|
0
|
1 library(reshape2)
|
|
|
2
|
|
|
3 args <- commandArgs(trailingOnly = TRUE)
|
|
|
4
|
|
|
5 before.unique.file = args[1]
|
|
|
6 merged.file = args[2]
|
|
|
7 outputdir = args[3]
|
|
|
8 gene.classes = unlist(strsplit(args[4], ","))
|
|
|
9 hotspot.analysis.sum.file = args[5]
|
|
|
10 NToverview.file = paste(outputdir, "ntoverview.txt", sep="/")
|
|
|
11 NTsum.file = paste(outputdir, "ntsum.txt", sep="/")
|
|
|
12 main.html = "index.html"
|
|
7
|
13 empty.region.filter = args[6]
|
|
|
14
|
|
0
|
15
|
|
|
16 setwd(outputdir)
|
|
|
17
|
|
|
18 before.unique = read.table(before.unique.file, header=T, sep="\t", fill=T, stringsAsFactors=F, quote="")
|
|
|
19 merged = read.table(merged.file, header=T, sep="\t", fill=T, stringsAsFactors=F, quote="")
|
|
|
20 hotspot.analysis.sum = read.table(hotspot.analysis.sum.file, header=F, sep=",", fill=T, stringsAsFactors=F, quote="")
|
|
|
21
|
|
|
22 #before.unique = before.unique[!grepl("unmatched", before.unique$best_match),]
|
|
|
23
|
|
7
|
24 if(empty.region.filter == "leader"){
|
|
31
|
25 before.unique$seq_conc = paste(before.unique$FR1.IMGT.seq, before.unique$CDR1.IMGT.seq, before.unique$FR2.IMGT.seq, before.unique$CDR2.IMGT.seq, before.unique$FR3.IMGT.seq, before.unique$CDR3.IMGT.seq)
|
|
7
|
26 } else if(empty.region.filter == "FR1"){
|
|
31
|
27 before.unique$seq_conc = paste(before.unique$CDR1.IMGT.seq, before.unique$FR2.IMGT.seq, before.unique$CDR2.IMGT.seq, before.unique$FR3.IMGT.seq, before.unique$CDR3.IMGT.seq)
|
|
7
|
28 } else if(empty.region.filter == "CDR1"){
|
|
31
|
29 before.unique$seq_conc = paste(before.unique$FR2.IMGT.seq, before.unique$CDR2.IMGT.seq, before.unique$FR3.IMGT.seq, before.unique$CDR3.IMGT.seq)
|
|
7
|
30 } else if(empty.region.filter == "FR2"){
|
|
31
|
31 before.unique$seq_conc = paste(before.unique$CDR2.IMGT.seq, before.unique$FR3.IMGT.seq, before.unique$CDR3.IMGT.seq)
|
|
7
|
32 }
|
|
0
|
33
|
|
|
34 IDs = before.unique[,c("Sequence.ID", "seq_conc", "best_match", "Functionality")]
|
|
|
35 IDs$best_match = as.character(IDs$best_match)
|
|
|
36
|
|
7
|
37 dat = data.frame(table(before.unique$seq_conc))
|
|
0
|
38
|
|
|
39 names(dat) = c("seq_conc", "Freq")
|
|
|
40
|
|
|
41 dat$seq_conc = factor(dat$seq_conc)
|
|
|
42
|
|
|
43 dat = dat[order(as.character(dat$seq_conc)),]
|
|
|
44
|
|
|
45 #writing html from R...
|
|
|
46 get.bg.color = function(val){
|
|
|
47 if(val %in% c("TRUE", "FALSE", "T", "F")){ #if its a logical value, give the background a green/red color
|
|
|
48 return(ifelse(val,"#eafaf1","#f9ebea"))
|
|
|
49 } else if (!is.na(as.numeric(val))) { #if its a numerical value, give it a grey tint if its >0
|
|
|
50 return(ifelse(val > 0,"#eaecee","white"))
|
|
|
51 } else {
|
|
|
52 return("white")
|
|
|
53 }
|
|
|
54 }
|
|
|
55 td = function(val) {
|
|
|
56 return(paste("<td bgcolor='", get.bg.color(val), "'>", val, "</td>", sep=""))
|
|
|
57 }
|
|
|
58 tr = function(val) {
|
|
|
59 return(paste(c("<tr>", sapply(val, td), "</tr>"), collapse=""))
|
|
|
60 }
|
|
|
61
|
|
|
62 make.link = function(id, clss, val) {
|
|
|
63 paste("<a href='", clss, "_", id, ".html'>", val, "</a>", sep="")
|
|
|
64 }
|
|
|
65 tbl = function(df) {
|
|
|
66 res = "<table border='1'>"
|
|
|
67 for(i in 1:nrow(df)){
|
|
|
68 res = paste(res, tr(df[i,]), sep="")
|
|
|
69 }
|
|
|
70 res = paste(res, "</table>")
|
|
|
71 }
|
|
|
72
|
|
|
73 cat("<table border='1' class='pure-table pure-table-striped'>", file=main.html, append=F)
|
|
7
|
74
|
|
|
75 if(empty.region.filter == "leader"){
|
|
|
76 cat("<caption>FR1+CDR1+FR2+CDR2+FR3+CDR3 sequences that show up more than once</caption>", file=main.html, append=T)
|
|
|
77 } else if(empty.region.filter == "FR1"){
|
|
|
78 cat("<caption>CDR1+FR2+CDR2+FR3+CDR3 sequences that show up more than once</caption>", file=main.html, append=T)
|
|
|
79 } else if(empty.region.filter == "CDR1"){
|
|
|
80 cat("<caption>FR2+CDR2+FR3+CDR3 sequences that show up more than once</caption>", file=main.html, append=T)
|
|
|
81 } else if(empty.region.filter == "FR2"){
|
|
|
82 cat("<caption>CDR2+FR3+CDR3 sequences that show up more than once</caption>", file=main.html, append=T)
|
|
|
83 }
|
|
|
84
|
|
0
|
85 cat("<tr>", file=main.html, append=T)
|
|
32
|
86 cat("<th>Sequence</th><th>Functionality</th><th>IGA1</th><th>IGA2</th><th>IGG1</th><th>IGG2</th><th>IGG3</th><th>IGG4</th><th>IGM</th><th>IGE</th><th>UN</th>", file=main.html, append=T)
|
|
|
87 cat("<th>total IGA</th><th>total IGG</th><th>total IGE</th><th>total IGM</th><th>number of subclasses</th><th>present in both IGA and IGG</th><th>present in IGA, IGG and IGM</th><th>present in IGA, IGG and IGE</th><th>present in IGA, IGG, IGM and IGE</th><th>IGA1+IGA2</th>", file=main.html, append=T)
|
|
|
88 cat("<th>IGG1+IGG2</th><th>IGG1+IGG3</th><th>IGG1+IGG4</th><th>IGG2+IGG3</th><th>IGG2+IGG4</th><th>IGG3+IGG4</th>", file=main.html, append=T)
|
|
|
89 cat("<th>IGG1+IGG2+IGG3</th><th>IGG2+IGG3+IGG4</th><th>IGG1+IGG2+IGG4</th><th>IGG1+IGG3+IGG4</th><th>IGG1+IGG2+IGG3+IGG4</th>", file=main.html, append=T)
|
|
0
|
90 cat("</tr>", file=main.html, append=T)
|
|
|
91
|
|
|
92
|
|
|
93
|
|
|
94 single.sequences=0 #sequence only found once, skipped
|
|
|
95 in.multiple=0 #same sequence across multiple subclasses
|
|
|
96 multiple.in.one=0 #same sequence multiple times in one subclass
|
|
|
97 unmatched=0 #all of the sequences are unmatched
|
|
|
98 some.unmatched=0 #one or more sequences in a clone are unmatched
|
|
|
99 matched=0 #should be the same als matched sequences
|
|
|
100
|
|
|
101 sequence.id.page="by_id.html"
|
|
|
102
|
|
|
103 for(i in 1:nrow(dat)){
|
|
|
104
|
|
|
105 ca1 = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGA1", IDs$best_match),]
|
|
|
106 ca2 = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGA2", IDs$best_match),]
|
|
|
107
|
|
|
108 cg1 = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGG1", IDs$best_match),]
|
|
|
109 cg2 = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGG2", IDs$best_match),]
|
|
|
110 cg3 = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGG3", IDs$best_match),]
|
|
|
111 cg4 = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGG4", IDs$best_match),]
|
|
|
112
|
|
|
113 cm = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGM", IDs$best_match),]
|
|
|
114
|
|
32
|
115 ce = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^IGE", IDs$best_match),]
|
|
|
116
|
|
0
|
117 un = IDs[IDs$seq_conc == dat[i,c("seq_conc")] & grepl("^unmatched", IDs$best_match),]
|
|
32
|
118
|
|
|
119 allc = rbind(ca1, ca2, cg1, cg2, cg3, cg4, cm, ce, un)
|
|
0
|
120
|
|
|
121 ca1.n = nrow(ca1)
|
|
|
122 ca2.n = nrow(ca2)
|
|
|
123
|
|
|
124 cg1.n = nrow(cg1)
|
|
|
125 cg2.n = nrow(cg2)
|
|
|
126 cg3.n = nrow(cg3)
|
|
|
127 cg4.n = nrow(cg4)
|
|
|
128
|
|
|
129 cm.n = nrow(cm)
|
|
|
130
|
|
32
|
131 ce.n = nrow(ce)
|
|
|
132
|
|
0
|
133 un.n = nrow(un)
|
|
|
134
|
|
32
|
135 classes = c(ca1.n, ca2.n, cg1.n, cg2.n, cg3.n, cg4.n, cm.n, ce.n, un.n)
|
|
0
|
136
|
|
|
137 classes.sum = sum(classes)
|
|
|
138
|
|
|
139 if(classes.sum == 1){
|
|
|
140 single.sequences = single.sequences + 1
|
|
|
141 next
|
|
|
142 }
|
|
|
143
|
|
|
144 if(un.n == classes.sum){
|
|
|
145 unmatched = unmatched + 1
|
|
|
146 next
|
|
|
147 }
|
|
|
148
|
|
|
149 in.classes = sum(classes > 0)
|
|
|
150
|
|
|
151 matched = matched + in.classes #count in how many subclasses the sequence occurs.
|
|
|
152
|
|
32
|
153 if(any(classes == classes.sum)){
|
|
0
|
154 multiple.in.one = multiple.in.one + 1
|
|
|
155 } else if (un.n > 0) {
|
|
|
156 some.unmatched = some.unmatched + 1
|
|
|
157 } else {
|
|
|
158 in.multiple = in.multiple + 1
|
|
|
159 }
|
|
|
160
|
|
|
161 id = as.numeric(dat[i,"seq_conc"])
|
|
|
162
|
|
|
163 functionality = paste(unique(allc[,"Functionality"]), collapse=",")
|
|
|
164
|
|
|
165 by.id.row = c()
|
|
|
166
|
|
|
167 if(ca1.n > 0){
|
|
|
168 cat(tbl(ca1), file=paste("IGA1_", id, ".html", sep=""))
|
|
|
169 }
|
|
|
170
|
|
|
171 if(ca2.n > 0){
|
|
|
172 cat(tbl(ca2), file=paste("IGA2_", id, ".html", sep=""))
|
|
|
173 }
|
|
|
174
|
|
|
175 if(cg1.n > 0){
|
|
|
176 cat(tbl(cg1), file=paste("IGG1_", id, ".html", sep=""))
|
|
|
177 }
|
|
|
178
|
|
|
179 if(cg2.n > 0){
|
|
|
180 cat(tbl(cg2), file=paste("IGG2_", id, ".html", sep=""))
|
|
|
181 }
|
|
|
182
|
|
|
183 if(cg3.n > 0){
|
|
|
184 cat(tbl(cg3), file=paste("IGG3_", id, ".html", sep=""))
|
|
|
185 }
|
|
|
186
|
|
|
187 if(cg4.n > 0){
|
|
|
188 cat(tbl(cg4), file=paste("IGG4_", id, ".html", sep=""))
|
|
|
189 }
|
|
|
190
|
|
|
191 if(cm.n > 0){
|
|
|
192 cat(tbl(cm), file=paste("IGM_", id, ".html", sep=""))
|
|
|
193 }
|
|
|
194
|
|
32
|
195 if(ce.n > 0){
|
|
|
196 cat(tbl(ce), file=paste("IGE_", id, ".html", sep=""))
|
|
|
197 }
|
|
|
198
|
|
0
|
199 if(un.n > 0){
|
|
|
200 cat(tbl(un), file=paste("un_", id, ".html", sep=""))
|
|
|
201 }
|
|
|
202
|
|
|
203 ca1.html = make.link(id, "IGA1", ca1.n)
|
|
|
204 ca2.html = make.link(id, "IGA2", ca2.n)
|
|
|
205
|
|
|
206 cg1.html = make.link(id, "IGG1", cg1.n)
|
|
|
207 cg2.html = make.link(id, "IGG2", cg2.n)
|
|
|
208 cg3.html = make.link(id, "IGG3", cg3.n)
|
|
|
209 cg4.html = make.link(id, "IGG4", cg4.n)
|
|
|
210
|
|
|
211 cm.html = make.link(id, "IGM", cm.n)
|
|
|
212
|
|
32
|
213 ce.html = make.link(id, "IGE", ce.n)
|
|
|
214
|
|
0
|
215 un.html = make.link(id, "un", un.n)
|
|
|
216
|
|
|
217 #extra columns
|
|
|
218 ca.n = ca1.n + ca2.n
|
|
|
219
|
|
|
220 cg.n = cg1.n + cg2.n + cg3.n + cg4.n
|
|
|
221
|
|
|
222 #in.classes
|
|
|
223
|
|
|
224 in.ca.cg = (ca.n > 0 & cg.n > 0)
|
|
|
225
|
|
32
|
226 in.ca.cg.cm = (ca.n > 0 & cg.n > 0 & cm.n > 0)
|
|
|
227
|
|
|
228 in.ca.cg.ce = (ca.n > 0 & cg.n > 0 & ce.n > 0)
|
|
|
229
|
|
|
230 in.ca.cg.cm.ce = (ca.n > 0 & cg.n > 0 & cm.n > 0 & ce.n > 0)
|
|
|
231
|
|
0
|
232 in.ca1.ca2 = (ca1.n > 0 & ca2.n > 0)
|
|
|
233
|
|
|
234 in.cg1.cg2 = (cg1.n > 0 & cg2.n > 0)
|
|
|
235 in.cg1.cg3 = (cg1.n > 0 & cg3.n > 0)
|
|
|
236 in.cg1.cg4 = (cg1.n > 0 & cg4.n > 0)
|
|
|
237 in.cg2.cg3 = (cg2.n > 0 & cg3.n > 0)
|
|
|
238 in.cg2.cg4 = (cg2.n > 0 & cg4.n > 0)
|
|
|
239 in.cg3.cg4 = (cg3.n > 0 & cg4.n > 0)
|
|
|
240
|
|
|
241 in.cg1.cg2.cg3 = (cg1.n > 0 & cg2.n > 0 & cg3.n > 0)
|
|
|
242 in.cg2.cg3.cg4 = (cg2.n > 0 & cg3.n > 0 & cg4.n > 0)
|
|
|
243 in.cg1.cg2.cg4 = (cg1.n > 0 & cg2.n > 0 & cg4.n > 0)
|
|
|
244 in.cg1.cg3.cg4 = (cg1.n > 0 & cg3.n > 0 & cg4.n > 0)
|
|
|
245
|
|
|
246 in.cg.all = (cg1.n > 0 & cg2.n > 0 & cg3.n > 0 & cg4.n > 0)
|
|
|
247
|
|
|
248 #rw = c(as.character(dat[i,"seq_conc"]), functionality, ca1.html, ca2.html, cg1.html, cg2.html, cg3.html, cg4.html, cm.html, un.html)
|
|
32
|
249 rw = c(as.character(dat[i,"seq_conc"]), functionality, ca1.html, ca2.html, cg1.html, cg2.html, cg3.html, cg4.html, cm.html, ce.html, un.html)
|
|
|
250 rw = c(rw, ca.n, cg.n, cm.n, ce.n, in.classes, in.ca.cg, in.ca.cg.cm, in.ca.cg.ce, in.ca.cg.cm.ce, in.ca1.ca2, in.cg1.cg2, in.cg1.cg3, in.cg1.cg4, in.cg2.cg3, in.cg2.cg4, in.cg3.cg4, in.cg1.cg2.cg3, in.cg2.cg3.cg4, in.cg1.cg2.cg4, in.cg1.cg3.cg4, in.cg.all)
|
|
0
|
251
|
|
|
252 cat(tr(rw), file=main.html, append=T)
|
|
|
253
|
|
|
254
|
|
|
255 for(i in 1:nrow(allc)){ #generate html by id
|
|
|
256 html = make.link(id, allc[i,"best_match"], allc[i,"Sequence.ID"])
|
|
|
257 cat(paste(html, "<br />"), file=sequence.id.page, append=T)
|
|
|
258 }
|
|
|
259 }
|
|
|
260
|
|
|
261 cat("</table>", file=main.html, append=T)
|
|
|
262
|
|
|
263 print(paste("Single sequences:", single.sequences))
|
|
|
264 print(paste("Sequences in multiple subclasses:", in.multiple))
|
|
|
265 print(paste("Multiple sequences in one subclass:", multiple.in.one))
|
|
|
266 print(paste("Matched with unmatched:", some.unmatched))
|
|
|
267 print(paste("Count that should match 'matched' sequences:", matched))
|
|
|
268
|
|
|
269 #ACGT overview
|
|
|
270
|
|
7
|
271 #NToverview = merged[!grepl("^unmatched", merged$best_match),]
|
|
|
272 NToverview = merged
|
|
0
|
273
|
|
7
|
274 if(empty.region.filter == "leader"){
|
|
|
275 NToverview$seq = paste(NToverview$FR1.IMGT.seq, NToverview$CDR1.IMGT.seq, NToverview$FR2.IMGT.seq, NToverview$CDR2.IMGT.seq, NToverview$FR3.IMGT.seq)
|
|
|
276 } else if(empty.region.filter == "FR1"){
|
|
|
277 NToverview$seq = paste(NToverview$CDR1.IMGT.seq, NToverview$FR2.IMGT.seq, NToverview$CDR2.IMGT.seq, NToverview$FR3.IMGT.seq)
|
|
|
278 } else if(empty.region.filter == "CDR1"){
|
|
|
279 NToverview$seq = paste(NToverview$FR2.IMGT.seq, NToverview$CDR2.IMGT.seq, NToverview$FR3.IMGT.seq)
|
|
|
280 } else if(empty.region.filter == "FR2"){
|
|
|
281 NToverview$seq = paste(NToverview$CDR2.IMGT.seq, NToverview$FR3.IMGT.seq)
|
|
|
282 }
|
|
0
|
283
|
|
|
284 NToverview$A = nchar(gsub("[^Aa]", "", NToverview$seq))
|
|
|
285 NToverview$C = nchar(gsub("[^Cc]", "", NToverview$seq))
|
|
|
286 NToverview$G = nchar(gsub("[^Gg]", "", NToverview$seq))
|
|
|
287 NToverview$T = nchar(gsub("[^Tt]", "", NToverview$seq))
|
|
|
288
|
|
|
289 #Nsum = data.frame(Sequence.ID="-", best_match="Sum", seq="-", A = sum(NToverview$A), C = sum(NToverview$C), G = sum(NToverview$G), T = sum(NToverview$T))
|
|
|
290
|
|
|
291 #NToverview = rbind(NToverview, NTsum)
|
|
|
292
|
|
|
293 NTresult = data.frame(nt=c("A", "C", "T", "G"))
|
|
|
294
|
|
|
295 for(clazz in gene.classes){
|
|
8
|
296 print(paste("class:", clazz))
|
|
0
|
297 NToverview.sub = NToverview[grepl(paste("^", clazz, sep=""), NToverview$best_match),]
|
|
8
|
298 print(paste("nrow:", nrow(NToverview.sub)))
|
|
0
|
299 new.col.x = c(sum(NToverview.sub$A), sum(NToverview.sub$C), sum(NToverview.sub$T), sum(NToverview.sub$G))
|
|
|
300 new.col.y = sum(new.col.x)
|
|
|
301 new.col.z = round(new.col.x / new.col.y * 100, 2)
|
|
|
302
|
|
|
303 tmp = names(NTresult)
|
|
|
304 NTresult = cbind(NTresult, data.frame(new.col.x, new.col.y, new.col.z))
|
|
|
305 names(NTresult) = c(tmp, paste(clazz, c("x", "y", "z"), sep=""))
|
|
|
306 }
|
|
|
307
|
|
|
308 write.table(NToverview[,c("Sequence.ID", "best_match", "seq", "A", "C", "G", "T")], NToverview.file, quote=F, sep="\t", row.names=F, col.names=T)
|
|
|
309
|
|
|
310 NToverview = NToverview[!grepl("unmatched", NToverview$best_match),]
|
|
|
311
|
|
|
312 new.col.x = c(sum(NToverview$A), sum(NToverview$C), sum(NToverview$T), sum(NToverview$G))
|
|
|
313 new.col.y = sum(new.col.x)
|
|
|
314 new.col.z = round(new.col.x / new.col.y * 100, 2)
|
|
|
315
|
|
|
316 tmp = names(NTresult)
|
|
|
317 NTresult = cbind(NTresult, data.frame(new.col.x, new.col.y, new.col.z))
|
|
|
318 names(NTresult) = c(tmp, paste("all", c("x", "y", "z"), sep=""))
|
|
|
319
|
|
|
320 names(hotspot.analysis.sum) = names(NTresult)
|
|
|
321
|
|
|
322 hotspot.analysis.sum = rbind(hotspot.analysis.sum, NTresult)
|
|
|
323
|
|
|
324 write.table(hotspot.analysis.sum, hotspot.analysis.sum.file, quote=F, sep=",", row.names=F, col.names=F, na="0")
|
|
|
325
|
|
|
326
|
|
|
327
|
|
|
328
|
|
|
329
|
|
|
330
|
|
|
331
|
|
|
332
|
|
|
333
|
|
|
334
|
|
|
335
|
|
|
336
|
|
|
337
|
|
|
338
|
|
|
339
|
|
|
340
|
|
|
341
|
|
|
342
|
|
|
343
|
|
|
344
|
|
|
345
|
|
|
346
|
|
|
347
|
|
|
348
|
|
|
349
|
|
|
350
|
|
|
351
|
|
|
352
|
|
|
353
|
|
|
354
|
