)
289 print(pD)
290 dev.off();
291 }
292
293 pJ = ggplot(PRODFJ)
294 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))
295 pJ = pJ + xlab("Summary of J gene") + ylab("Frequency") + ggtitle("Relative frequency of J gene usage")
296 write.table(x=PRODFJ, file="JFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
297
298 png("JPlot.png",width = 800, height = 600)
299 pJ
300 dev.off();
301
302 pJ = ggplot(PRODFJ)
303 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))
304 pJ = pJ + xlab("Summary of J gene") + ylab("Frequency") + ggtitle("Relative frequency of J gene usage")
305 write.table(x=PRODFJ, file="JFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
306
307 png("JPlot.png",width = 800, height = 600)
308 pJ
309 dev.off();
310
311 # ---------------------- Now the frequency plots of the V, D and J families ----------------------
312
313 print("Report Clonality - V, D and J family plots")
314
315 VGenes = PRODF[,c("Sample", "Top.V.Gene")]
316 VGenes$Top.V.Gene = gsub("-.*", "", VGenes$Top.V.Gene)
317 VGenes = data.frame(data.table(VGenes)[, list(Count=.N), by=c("Sample", "Top.V.Gene")])
318 TotalPerSample = data.frame(data.table(VGenes)[, list(total=sum(.SD$Count)), by=Sample])
319 VGenes = merge(VGenes, TotalPerSample, by="Sample")
320 VGenes$Frequency = VGenes$Count * 100 / VGenes$total
321 VPlot = ggplot(VGenes)
322 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)) +
323 ggtitle("Distribution of V gene families") +
324 ylab("Percentage of sequences")
325 png("VFPlot.png")
326 VPlot
327 dev.off();
328 write.table(x=VGenes, file="VFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
329
330 if(useD){
331 DGenes = PRODF[,c("Sample", "Top.D.Gene")]
332 DGenes$Top.D.Gene = gsub("-.*", "", DGenes$Top.D.Gene)
333 DGenes = data.frame(data.table(DGenes)[, list(Count=.N), by=c("Sample", "Top.D.Gene")])
334 TotalPerSample = data.frame(data.table(DGenes)[, list(total=sum(.SD$Count)), by=Sample])
335 DGenes = merge(DGenes, TotalPerSample, by="Sample")
336 DGenes$Frequency = DGenes$Count * 100 / DGenes$total
337 DPlot = ggplot(DGenes)
338 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)) +
339 ggtitle("Distribution of D gene families") +
340 ylab("Percentage of sequences")
341 png("DFPlot.png")
342 print(DPlot)
343 dev.off();
344 write.table(x=DGenes, file="DFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
345 }
346
347 JGenes = PRODF[,c("Sample", "Top.J.Gene")]
348 JGenes$Top.J.Gene = gsub("-.*", "", JGenes$Top.J.Gene)
349 JGenes = data.frame(data.table(JGenes)[, list(Count=.N), by=c("Sample", "Top.J.Gene")])
350 TotalPerSample = data.frame(data.table(JGenes)[, list(total=sum(.SD$Count)), by=Sample])
351 JGenes = merge(JGenes, TotalPerSample, by="Sample")
352 JGenes$Frequency = JGenes$Count * 100 / JGenes$total
353 JPlot = ggplot(JGenes)
354 JPlot = JPlot + geom_bar(aes( x = Top.J.Gene, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
355 ggtitle("Distribution of J gene families") +
356 ylab("Percentage of sequences")
357 png("JFPlot.png")
358 JPlot
359 dev.off();
360 write.table(x=JGenes, file="JFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
361
362 # ---------------------- Plotting the cdr3 length ----------------------
363
364 print("Report Clonality - CDR3 length plot")
365
366 CDR3Length = data.frame(data.table(PRODF)[, list(Count=.N), by=c("Sample", "CDR3.Length.DNA")])
367 TotalPerSample = data.frame(data.table(CDR3Length)[, list(total=sum(.SD$Count)), by=Sample])
368 CDR3Length = merge(CDR3Length, TotalPerSample, by="Sample")
369 CDR3Length$Frequency = CDR3Length$Count * 100 / CDR3Length$total
370 CDR3LengthPlot = ggplot(CDR3Length)
371 CDR3LengthPlot = CDR3LengthPlot + geom_bar(aes( x = CDR3.Length.DNA, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
372 ggtitle("Length distribution of CDR3") +
373 xlab("CDR3 Length") +
374 ylab("Percentage of sequences")
375 png("CDR3LengthPlot.png",width = 1280, height = 720)
376 CDR3LengthPlot
377 dev.off()
378 write.table(x=CDR3Length, file="CDR3LengthPlot.csv", sep=",",quote=F,row.names=F,col.names=T)
379
380 # ---------------------- Plot the heatmaps ----------------------
381
382 #get the reverse order for the V and D genes
383 revVchain = Vchain
384 revDchain = Dchain
385 revVchain$chr.orderV = rev(revVchain$chr.orderV)
386 revDchain$chr.orderD = rev(revDchain$chr.orderD)
387
388 if(useD){
389 print("Report Clonality - Heatmaps VD")
390 plotVD <- function(dat){
391 if(length(dat[,1]) == 0){
392 return()
393 }
394
395 img = ggplot() +
396 geom_tile(data=dat, aes(x=factor(reorder(Top.D.Gene, chr.orderD)), y=factor(reorder(Top.V.Gene, chr.orderV)), fill=relLength)) +
397 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
398 scale_fill_gradient(low="gold", high="blue", na.value="white") +
399 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
400 xlab("D genes") +
401 ylab("V Genes")
402
403 png(paste("HeatmapVD_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Dchain$v.name)), height=100+(15*length(Vchain$v.name)))
404 print(img)
405 dev.off()
406 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)
407 }
408
409 VandDCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.V.Gene", "Top.D.Gene", "Sample")])
410
411 VandDCount$l = log(VandDCount$Length)
412 maxVD = data.frame(data.table(VandDCount)[, list(max=max(l)), by=c("Sample")])
413 VandDCount = merge(VandDCount, maxVD, by.x="Sample", by.y="Sample", all.x=T)
414 VandDCount$relLength = VandDCount$l / VandDCount$max
415
416 cartegianProductVD = expand.grid(Top.V.Gene = Vchain$v.name, Top.D.Gene = Dchain$v.name)
417
418 completeVD = merge(VandDCount, cartegianProductVD, by.x=c("Top.V.Gene", "Top.D.Gene"), by.y=c("Top.V.Gene", "Top.D.Gene"), all=TRUE)
419
420 completeVD = merge(completeVD, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE)
421
422 completeVD = merge(completeVD, Dchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE)
423
424 fltr = is.nan(completeVD$relLength)
425 if(all(fltr)){
426 completeVD[fltr,"relLength"] = 0
427 }
428
429 VDList = split(completeVD, f=completeVD[,"Sample"])
430 lapply(VDList, FUN=plotVD)
431 }
432
433 print("Report Clonality - Heatmaps VJ")
434
435 plotVJ <- function(dat){
436 if(length(dat[,1]) == 0){
437 return()
438 }
439 cat(paste(unique(dat[3])[1,1]))
440 img = ggplot() +
441 geom_tile(data=dat, aes(x=factor(reorder(Top.J.Gene, chr.orderJ)), y=factor(reorder(Top.V.Gene, chr.orderV)), fill=relLength)) +
442 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
443 scale_fill_gradient(low="gold", high="blue", na.value="white") +
444 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
445 xlab("J genes") +
446 ylab("V Genes")
447
448 png(paste("HeatmapVJ_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Jchain$v.name)), height=100+(15*length(Vchain$v.name)))
449 print(img)
450 dev.off()
451 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)
452 }
453
454 VandJCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.V.Gene", "Top.J.Gene", "Sample")])
455
456 VandJCount$l = log(VandJCount$Length)
457 maxVJ = data.frame(data.table(VandJCount)[, list(max=max(l)), by=c("Sample")])
458 VandJCount = merge(VandJCount, maxVJ, by.x="Sample", by.y="Sample", all.x=T)
459 VandJCount$relLength = VandJCount$l / VandJCount$max
460
461 cartegianProductVJ = expand.grid(Top.V.Gene = Vchain$v.name, Top.J.Gene = Jchain$v.name)
462
463 completeVJ = merge(VandJCount, cartegianProductVJ, all.y=TRUE)
464 completeVJ = merge(completeVJ, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE)
465 completeVJ = merge(completeVJ, Jchain, by.x="Top.J.Gene", by.y="v.name", all.x=TRUE)
466
467 fltr = is.nan(completeVJ$relLength)
468 if(any(fltr)){
469 completeVJ[fltr,"relLength"] = 1
470 }
471
472 VJList = split(completeVJ, f=completeVJ[,"Sample"])
473 lapply(VJList, FUN=plotVJ)
474
475
476
477 if(useD){
478 print("Report Clonality - Heatmaps DJ")
479 plotDJ <- function(dat){
480 if(length(dat[,1]) == 0){
481 return()
482 }
483 img = ggplot() +
484 geom_tile(data=dat, aes(x=factor(reorder(Top.J.Gene, chr.orderJ)), y=factor(reorder(Top.D.Gene, chr.orderD)), fill=relLength)) +
485 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
486 scale_fill_gradient(low="gold", high="blue", na.value="white") +
487 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
488 xlab("J genes") +
489 ylab("D Genes")
490
491 png(paste("HeatmapDJ_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Jchain$v.name)), height=100+(15*length(Dchain$v.name)))
492 print(img)
493 dev.off()
494 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)
495 }
496
497
498 DandJCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.D.Gene", "Top.J.Gene", "Sample")])
499
500 DandJCount$l = log(DandJCount$Length)
501 maxDJ = data.frame(data.table(DandJCount)[, list(max=max(l)), by=c("Sample")])
502 DandJCount = merge(DandJCount, maxDJ, by.x="Sample", by.y="Sample", all.x=T)
503 DandJCount$relLength = DandJCount$l / DandJCount$max
504
505 cartegianProductDJ = expand.grid(Top.D.Gene = Dchain$v.name, Top.J.Gene = Jchain$v.name)
506
507 completeDJ = merge(DandJCount, cartegianProductDJ, all.y=TRUE)
508 completeDJ = merge(completeDJ, revDchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE)
509 completeDJ = merge(completeDJ, Jchain, by.x="Top.J.Gene", by.y="v.name", all.x=TRUE)
510
511 fltr = is.nan(completeDJ$relLength)
512 if(any(fltr)){
513 completeDJ[fltr, "relLength"] = 1
514 }
515
516 DJList = split(completeDJ, f=completeDJ[,"Sample"])
517 lapply(DJList, FUN=plotDJ)
518 }
519
520
521 # ---------------------- output tables for the circos plots ----------------------
522
523 print("Report Clonality - Circos data")
524
525 for(smpl in unique(PRODF$Sample)){
526 PRODF.sample = PRODF[PRODF$Sample == smpl,]
527
528 fltr = PRODF.sample$Top.V.Gene == ""
529 if(any(fltr, na.rm=T)){
530 PRODF.sample[fltr, "Top.V.Gene"] = "NA"
531 }
532
533 fltr = PRODF.sample$Top.D.Gene == ""
534 if(any(fltr, na.rm=T)){
535 PRODF.sample[fltr, "Top.D.Gene"] = "NA"
536 }
537
538 fltr = PRODF.sample$Top.J.Gene == ""
539 if(any(fltr, na.rm=T)){
540 PRODF.sample[fltr, "Top.J.Gene"] = "NA"
541 }
542
543 v.d = table(PRODF.sample$Top.V.Gene, PRODF.sample$Top.D.Gene)
544 v.j = table(PRODF.sample$Top.V.Gene, PRODF.sample$Top.J.Gene)
545 d.j = table(PRODF.sample$Top.D.Gene, PRODF.sample$Top.J.Gene)
546
547 write.table(v.d, file=paste(smpl, "_VD_circos.txt", sep=""), sep="\t", quote=F, row.names=T, col.names=NA)
548 write.table(v.j, file=paste(smpl, "_VJ_circos.txt", sep=""), sep="\t", quote=F, row.names=T, col.names=NA)
549 write.table(d.j, file=paste(smpl, "_DJ_circos.txt", sep=""), sep="\t", quote=F, row.names=T, col.names=NA)
550 }
551
552 # ---------------------- calculating the clonality score ----------------------
553
554 if("Replicate" %in% colnames(inputdata)) #can only calculate clonality score when replicate information is available
555 {
556 print("Report Clonality - Clonality")
557 write.table(clonalityFrame, "clonalityComplete.csv", sep=",",quote=F,row.names=F,col.names=T)
558 if(clonality_method == "boyd"){
559 samples = split(clonalityFrame, clonalityFrame$Sample, drop=T)
560
561 for (sample in samples){
562 res = data.frame(paste=character(0))
563 sample_id = unique(sample$Sample)[[1]]
564 for(replicate in unique(sample$Replicate)){
565 tmp = sample[sample$Replicate == replicate,]
566 clone_table = data.frame(table(tmp$clonaltype))
567 clone_col_name = paste("V", replicate, sep="")
568 colnames(clone_table) = c("paste", clone_col_name)
569 res = merge(res, clone_table, by="paste", all=T)
570 }
571
572 res[is.na(res)] = 0
573 infer.result = infer.clonality(as.matrix(res[,2:ncol(res)]))
574
575 print(infer.result)
576
577 write.table(data.table(infer.result[[12]]), file=paste("lymphclon_clonality_", sample_id, ".csv", sep=""), sep=",",quote=F,row.names=F,col.names=F)
578
579 res$type = rowSums(res[,2:ncol(res)])
580
581 coincidence.table = data.frame(table(res$type))
582 colnames(coincidence.table) = c("Coincidence Type", "Raw Coincidence Freq")
583 write.table(coincidence.table, file=paste("lymphclon_coincidences_", sample_id, ".csv", sep=""), sep=",",quote=F,row.names=F,col.names=T)
584 }
585 } else {
586 clonalFreq = data.frame(data.table(clonalityFrame)[, list(Type=.N), by=c("Sample", "clonaltype")])
587
588 #write files for every coincidence group of >1
589 samples = unique(clonalFreq$Sample)
590 for(sample in samples){
591 clonalFreqSample = clonalFreq[clonalFreq$Sample == sample,]
592 if(max(clonalFreqSample$Type) > 1){
593 for(i in 2:max(clonalFreqSample$Type)){
594 clonalFreqSampleType = clonalFreqSample[clonalFreqSample$Type == i,]
595 clonalityFrame.sub = clonalityFrame[clonalityFrame$clonaltype %in% clonalFreqSampleType$clonaltype,]
596 clonalityFrame.sub = clonalityFrame.sub[order(clonalityFrame.sub$clonaltype),]
597 write.table(clonalityFrame.sub, file=paste("coincidences_", sample, "_", i, ".txt", sep=""), sep="\t",quote=F,row.names=F,col.names=T)
598 }
599 }
600 }
601
602 clonalFreqCount = data.frame(data.table(clonalFreq)[, list(Count=.N), by=c("Sample", "Type")])
603 clonalFreqCount$realCount = clonalFreqCount$Type * clonalFreqCount$Count
604 clonalSum = data.frame(data.table(clonalFreqCount)[, list(Reads=sum(realCount)), by=c("Sample")])
605 clonalFreqCount = merge(clonalFreqCount, clonalSum, by.x="Sample", by.y="Sample")
606
607 ct = c('Type\tWeight\n2\t1\n3\t3\n4\t6\n5\t10\n6\t15')
608 tcct = textConnection(ct)
609 CT = read.table(tcct, sep="\t", header=TRUE)
610 close(tcct)
611 clonalFreqCount = merge(clonalFreqCount, CT, by.x="Type", by.y="Type", all.x=T)
612 clonalFreqCount$WeightedCount = clonalFreqCount$Count * clonalFreqCount$Weight
613
614 ReplicateReads = data.frame(data.table(clonalityFrame)[, list(Type=.N), by=c("Sample", "Replicate", "clonaltype")])
615 ReplicateReads = data.frame(data.table(ReplicateReads)[, list(Reads=.N), by=c("Sample", "Replicate")])
616 clonalFreqCount$Reads = as.numeric(clonalFreqCount$Reads)
617 ReplicateReads$Reads = as.numeric(ReplicateReads$Reads)
618 ReplicateReads$squared = as.numeric(ReplicateReads$Reads * ReplicateReads$Reads)
619
620 ReplicatePrint <- function(dat){
621 write.table(dat[-1], paste("ReplicateReads_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
622 }
623
624 ReplicateSplit = split(ReplicateReads, f=ReplicateReads[,"Sample"])
625 lapply(ReplicateSplit, FUN=ReplicatePrint)
626
627 ReplicateReads = data.frame(data.table(ReplicateReads)[, list(ReadsSum=sum(as.numeric(Reads)), ReadsSquaredSum=sum(as.numeric(squared))), by=c("Sample")])
628 clonalFreqCount = merge(clonalFreqCount, ReplicateReads, by.x="Sample", by.y="Sample", all.x=T)
629
630 ReplicateSumPrint <- function(dat){
631 write.table(dat[-1], paste("ReplicateSumReads_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
632 }
633
634 ReplicateSumSplit = split(ReplicateReads, f=ReplicateReads[,"Sample"])
635 lapply(ReplicateSumSplit, FUN=ReplicateSumPrint)
636
637 clonalFreqCountSum = data.frame(data.table(clonalFreqCount)[, list(Numerator=sum(WeightedCount, na.rm=T)), by=c("Sample")])
638 clonalFreqCount = merge(clonalFreqCount, clonalFreqCountSum, by.x="Sample", by.y="Sample", all.x=T)
639 clonalFreqCount$ReadsSum = as.numeric(clonalFreqCount$ReadsSum) #prevent integer overflow
640 clonalFreqCount$Denominator = (((clonalFreqCount$ReadsSum * clonalFreqCount$ReadsSum) - clonalFreqCount$ReadsSquaredSum) / 2)
641 clonalFreqCount$Result = (clonalFreqCount$Numerator + 1) / (clonalFreqCount$Denominator + 1)
642
643 ClonalityScorePrint <- function(dat){
644 write.table(dat$Result, paste("ClonalityScore_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
645 }
646
647 clonalityScore = clonalFreqCount[c("Sample", "Result")]
648 clonalityScore = unique(clonalityScore)
649
650 clonalityScoreSplit = split(clonalityScore, f=clonalityScore[,"Sample"])
651 lapply(clonalityScoreSplit, FUN=ClonalityScorePrint)
652
653 clonalityOverview = clonalFreqCount[c("Sample", "Type", "Count", "Weight", "WeightedCount")]
654
655
656
657 ClonalityOverviewPrint <- function(dat){
658 dat = dat[order(dat[,2]),]
659 write.table(dat[-1], paste("ClonalityOverView_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
660 }
661
662 clonalityOverviewSplit = split(clonalityOverview, f=clonalityOverview$Sample)
663 lapply(clonalityOverviewSplit, FUN=ClonalityOverviewPrint)
664 }
665 }
666
667 bak = PRODF
668
669 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")
670 if(all(imgtcolumns %in% colnames(inputdata)))
671 {
672 print("found IMGT columns, running junction analysis")
673
674 if(locus %in% c("IGK","IGL", "TRA", "TRG")){
675 print("VJ recombination, no filtering on absent D")
676 } else {
677 print("VDJ recombination, using N column for junction analysis")
678 fltr = nchar(PRODF$Top.D.Gene) < 4
679 print(paste("Removing", sum(fltr), "sequences without a identified D"))
680 PRODF = PRODF[!fltr,]
681 }
682
683
684 #ensure certain columns are in the data (files generated with older versions of IMGT Loader)
685 col.checks = c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb")
686 for(col.check in col.checks){
687 if(!(col.check %in% names(PRODF))){
688 print(paste(col.check, "not found adding new column"))
689 if(nrow(PRODF) > 0){ #because R is anoying...
690 PRODF[,col.check] = 0
691 } else {
692 PRODF = cbind(PRODF, data.frame(N3.REGION.nt.nb=numeric(0), N4.REGION.nt.nb=numeric(0)))
693 }
694 if(nrow(UNPROD) > 0){
695 UNPROD[,col.check] = 0
696 } else {
697 UNPROD = cbind(UNPROD, data.frame(N3.REGION.nt.nb=numeric(0), N4.REGION.nt.nb=numeric(0)))
698 }
699 }
700 }
701
702 num_median = function(x, na.rm=T) { as.numeric(median(x, na.rm=na.rm)) }
703
704 newData = data.frame(data.table(PRODF)[,list(unique=.N,
705 VH.DEL=mean(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
706 P1=mean(.SD$P3V.nt.nb, na.rm=T),
707 N1=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
708 P2=mean(.SD$P5D.nt.nb, na.rm=T),
709 DEL.DH=mean(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
710 DH.DEL=mean(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
711 P3=mean(.SD$P3D.nt.nb, na.rm=T),
712 N2=mean(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
713 P4=mean(.SD$P5J.nt.nb, na.rm=T),
714 DEL.JH=mean(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
715 Total.Del=mean(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
716 Total.N=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
717 Total.P=mean(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T))),
718 by=c("Sample")])
719 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
720 write.table(newData, "junctionAnalysisProd_mean.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
721
722 newData = data.frame(data.table(PRODF)[,list(unique=.N,
723 VH.DEL=num_median(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
724 P1=num_median(.SD$P3V.nt.nb, na.rm=T),
725 N1=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
726 P2=num_median(.SD$P5D.nt.nb, na.rm=T),
727 DEL.DH=num_median(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
728 DH.DEL=num_median(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
729 P3=num_median(.SD$P3D.nt.nb, na.rm=T),
730 N2=num_median(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
731 P4=num_median(.SD$P5J.nt.nb, na.rm=T),
732 DEL.JH=num_median(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
733 Total.Del=num_median(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
734 Total.N=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
735 Total.P=num_median(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T))),
736 by=c("Sample")])
737 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
738 write.table(newData, "junctionAnalysisProd_median.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
739
740 newData = data.frame(data.table(UNPROD)[,list(unique=.N,
741 VH.DEL=mean(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
742 P1=mean(.SD$P3V.nt.nb, na.rm=T),
743 N1=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
744 P2=mean(.SD$P5D.nt.nb, na.rm=T),
745 DEL.DH=mean(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
746 DH.DEL=mean(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
747 P3=mean(.SD$P3D.nt.nb, na.rm=T),
748 N2=mean(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
749 P4=mean(.SD$P5J.nt.nb, na.rm=T),
750 DEL.JH=mean(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
751 Total.Del=mean(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
752 Total.N=mean(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
753 Total.P=mean(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T))),
754 by=c("Sample")])
755 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
756 write.table(newData, "junctionAnalysisUnProd_mean.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
757
758 newData = data.frame(data.table(UNPROD)[,list(unique=.N,
759 VH.DEL=num_median(.SD$X3V.REGION.trimmed.nt.nb, na.rm=T),
760 P1=num_median(.SD$P3V.nt.nb, na.rm=T),
761 N1=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb"), with=F], na.rm=T)),
762 P2=num_median(.SD$P5D.nt.nb, na.rm=T),
763 DEL.DH=num_median(.SD$X5D.REGION.trimmed.nt.nb, na.rm=T),
764 DH.DEL=num_median(.SD$X3D.REGION.trimmed.nt.nb, na.rm=T),
765 P3=num_median(.SD$P3D.nt.nb, na.rm=T),
766 N2=num_median(rowSums(.SD[,c("N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
767 P4=num_median(.SD$P5J.nt.nb, na.rm=T),
768 DEL.JH=num_median(.SD$X5J.REGION.trimmed.nt.nb, na.rm=T),
769 Total.Del=num_median(rowSums(.SD[,c("X3V.REGION.trimmed.nt.nb", "X5D.REGION.trimmed.nt.nb", "X3D.REGION.trimmed.nt.nb", "X5J.REGION.trimmed.nt.nb"), with=F], na.rm=T)),
770 Total.N=num_median(rowSums(.SD[,c("N.REGION.nt.nb", "N1.REGION.nt.nb", "N2.REGION.nt.nb", "N3.REGION.nt.nb", "N4.REGION.nt.nb"), with=F], na.rm=T)),
771 Total.P=num_median(rowSums(.SD[,c("P3V.nt.nb", "P5D.nt.nb", "P3D.nt.nb", "P5J.nt.nb"), with=F], na.rm=T))),
772 by=c("Sample")])
773
774 newData[,sapply(newData, is.numeric)] = round(newData[,sapply(newData, is.numeric)],1)
775 write.table(newData, "junctionAnalysisUnProd_median.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
776 }
777
778 PRODF = bak
779
780
781 # ---------------------- D reading frame ----------------------
782
783 D.REGION.reading.frame = PRODF$D.REGION.reading.frame
784
785 D.REGION.reading.frame[is.na(D.REGION.reading.frame)] = "No D"
786
787 D.REGION.reading.frame = data.frame(table(D.REGION.reading.frame))
788
789 write.table(D.REGION.reading.frame, "DReadingFrame.csv" , sep="\t",quote=F,row.names=F,col.names=T)
790
791 D.REGION.reading.frame = ggplot(D.REGION.reading.frame)
792 D.REGION.reading.frame = D.REGION.reading.frame + geom_bar(aes( x = D.REGION.reading.frame, y = Freq), stat='identity', position='dodge' ) + ggtitle("D reading frame") + xlab("Frequency") + ylab("Frame")
793
794 png("DReadingFrame.png")
795 D.REGION.reading.frame
796 dev.off()
797
798
799
800
801 # ---------------------- AA composition in CDR3 ----------------------
802
803 AACDR3 = PRODF[,c("Sample", "CDR3.Seq")]
804
805 TotalPerSample = data.frame(data.table(AACDR3)[, list(total=sum(nchar(as.character(.SD$CDR3.Seq)))), by=Sample])
806
807 AAfreq = list()
808
809 for(i in 1:nrow(TotalPerSample)){
810 sample = TotalPerSample$Sample[i]
811 AAfreq[[i]] = data.frame(table(unlist(strsplit(as.character(AACDR3[AACDR3$Sample == sample,c("CDR3.Seq")]), ""))))
812 AAfreq[[i]]$Sample = sample
813 }
814
815 AAfreq = ldply(AAfreq, data.frame)
816 AAfreq = merge(AAfreq, TotalPerSample, by="Sample", all.x = T)
817 AAfreq$freq_perc = as.numeric(AAfreq$Freq / AAfreq$total * 100)
818
819
820 AAorder = read.table(sep="\t", header=TRUE, text="order.aa\tAA\n1\tR\n2\tK\n3\tN\n4\tD\n5\tQ\n6\tE\n7\tH\n8\tP\n9\tY\n10\tW\n11\tS\n12\tT\n13\tG\n14\tA\n15\tM\n16\tC\n17\tF\n18\tL\n19\tV\n20\tI")
821 AAfreq = merge(AAfreq, AAorder, by.x='Var1', by.y='AA', all.x=TRUE)
822
823 AAfreq = AAfreq[!is.na(AAfreq$order.aa),]
824
825 AAfreqplot = ggplot(AAfreq)
826 AAfreqplot = AAfreqplot + geom_bar(aes( x=factor(reorder(Var1, order.aa)), y = freq_perc, fill = Sample), stat='identity', position='dodge' )
827 AAfreqplot = AAfreqplot + annotate("rect", xmin = 0.5, xmax = 2.5, ymin = 0, ymax = Inf, fill = "red", alpha = 0.2)
828 AAfreqplot = AAfreqplot + annotate("rect", xmin = 3.5, xmax = 4.5, ymin = 0, ymax = Inf, fill = "blue", alpha = 0.2)
829 AAfreqplot = AAfreqplot + annotate("rect", xmin = 5.5, xmax = 6.5, ymin = 0, ymax = Inf, fill = "blue", alpha = 0.2)
830 AAfreqplot = AAfreqplot + annotate("rect", xmin = 6.5, xmax = 7.5, ymin = 0, ymax = Inf, fill = "red", alpha = 0.2)
831 AAfreqplot = AAfreqplot + ggtitle("Amino Acid Composition in the CDR3") + xlab("Amino Acid, from Hydrophilic (left) to Hydrophobic (right)") + ylab("Percentage")
832
833 png("AAComposition.png",width = 1280, height = 720)
834 AAfreqplot
835 dev.off()
836 write.table(AAfreq, "AAComposition.csv" , sep=",",quote=F,na="-",row.names=F,col.names=T)
837
838