|
53
|
1 library(data.table)
|
|
|
2 library(ggplot2)
|
|
|
3
|
|
|
4 args <- commandArgs(trailingOnly = TRUE)
|
|
|
5
|
|
|
6 input = args[1]
|
|
|
7 genes = unlist(strsplit(args[2], ","))
|
|
|
8 outputdir = args[3]
|
|
|
9 print(args[4])
|
|
|
10 include_fr1 = ifelse(args[4] == "yes", T, F)
|
|
|
11 setwd(outputdir)
|
|
|
12
|
|
|
13 dat = read.table(input, header=T, sep="\t", fill=T, stringsAsFactors=F)
|
|
|
14
|
|
|
15 if(length(dat$Sequence.ID) == 0){
|
|
|
16 setwd(outputdir)
|
|
|
17 result = data.frame(x = rep(0, 5), y = rep(0, 5), z = rep(NA, 5))
|
|
|
18 row.names(result) = c("Number of Mutations (%)", "Transition (%)", "Transversions (%)", "Transitions at G C (%)", "Targeting of C G (%)")
|
|
|
19 write.table(x=result, file="mutations.txt", sep=",",quote=F,row.names=T,col.names=F)
|
|
|
20 transitionTable = data.frame(A=rep(0, 4),C=rep(0, 4),G=rep(0, 4),T=rep(0, 4))
|
|
|
21 row.names(transitionTable) = c("A", "C", "G", "T")
|
|
|
22 transitionTable["A","A"] = NA
|
|
|
23 transitionTable["C","C"] = NA
|
|
|
24 transitionTable["G","G"] = NA
|
|
|
25 transitionTable["T","T"] = NA
|
|
|
26 write.table(x=transitionTable, file="transitions.txt", sep=",",quote=F,row.names=T,col.names=NA)
|
|
|
27 cat("0", file="n.txt")
|
|
|
28 stop("No data")
|
|
|
29 }
|
|
|
30
|
|
|
31
|
|
|
32
|
|
|
33 cleanup_columns = c("FR1.IMGT.c.a",
|
|
|
34 "FR2.IMGT.g.t",
|
|
|
35 "CDR1.IMGT.Nb.of.nucleotides",
|
|
|
36 "CDR2.IMGT.t.a",
|
|
|
37 "FR1.IMGT.c.g",
|
|
|
38 "CDR1.IMGT.c.t",
|
|
|
39 "FR2.IMGT.a.c",
|
|
|
40 "FR2.IMGT.Nb.of.mutations",
|
|
|
41 "FR2.IMGT.g.c",
|
|
|
42 "FR2.IMGT.a.g",
|
|
|
43 "FR3.IMGT.t.a",
|
|
|
44 "FR3.IMGT.t.c",
|
|
|
45 "FR2.IMGT.g.a",
|
|
|
46 "FR3.IMGT.c.g",
|
|
|
47 "FR1.IMGT.Nb.of.mutations",
|
|
|
48 "CDR1.IMGT.g.a",
|
|
|
49 "CDR1.IMGT.t.g",
|
|
|
50 "CDR1.IMGT.g.c",
|
|
|
51 "CDR2.IMGT.Nb.of.nucleotides",
|
|
|
52 "FR2.IMGT.a.t",
|
|
|
53 "CDR1.IMGT.Nb.of.mutations",
|
|
|
54 "CDR1.IMGT.a.g",
|
|
|
55 "FR3.IMGT.a.c",
|
|
|
56 "FR1.IMGT.g.a",
|
|
|
57 "FR3.IMGT.a.g",
|
|
|
58 "FR1.IMGT.a.t",
|
|
|
59 "CDR2.IMGT.a.g",
|
|
|
60 "CDR2.IMGT.Nb.of.mutations",
|
|
|
61 "CDR2.IMGT.g.t",
|
|
|
62 "CDR2.IMGT.a.c",
|
|
|
63 "CDR1.IMGT.t.c",
|
|
|
64 "FR3.IMGT.g.c",
|
|
|
65 "FR1.IMGT.g.t",
|
|
|
66 "FR3.IMGT.g.t",
|
|
|
67 "CDR1.IMGT.a.t",
|
|
|
68 "FR1.IMGT.a.g",
|
|
|
69 "FR3.IMGT.a.t",
|
|
|
70 "FR3.IMGT.Nb.of.nucleotides",
|
|
|
71 "FR2.IMGT.t.c",
|
|
|
72 "CDR2.IMGT.g.a",
|
|
|
73 "FR2.IMGT.t.a",
|
|
|
74 "CDR1.IMGT.t.a",
|
|
|
75 "FR2.IMGT.t.g",
|
|
|
76 "FR3.IMGT.t.g",
|
|
|
77 "FR2.IMGT.Nb.of.nucleotides",
|
|
|
78 "FR1.IMGT.t.a",
|
|
|
79 "FR1.IMGT.t.g",
|
|
|
80 "FR3.IMGT.c.t",
|
|
|
81 "FR1.IMGT.t.c",
|
|
|
82 "CDR2.IMGT.a.t",
|
|
|
83 "FR2.IMGT.c.t",
|
|
|
84 "CDR1.IMGT.g.t",
|
|
|
85 "CDR2.IMGT.t.g",
|
|
|
86 "FR1.IMGT.Nb.of.nucleotides",
|
|
|
87 "CDR1.IMGT.c.g",
|
|
|
88 "CDR2.IMGT.t.c",
|
|
|
89 "FR3.IMGT.g.a",
|
|
|
90 "CDR1.IMGT.a.c",
|
|
|
91 "FR2.IMGT.c.a",
|
|
|
92 "FR3.IMGT.Nb.of.mutations",
|
|
|
93 "FR2.IMGT.c.g",
|
|
|
94 "CDR2.IMGT.g.c",
|
|
|
95 "FR1.IMGT.g.c",
|
|
|
96 "CDR2.IMGT.c.t",
|
|
|
97 "FR3.IMGT.c.a",
|
|
|
98 "CDR1.IMGT.c.a",
|
|
|
99 "CDR2.IMGT.c.g",
|
|
|
100 "CDR2.IMGT.c.a",
|
|
|
101 "FR1.IMGT.c.t",
|
|
|
102 "FR1.IMGT.Nb.of.silent.mutations",
|
|
|
103 "FR2.IMGT.Nb.of.silent.mutations",
|
|
|
104 "FR3.IMGT.Nb.of.silent.mutations",
|
|
|
105 "FR1.IMGT.Nb.of.nonsilent.mutations",
|
|
|
106 "FR2.IMGT.Nb.of.nonsilent.mutations",
|
|
|
107 "FR3.IMGT.Nb.of.nonsilent.mutations")
|
|
|
108
|
|
|
109 for(col in cleanup_columns){
|
|
|
110 dat[,col] = gsub("\\(.*\\)", "", dat[,col])
|
|
|
111 #dat[dat[,col] == "",] = "0"
|
|
|
112 dat[,col] = as.numeric(dat[,col])
|
|
|
113 dat[is.na(dat[,col]),] = 0
|
|
|
114 }
|
|
|
115
|
|
|
116 regions = c("FR1", "CDR1", "FR2", "CDR2", "FR3")
|
|
|
117 if(!include_fr1){
|
|
|
118 regions = c("CDR1", "FR2", "CDR2", "FR3")
|
|
|
119 }
|
|
|
120
|
|
|
121 sum_by_row = function(x, columns) { sum(as.numeric(x[columns]), na.rm=T) }
|
|
|
122
|
|
|
123 VRegionMutations_columns = paste(regions, ".IMGT.Nb.of.mutations", sep="")
|
|
|
124 dat$VRegionMutations = apply(dat, FUN=sum_by_row, 1, columns=VRegionMutations_columns)
|
|
|
125
|
|
|
126 VRegionNucleotides_columns = paste(regions, ".IMGT.Nb.of.nucleotides", sep="")
|
|
|
127 dat$VRegionNucleotides = apply(dat, FUN=sum_by_row, 1, columns=VRegionNucleotides_columns)
|
|
|
128
|
|
|
129 transitionMutations_columns = paste(rep(regions, each=4), c(".IMGT.a.g", ".IMGT.g.a", ".IMGT.c.t", ".IMGT.t.c"), sep="")
|
|
|
130 dat$transitionMutations = apply(dat, FUN=sum_by_row, 1, columns=transitionMutations_columns)
|
|
|
131
|
|
|
132 transversionMutations_columns = paste(rep(regions, each=8), c(".IMGT.a.c",".IMGT.c.a",".IMGT.a.t",".IMGT.t.a",".IMGT.g.c",".IMGT.c.g",".IMGT.g.t",".IMGT.t.g"), sep="")
|
|
|
133 dat$transversionMutations = apply(dat, FUN=sum_by_row, 1, columns=transversionMutations_columns)
|
|
|
134
|
|
|
135
|
|
|
136 transitionMutationsAtGC_columns = paste(rep(regions, each=2), c(".IMGT.g.a",".IMGT.c.t"), sep="")
|
|
|
137 dat$transitionMutationsAtGC = apply(dat, FUN=sum_by_row, 1, columns=transitionMutationsAtGC_columns)
|
|
|
138
|
|
|
139
|
|
|
140 totalMutationsAtGC_columns = paste(rep(regions, each=6), c(".IMGT.c.g",".IMGT.c.t",".IMGT.c.a",".IMGT.g.c",".IMGT.g.a",".IMGT.g.t"), sep="")
|
|
|
141 #totalMutationsAtGC_columns = paste(rep(regions, each=6), c(".IMGT.g.a",".IMGT.c.t",".IMGT.c.a",".IMGT.c.g",".IMGT.g.t"), sep="")
|
|
|
142 dat$totalMutationsAtGC = apply(dat, FUN=sum_by_row, 1, columns=totalMutationsAtGC_columns)
|
|
|
143
|
|
|
144 transitionMutationsAtAT_columns = paste(rep(regions, each=2), c(".IMGT.a.g",".IMGT.t.c"), sep="")
|
|
|
145 dat$transitionMutationsAtAT = apply(dat, FUN=sum_by_row, 1, columns=transitionMutationsAtAT_columns)
|
|
|
146
|
|
|
147 totalMutationsAtAT_columns = paste(rep(regions, each=6), c(".IMGT.a.g",".IMGT.a.c",".IMGT.a.t",".IMGT.t.g",".IMGT.t.c",".IMGT.t.a"), sep="")
|
|
|
148 #totalMutationsAtAT_columns = paste(rep(regions, each=5), c(".IMGT.a.g",".IMGT.t.c",".IMGT.a.c",".IMGT.g.c",".IMGT.t.g"), sep="")
|
|
|
149 dat$totalMutationsAtAT = apply(dat, FUN=sum_by_row, 1, columns=totalMutationsAtAT_columns)
|
|
|
150
|
|
|
151
|
|
|
152 FRRegions = regions[grepl("FR", regions)]
|
|
|
153 CDRRegions = regions[grepl("CDR", regions)]
|
|
|
154
|
|
|
155 FR_silentMutations_columns = paste(FRRegions, ".IMGT.Nb.of.silent.mutations", sep="")
|
|
|
156 dat$silentMutationsFR = apply(dat, FUN=sum_by_row, 1, columns=FR_silentMutations_columns)
|
|
|
157
|
|
|
158 CDR_silentMutations_columns = paste(CDRRegions, ".IMGT.Nb.of.silent.mutations", sep="")
|
|
|
159 dat$silentMutationsCDR = apply(dat, FUN=sum_by_row, 1, columns=CDR_silentMutations_columns)
|
|
|
160
|
|
|
161 FR_nonSilentMutations_columns = paste(FRRegions, ".IMGT.Nb.of.nonsilent.mutations", sep="")
|
|
|
162 dat$nonSilentMutationsFR = apply(dat, FUN=sum_by_row, 1, columns=FR_nonSilentMutations_columns)
|
|
|
163
|
|
|
164 CDR_nonSilentMutations_columns = paste(CDRRegions, ".IMGT.Nb.of.nonsilent.mutations", sep="")
|
|
|
165 dat$nonSilentMutationsCDR = apply(dat, FUN=sum_by_row, 1, columns=CDR_nonSilentMutations_columns)
|
|
|
166
|
|
|
167 mutation.sum.columns = c("Sequence.ID", "VRegionMutations", "VRegionNucleotides", "transitionMutations", "transversionMutations", "transitionMutationsAtGC", "transitionMutationsAtAT", "silentMutationsFR", "nonSilentMutationsFR", "silentMutationsCDR", "nonSilentMutationsCDR")
|
|
|
168
|
|
|
169 write.table(dat[,mutation.sum.columns], "mutation_by_id.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
170
|
|
|
171 setwd(outputdir)
|
|
|
172
|
|
|
173 nts = c("a", "c", "g", "t")
|
|
|
174 zeros=rep(0, 4)
|
|
|
175 matrx = matrix(data = 0, ncol=((length(genes) + 1) * 3),nrow=9)
|
|
|
176 for(i in 1:length(genes)){
|
|
|
177 gene = genes[i]
|
|
|
178 tmp = dat[grepl(paste(".*", gene, ".*", sep=""), dat$best_match),]
|
|
|
179 if(gene == "."){
|
|
|
180 tmp = dat
|
|
|
181 }
|
|
|
182 j = i - 1
|
|
|
183 x = (j * 3) + 1
|
|
|
184 y = (j * 3) + 2
|
|
|
185 z = (j * 3) + 3
|
|
|
186 matrx[1,x] = sum(tmp$VRegionMutations)
|
|
|
187 matrx[1,y] = sum(tmp$VRegionNucleotides)
|
|
|
188 matrx[1,z] = round(matrx[1,x] / matrx[1,y] * 100, digits=1)
|
|
|
189
|
|
|
190 matrx[2,x] = sum(tmp$transitionMutations)
|
|
|
191 matrx[2,y] = sum(tmp$VRegionMutations)
|
|
|
192 matrx[2,z] = round(matrx[2,x] / matrx[2,y] * 100, digits=1)
|
|
|
193
|
|
|
194 matrx[3,x] = sum(tmp$transversionMutations)
|
|
|
195 matrx[3,y] = sum(tmp$VRegionMutations)
|
|
|
196 matrx[3,z] = round(matrx[3,x] / matrx[3,y] * 100, digits=1)
|
|
|
197
|
|
|
198 matrx[4,x] = sum(tmp$transitionMutationsAtGC)
|
|
|
199 matrx[4,y] = sum(tmp$totalMutationsAtGC)
|
|
|
200 matrx[4,z] = round(matrx[4,x] / matrx[4,y] * 100, digits=1)
|
|
|
201
|
|
|
202 matrx[5,x] = sum(tmp$totalMutationsAtGC)
|
|
|
203 matrx[5,y] = sum(tmp$VRegionMutations)
|
|
|
204 matrx[5,z] = round(matrx[5,x] / matrx[5,y] * 100, digits=1)
|
|
|
205
|
|
|
206 matrx[6,x] = sum(tmp$transitionMutationsAtAT)
|
|
|
207 matrx[6,y] = sum(tmp$totalMutationsAtAT)
|
|
|
208 matrx[6,z] = round(matrx[6,x] / matrx[6,y] * 100, digits=1)
|
|
|
209
|
|
|
210 matrx[7,x] = sum(tmp$totalMutationsAtAT)
|
|
|
211 matrx[7,y] = sum(tmp$VRegionMutations)
|
|
|
212 matrx[7,z] = round(matrx[7,x] / matrx[7,y] * 100, digits=1)
|
|
|
213
|
|
|
214 matrx[8,x] = sum(tmp$nonSilentMutationsFR)
|
|
|
215 matrx[8,y] = sum(tmp$silentMutationsFR)
|
|
|
216 matrx[8,z] = round(matrx[8,x] / matrx[8,y], digits=1)
|
|
|
217
|
|
|
218 matrx[9,x] = sum(tmp$nonSilentMutationsCDR)
|
|
|
219 matrx[9,y] = sum(tmp$silentMutationsCDR)
|
|
|
220 matrx[9,z] = round(matrx[9,x] / matrx[9,y], digits=1)
|
|
|
221
|
|
|
222
|
|
|
223 transitionTable = data.frame(A=zeros,C=zeros,G=zeros,T=zeros)
|
|
|
224 row.names(transitionTable) = c("A", "C", "G", "T")
|
|
|
225 transitionTable["A","A"] = NA
|
|
|
226 transitionTable["C","C"] = NA
|
|
|
227 transitionTable["G","G"] = NA
|
|
|
228 transitionTable["T","T"] = NA
|
|
|
229
|
|
|
230 if(nrow(tmp) > 0){
|
|
|
231 for(nt1 in nts){
|
|
|
232 for(nt2 in nts){
|
|
|
233 if(nt1 == nt2){
|
|
|
234 next
|
|
|
235 }
|
|
|
236 NT1 = LETTERS[letters == nt1]
|
|
|
237 NT2 = LETTERS[letters == nt2]
|
|
|
238 FR1 = paste("FR1.IMGT.", nt1, ".", nt2, sep="")
|
|
|
239 CDR1 = paste("CDR1.IMGT.", nt1, ".", nt2, sep="")
|
|
|
240 FR2 = paste("FR2.IMGT.", nt1, ".", nt2, sep="")
|
|
|
241 CDR2 = paste("CDR2.IMGT.", nt1, ".", nt2, sep="")
|
|
|
242 FR3 = paste("FR3.IMGT.", nt1, ".", nt2, sep="")
|
|
|
243 if(include_fr1){
|
|
|
244 transitionTable[NT1,NT2] = sum(tmp[,c(FR1, CDR1, FR2, CDR2, FR3)])
|
|
|
245 } else {
|
|
|
246 transitionTable[NT1,NT2] = sum(tmp[,c(CDR1, FR2, CDR2, FR3)])
|
|
|
247 }
|
|
|
248 }
|
|
|
249 }
|
|
|
250 }
|
|
|
251
|
|
|
252
|
|
|
253 write.table(x=transitionTable, file=paste("transitions_", gene ,".txt", sep=""), sep=",",quote=F,row.names=T,col.names=NA)
|
|
|
254 write.table(x=tmp[,c("Sequence.ID", "best_match", "chunk_hit_percentage", "nt_hit_percentage", "start_locations")], file=paste("matched_", gene ,".txt", sep=""), sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
255
|
|
|
256 cat(matrx[1,x], file=paste(gene, "_value.txt" ,sep=""))
|
|
|
257 cat(length(tmp$Sequence.ID), file=paste(gene, "_n.txt" ,sep=""))
|
|
|
258 }
|
|
|
259
|
|
|
260 #again for all of the data
|
|
|
261 tmp = dat
|
|
|
262 j = i
|
|
|
263 x = (j * 3) + 1
|
|
|
264 y = (j * 3) + 2
|
|
|
265 z = (j * 3) + 3
|
|
|
266 matrx[1,x] = sum(tmp$VRegionMutations)
|
|
|
267 matrx[1,y] = sum(tmp$VRegionNucleotides)
|
|
|
268 matrx[1,z] = round(matrx[1,x] / matrx[1,y] * 100, digits=1)
|
|
|
269
|
|
|
270 matrx[2,x] = sum(tmp$transitionMutations)
|
|
|
271 matrx[2,y] = sum(tmp$VRegionMutations)
|
|
|
272 matrx[2,z] = round(matrx[2,x] / matrx[2,y] * 100, digits=1)
|
|
|
273
|
|
|
274 matrx[3,x] = sum(tmp$transversionMutations)
|
|
|
275 matrx[3,y] = sum(tmp$VRegionMutations)
|
|
|
276 matrx[3,z] = round(matrx[3,x] / matrx[3,y] * 100, digits=1)
|
|
|
277
|
|
|
278 matrx[4,x] = sum(tmp$transitionMutationsAtGC)
|
|
|
279 matrx[4,y] = sum(tmp$totalMutationsAtGC)
|
|
|
280 matrx[4,z] = round(matrx[4,x] / matrx[4,y] * 100, digits=1)
|
|
|
281
|
|
|
282 matrx[5,x] = sum(tmp$totalMutationsAtGC)
|
|
|
283 matrx[5,y] = sum(tmp$VRegionMutations)
|
|
|
284 matrx[5,z] = round(matrx[5,x] / matrx[5,y] * 100, digits=1)
|
|
|
285
|
|
|
286 matrx[6,x] = sum(tmp$transitionMutationsAtAT)
|
|
|
287 matrx[6,y] = sum(tmp$totalMutationsAtAT)
|
|
|
288 matrx[6,z] = round(matrx[6,x] / matrx[6,y] * 100, digits=1)
|
|
|
289
|
|
|
290 matrx[7,x] = sum(tmp$totalMutationsAtAT)
|
|
|
291 matrx[7,y] = sum(tmp$VRegionMutations)
|
|
|
292 matrx[7,z] = round(matrx[7,x] / matrx[7,y] * 100, digits=1)
|
|
|
293
|
|
|
294 matrx[8,x] = sum(tmp$nonSilentMutationsFR)
|
|
|
295 matrx[8,y] = sum(tmp$silentMutationsFR)
|
|
|
296 matrx[8,z] = round(matrx[8,x] / matrx[8,y], digits=1)
|
|
|
297
|
|
|
298 matrx[9,x] = sum(tmp$nonSilentMutationsCDR)
|
|
|
299 matrx[9,y] = sum(tmp$silentMutationsCDR)
|
|
|
300 matrx[9,z] = round(matrx[9,x] / matrx[9,y], digits=1)
|
|
|
301
|
|
|
302 transitionTable = data.frame(A=1:4,C=1:4,G=1:4,T=1:4)
|
|
|
303 row.names(transitionTable) = c("A", "C", "G", "T")
|
|
|
304 transitionTable["A","A"] = NA
|
|
|
305 transitionTable["C","C"] = NA
|
|
|
306 transitionTable["G","G"] = NA
|
|
|
307 transitionTable["T","T"] = NA
|
|
|
308
|
|
|
309
|
|
|
310 for(nt1 in nts){
|
|
|
311 for(nt2 in nts){
|
|
|
312 if(nt1 == nt2){
|
|
|
313 next
|
|
|
314 }
|
|
|
315 NT1 = LETTERS[letters == nt1]
|
|
|
316 NT2 = LETTERS[letters == nt2]
|
|
|
317 FR1 = paste("FR1.IMGT.", nt1, ".", nt2, sep="")
|
|
|
318 CDR1 = paste("CDR1.IMGT.", nt1, ".", nt2, sep="")
|
|
|
319 FR2 = paste("FR2.IMGT.", nt1, ".", nt2, sep="")
|
|
|
320 CDR2 = paste("CDR2.IMGT.", nt1, ".", nt2, sep="")
|
|
|
321 FR3 = paste("FR3.IMGT.", nt1, ".", nt2, sep="")
|
|
|
322 if(include_fr1){
|
|
|
323 transitionTable[NT1,NT2] = sum(tmp[,c(FR1, CDR1, FR2, CDR2, FR3)])
|
|
|
324 } else {
|
|
|
325 transitionTable[NT1,NT2] = sum(tmp[,c(CDR1, FR2, CDR2, FR3)])
|
|
|
326 }
|
|
|
327 }
|
|
|
328 }
|
|
|
329 write.table(x=transitionTable, file="transitions.txt", sep=",",quote=F,row.names=T,col.names=NA)
|
|
|
330 write.table(x=tmp[,c("Sequence.ID", "best_match", "chunk_hit_percentage", "nt_hit_percentage", "start_locations")], file="matched_all.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
331 cat(matrx[1,x], file="total_value.txt")
|
|
|
332 cat(length(tmp$Sequence.ID), file="total_n.txt")
|
|
|
333
|
|
|
334
|
|
|
335
|
|
|
336 result = data.frame(matrx)
|
|
|
337 row.names(result) = c("Number of Mutations (%)", "Transition (%)", "Transversions (%)", "Transitions at G C (%)", "Targeting of C G (%)", "Transitions at A T (%)", "Targeting of A T (%)", "FR R/S (ratio)", "CDR R/S (ratio)")
|
|
|
338
|
|
|
339 write.table(x=result, file="mutations.txt", sep=",",quote=F,row.names=T,col.names=F)
|
|
|
340
|
|
|
341
|
|
|
342 if (!("ggplot2" %in% rownames(installed.packages()))) {
|
|
|
343 install.packages("ggplot2", repos="http://cran.xl-mirror.nl/")
|
|
|
344 }
|
|
|
345
|
|
|
346
|
|
|
347 genesForPlot = gsub("[0-9]", "", dat$best_match)
|
|
|
348 genesForPlot = data.frame(table(genesForPlot))
|
|
|
349 colnames(genesForPlot) = c("Gene","Freq")
|
|
|
350 genesForPlot$label = paste(genesForPlot$Gene, "-", genesForPlot$Freq)
|
|
|
351 write.table(genesForPlot, "all.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
352
|
|
|
353
|
|
|
354 pc = ggplot(genesForPlot, aes(x = factor(1), y=Freq, fill=label))
|
|
|
355 pc = pc + geom_bar(width = 1, stat = "identity")
|
|
|
356 pc = pc + coord_polar(theta="y")
|
|
|
357 pc = pc + xlab(" ") + ylab(" ") + ggtitle(paste("Classes", "( n =", sum(genesForPlot$Freq), ")"))
|
|
|
358
|
|
|
359 png(filename="all.png")
|
|
|
360 pc
|
|
|
361 dev.off()
|
|
|
362
|
|
|
363
|
|
|
364 #blegh
|
|
|
365 genesForPlot = dat[grepl("ca", dat$best_match),]$best_match
|
|
|
366 if(length(genesForPlot) > 0){
|
|
|
367 genesForPlot = data.frame(table(genesForPlot))
|
|
|
368 colnames(genesForPlot) = c("Gene","Freq")
|
|
|
369 genesForPlot$label = paste(genesForPlot$Gene, "-", genesForPlot$Freq)
|
|
|
370
|
|
|
371 pc = ggplot(genesForPlot, aes(x = factor(1), y=Freq, fill=label))
|
|
|
372 pc = pc + geom_bar(width = 1, stat = "identity")
|
|
|
373 pc = pc + coord_polar(theta="y")
|
|
|
374 pc = pc + xlab(" ") + ylab(" ") + ggtitle(paste("IgA subclasses", "( n =", sum(genesForPlot$Freq), ")"))
|
|
|
375 write.table(genesForPlot, "ca.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
376
|
|
|
377 png(filename="ca.png")
|
|
|
378 print(pc)
|
|
|
379 dev.off()
|
|
|
380 }
|
|
|
381
|
|
|
382 genesForPlot = dat[grepl("cg", dat$best_match),]$best_match
|
|
|
383 if(length(genesForPlot) > 0){
|
|
|
384 genesForPlot = data.frame(table(genesForPlot))
|
|
|
385 colnames(genesForPlot) = c("Gene","Freq")
|
|
|
386 genesForPlot$label = paste(genesForPlot$Gene, "-", genesForPlot$Freq)
|
|
|
387
|
|
|
388 pc = ggplot(genesForPlot, aes(x = factor(1), y=Freq, fill=label))
|
|
|
389 pc = pc + geom_bar(width = 1, stat = "identity")
|
|
|
390 pc = pc + coord_polar(theta="y")
|
|
|
391 pc = pc + xlab(" ") + ylab(" ") + ggtitle(paste("IgG subclasses", "( n =", sum(genesForPlot$Freq), ")"))
|
|
|
392 write.table(genesForPlot, "cg.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
393
|
|
|
394 png(filename="cg.png")
|
|
|
395 print(pc)
|
|
|
396 dev.off()
|
|
|
397 }
|
|
|
398
|
|
|
399 dat$percentage_mutations = round(dat$VRegionMutations / dat$VRegionNucleotides * 100, 2)
|
|
|
400
|
|
|
401 p = ggplot(dat, aes(best_match, percentage_mutations))
|
|
|
402 p = p + geom_point(aes(colour=best_match), position="jitter") + geom_boxplot(aes(middle=mean(percentage_mutations)), alpha=0.1, outlier.shape = NA)
|
|
|
403 p = p + xlab("Subclass") + ylab("Frequency") + ggtitle("Frequency scatter plot")
|
|
|
404
|
|
|
405 png(filename="scatter.png")
|
|
|
406 print(p)
|
|
|
407 dev.off()
|
|
|
408
|
|
|
409 write.table(dat[,c("Sequence.ID", "best_match", "VRegionMutations", "VRegionNucleotides", "percentage_mutations")], "scatter.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
410
|
|
|
411 write.table(dat, input, sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
412
|
|
|
413
|
|
|
414
|
|
|
415
|
|
|
416
|
|
|
417
|
|
|
418 dat$best_match_class = substr(dat$best_match, 0, 2)
|
|
|
419 freq_labels = c("0", "0-2", "2-5", "5-10", "10-15", "15-20", "20")
|
|
|
420 dat$frequency_bins = cut(dat$percentage_mutations, breaks=c(-Inf, 0, 2,5,10,15,20, Inf), labels=freq_labels)
|
|
|
421
|
|
|
422 frequency_bins_data = data.frame(data.table(dat)[, list(frequency_count=.N), by=c("best_match_class", "frequency_bins")])
|
|
|
423
|
|
|
424 p = ggplot(frequency_bins_data, aes(frequency_bins, frequency_count))
|
|
|
425 p = p + geom_bar(aes(fill=best_match_class), stat="identity", position="dodge")
|
|
|
426 p = p + xlab("Frequency ranges") + ylab("Frequency") + ggtitle("Mutation Frequencies by class")
|
|
|
427
|
|
|
428 png(filename="frequency_ranges.png")
|
|
|
429 print(p)
|
|
|
430 dev.off()
|
|
|
431
|
|
|
432 frequency_bins_data_by_class = frequency_bins_data
|
|
|
433
|
|
|
434 write.table(frequency_bins_data_by_class, "frequency_ranges_classes.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
435
|
|
|
436 frequency_bins_data = data.frame(data.table(dat)[, list(frequency_count=.N), by=c("best_match", "frequency_bins")])
|
|
|
437
|
|
|
438 write.table(frequency_bins_data, "frequency_ranges_subclasses.txt", sep="\t",quote=F,row.names=F,col.names=T)
|
|
|
439
|
|
|
440
|
|
|
441 #frequency_bins_data_by_class
|
|
|
442 #frequency_ranges_subclasses.txt
|
|
|
443
|
|
|
444
|
|
|
445
|
|
|
446
|
|
|
447
|
|
|
448
|
|
|
449
|
|
|
450
|
|
|
451
|
|
|
452
|
|
|
453
|
|
|
454
|
|
|
455
|
|
|
456
|
|
|
457
|
|
|
458
|
|
|
459
|
|
|
460
|
|
|
461
|
|
|
462
|
|
|
463
|
|
|
464
|
|
|
465
|
|
|
466
|
|
|
467
|
|
|
468
|
|
|
469
|
