Mercurial > repos > davidvanzessen > complete_immunerepertoire_igg

comparison RScript.r @ 7:a9053212a462 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Mon, 05 Jan 2015 09:30:08 -0500
parents
children 043fd6613fd9
comparison
equal deleted inserted replaced
6:8b46fca04595 7:a9053212a462
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 # ---------------------- parameters ----------------------
27
28 args <- commandArgs(trailingOnly = TRUE)
29
30 infile = args[1] #path to input file
31 outfile = args[2] #path to output file
32 outdir = args[3] #path to output folder (html/images/data)
33 clonaltype = args[4] #clonaltype definition, or 'none' for no unique filtering
34 species = args[5] #human or mouse
35 locus = args[6] # IGH, IGK, IGL, TRB, TRA, TRG or TRD
36 filterproductive = ifelse(args[7] == "yes", T, F) #should unproductive sequences be filtered out? (yes/no)
37
38 # ---------------------- Data preperation ----------------------
39
40 inputdata = read.table(infile, sep="\t", header=TRUE, fill=T, comment.char="")
41
42 setwd(outdir)
43
44 # remove weird rows
45 inputdata = inputdata[inputdata$Sample != "",]
46
47 #remove the allele from the V,D and J genes
48 inputdata$Top.V.Gene = gsub("[*]([0-9]+)", "", inputdata$Top.V.Gene)
49 inputdata$Top.D.Gene = gsub("[*]([0-9]+)", "", inputdata$Top.D.Gene)
50 inputdata$Top.J.Gene = gsub("[*]([0-9]+)", "", inputdata$Top.J.Gene)
51 inputdata$clonaltype = 1:nrow(inputdata)
52 PRODF = inputdata
53 if(filterproductive){
54 if("Functionality" %in% colnames(inputdata)) { # "Functionality" is an IMGT column
55 PRODF = inputdata[inputdata$Functionality == "productive" | inputdata$Functionality == "productive (see comment)", ]
56 } else {
57 PRODF = inputdata[inputdata$VDJ.Frame != "In-frame with stop codon" & inputdata$VDJ.Frame != "Out-of-frame" & inputdata$CDR3.Found.How != "NOT_FOUND" , ]
58 }
59 }
60
61 #remove duplicates based on the clonaltype
62 if(clonaltype != "none"){
63 PRODF$clonaltype = do.call(paste, c(PRODF[unlist(strsplit(clonaltype, ","))], sep = ":"))
64 PRODF = PRODF[!duplicated(PRODF$clonaltype), ]
65 }
66
67 PRODF$freq = 1
68
69 if(any(grepl(pattern="_", x=PRODF$ID))){ #the frequency can be stored in the ID with the pattern ".*_freq_.*"
70 PRODF$freq = gsub("^[0-9]+_", "", PRODF$ID)
71 PRODF$freq = gsub("_.*", "", PRODF$freq)
72 PRODF$freq = as.numeric(PRODF$freq)
73 if(any(is.na(PRODF$freq))){ #if there was an "_" in the ID, but not the frequency, go back to frequency of 1 for every sequence
74 PRODF$freq = 1
75 }
76 }
77
78
79
80 #write the complete dataset that is left over, will be the input if 'none' for clonaltype and 'no' for filterproductive
81 write.table(PRODF, "allUnique.csv", sep=",",quote=F,row.names=F,col.names=T)
82
83 #write the samples to a file
84 sampleFile <- file("samples.txt")
85 un = unique(inputdata$Sample)
86 un = paste(un, sep="\n")
87 writeLines(un, sampleFile)
88 close(sampleFile)
89
90 # ---------------------- Counting the productive/unproductive and unique sequences ----------------------
91
92 inputdata.dt = data.table(inputdata) #for speed
93
94 ct = unlist(strsplit(clonaltype, ","))
95 if(clonaltype == "none"){
96 ct = c("ID")
97 }
98
99 inputdata.dt$samples_replicates = paste(inputdata.dt$Sample, inputdata.dt$Replicate, sep="_")
100 samples_replicates = c(unique(inputdata.dt$samples_replicates), unique(as.character(inputdata.dt$Sample)))
101 frequency_table = data.frame(ID = samples_replicates[order(samples_replicates)])
102
103
104 sample_productive_count = inputdata.dt[, list(All=.N,
105 Productive = nrow(.SD[.SD$Functionality == "productive" | .SD$Functionality == "productive (see comment)",]),
106 perc_prod = 1,
107 Productive_unique = nrow(.SD[.SD$Functionality == "productive" | .SD$Functionality == "productive (see comment)",list(count=.N),by=ct]),
108 perc_prod_un = 1,
109 Unproductive= nrow(.SD[.SD$Functionality != "productive" & .SD$Functionality != "productive (see comment)",]),
110 perc_unprod = 1,
111 Unproductive_unique =nrow(.SD[.SD$Functionality != "productive" & .SD$Functionality != "productive (see comment)",list(count=.N),by=ct]),
112 perc_unprod_un = 1),
113 by=c("Sample")]
114
115 sample_productive_count$perc_prod = round(sample_productive_count$Productive / sample_productive_count$All * 100)
116 sample_productive_count$perc_prod_un = round(sample_productive_count$Productive_unique / sample_productive_count$All * 100)
117
118 sample_productive_count$perc_unprod = round(sample_productive_count$Unproductive / sample_productive_count$All * 100)
119 sample_productive_count$perc_unprod_un = round(sample_productive_count$Unproductive_unique / sample_productive_count$All * 100)
120
121
122 sample_replicate_productive_count = inputdata.dt[, list(All=.N,
123 Productive = nrow(.SD[.SD$Functionality == "productive" | .SD$Functionality == "productive (see comment)",]),
124 perc_prod = 1,
125 Productive_unique = nrow(.SD[.SD$Functionality == "productive" | .SD$Functionality == "productive (see comment)",list(count=.N),by=ct]),
126 perc_prod_un = 1,
127 Unproductive= nrow(.SD[.SD$Functionality != "productive" & .SD$Functionality != "productive (see comment)",]),
128 perc_unprod = 1,
129 Unproductive_unique =nrow(.SD[.SD$Functionality != "productive" & .SD$Functionality != "productive (see comment)",list(count=.N),by=ct]),
130 perc_unprod_un = 1),
131 by=c("samples_replicates")]
132
133 sample_replicate_productive_count$perc_prod = round(sample_replicate_productive_count$Productive / sample_replicate_productive_count$All * 100)
134 sample_replicate_productive_count$perc_prod_un = round(sample_replicate_productive_count$Productive_unique / sample_replicate_productive_count$All * 100)
135
136 sample_replicate_productive_count$perc_unprod = round(sample_replicate_productive_count$Unproductive / sample_replicate_productive_count$All * 100)
137 sample_replicate_productive_count$perc_unprod_un = round(sample_replicate_productive_count$Unproductive_unique / sample_replicate_productive_count$All * 100)
138
139 setnames(sample_replicate_productive_count, colnames(sample_productive_count))
140
141 counts = rbind(sample_replicate_productive_count, sample_productive_count)
142 counts = counts[order(counts$Sample),]
143
144 write.table(x=counts, file="productive_counting.txt", sep=",",quote=F,row.names=F,col.names=F)
145
146 # ---------------------- Frequency calculation for V, D and J ----------------------
147
148 PRODFV = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.V.Gene")])
149 Total = ddply(PRODFV, .(Sample), function(x) data.frame(Total = sum(x$Length)))
150 PRODFV = merge(PRODFV, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
151 PRODFV = ddply(PRODFV, c("Sample", "Top.V.Gene"), summarise, relFreq= (Length*100 / Total))
152
153 PRODFD = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.D.Gene")])
154 Total = ddply(PRODFD, .(Sample), function(x) data.frame(Total = sum(x$Length)))
155 PRODFD = merge(PRODFD, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
156 PRODFD = ddply(PRODFD, c("Sample", "Top.D.Gene"), summarise, relFreq= (Length*100 / Total))
157
158 PRODFJ = data.frame(data.table(PRODF)[, list(Length=sum(freq)), by=c("Sample", "Top.J.Gene")])
159 Total = ddply(PRODFJ, .(Sample), function(x) data.frame(Total = sum(x$Length)))
160 PRODFJ = merge(PRODFJ, Total, by.x='Sample', by.y='Sample', all.x=TRUE)
161 PRODFJ = ddply(PRODFJ, c("Sample", "Top.J.Gene"), summarise, relFreq= (Length*100 / Total))
162
163 # ---------------------- Setting up the gene names for the different T/B, human/mouse and locus ----------------------
164
165 V = c("v.name\tchr.orderV\n")
166 D = c("v.name\tchr.orderD\n")
167 J = c("v.name\tchr.orderJ\n")
168
169 if(species == "human"){
170 if(locus == "trb"){
171 V = c("v.name\tchr.orderV\nTRBV2\t1\nTRBV3-1\t2\nTRBV4-1\t3\nTRBV5-1\t4\nTRBV6-1\t5\nTRBV4-2\t6\nTRBV6-2\t7\nTRBV4-3\t8\nTRBV6-3\t9\nTRBV7-2\t10\nTRBV6-4\t11\nTRBV7-3\t12\nTRBV9\t13\nTRBV10-1\t14\nTRBV11-1\t15\nTRBV10-2\t16\nTRBV11-2\t17\nTRBV6-5\t18\nTRBV7-4\t19\nTRBV5-4\t20\nTRBV6-6\t21\nTRBV5-5\t22\nTRBV7-6\t23\nTRBV5-6\t24\nTRBV6-8\t25\nTRBV7-7\t26\nTRBV6-9\t27\nTRBV7-8\t28\nTRBV5-8\t29\nTRBV7-9\t30\nTRBV13\t31\nTRBV10-3\t32\nTRBV11-3\t33\nTRBV12-3\t34\nTRBV12-4\t35\nTRBV12-5\t36\nTRBV14\t37\nTRBV15\t38\nTRBV16\t39\nTRBV18\t40\nTRBV19\t41\nTRBV20-1\t42\nTRBV24-1\t43\nTRBV25-1\t44\nTRBV27\t45\nTRBV28\t46\nTRBV29-1\t47\nTRBV30\t48")
172 D = c("v.name\tchr.orderD\nTRBD1\t1\nTRBD2\t2\n")
173 J = c("v.name\tchr.orderJ\nTRBJ1-1\t1\nTRBJ1-2\t2\nTRBJ1-3\t3\nTRBJ1-4\t4\nTRBJ1-5\t5\nTRBJ1-6\t6\nTRBJ2-1\t7\nTRBJ2-2\t8\nTRBJ2-3\t9\nTRBJ2-4\t10\nTRBJ2-5\t11\nTRBJ2-6\t12\nTRBJ2-7\t13")
174 } else if (locus == "tra"){
175 V = c("v.name\tchr.orderVTRAV1-1\t1\nTRAV1-2\t2\nTRAV2\t3\nTRAV3\t4\nTRAV4\t5\nTRAV5\t6\nTRAV6\t7\nTRAV7\t8\nTRAV8-1\t9\nTRAV9-1\t10\nTRAV10\t11\nTRAV12-1\t12\nTRAV8-2\t13\nTRAV8-3\t14\nTRAV13-1\t15\nTRAV12-2\t16\nTRAV8-4\t17\nTRAV13-2\t18\nTRAV14/DV4\t19\nTRAV9-2\t20\nTRAV12-3\t21\nTRAV8-6\t22\nTRAV16\t23\nTRAV17\t24\nTRAV18\t25\nTRAV19\t26\nTRAV20\t27\nTRAV21\t28\nTRAV22\t29\nTRAV23/DV6\t30\nTRAV24\t31\nTRAV25\t32\nTRAV26-1\t33\nTRAV27\t34\nTRAV29/DV5\t35\nTRAV30\t36\nTRAV26-2\t37\nTRAV34\t38\nTRAV35\t39\nTRAV36/DV7\t40\nTRAV38-1\t41\nTRAV38-2/DV8\t42\nTRAV39\t43\nTRAV40\t44\nTRAV41\t45\n")
176 D = c("v.name\tchr.orderD\n")
177 J = c("v.name\tchr.orderJ\nTRAJ57\t1\nTRAJ56\t2\nTRAJ54\t3\nTRAJ53\t4\nTRAJ52\t5\nTRAJ50\t6\nTRAJ49\t7\nTRAJ48\t8\nTRAJ47\t9\nTRAJ46\t10\nTRAJ45\t11\nTRAJ44\t12\nTRAJ43\t13\nTRAJ42\t14\nTRAJ41\t15\nTRAJ40\t16\nTRAJ39\t17\nTRAJ38\t18\nTRAJ37\t19\nTRAJ36\t20\nTRAJ34\t21\nTRAJ33\t22\nTRAJ32\t23\nTRAJ31\t24\nTRAJ30\t25\nTRAJ29\t26\nTRAJ28\t27\nTRAJ27\t28\nTRAJ26\t29\nTRAJ24\t30\nTRAJ23\t31\nTRAJ22\t32\nTRAJ21\t33\nTRAJ20\t34\nTRAJ18\t35\nTRAJ17\t36\nTRAJ16\t37\nTRAJ15\t38\nTRAJ14\t39\nTRAJ13\t40\nTRAJ12\t41\nTRAJ11\t42\nTRAJ10\t43\nTRAJ9\t44\nTRAJ8\t45\nTRAJ7\t46\nTRAJ6\t47\nTRAJ5\t48\nTRAJ4\t49\nTRAJ3\t50")
178 } else if (locus == "trg"){
179 V = c("v.name\tchr.orderV\nTRGV9\t1\nTRGV8\t2\nTRGV5\t3\nTRGV4\t4\nTRGV3\t5\nTRGV2\t6")
180 D = c("v.name\tchr.orderD\n")
181 J = c("v.name\tchr.orderJ\nTRGJ2\t1\nTRGJP2\t2\nTRGJ1\t3\nTRGJP1\t4")
182 } else if (locus == "trd"){
183 V = c("v.name\tchr.orderV\nTRDV1\t1\nTRDV2\t2\nTRDV3\t3")
184 D = c("v.name\tchr.orderD\nTRDD1\t1\nTRDD2\t2\nTRDD3\t3")
185 J = c("v.name\tchr.orderJ\nTRDJ1\t1\nTRDJ4\t2\nTRDJ2\t3\nTRDJ3\t4")
186 } else if(locus == "igh"){
187 V = c("v.name\tchr.orderV\nIGHV3-74\t1\nIGHV3-73\t2\nIGHV3-72\t3\nIGHV2-70\t4\nIGHV1-69D\t5\nIGHV1-69-2\t6\nIGHV2-70D\t7\nIGHV1-69\t8\nIGHV3-66\t9\nIGHV3-64\t10\nIGHV4-61\t11\nIGHV4-59\t12\nIGHV1-58\t13\nIGHV3-53\t14\nIGHV5-51\t15\nIGHV3-49\t16\nIGHV3-48\t17\nIGHV1-46\t18\nIGHV1-45\t19\nIGHV3-43\t20\nIGHV4-39\t21\nIGHV3-43D\t22\nIGHV4-38-2\t23\nIGHV4-34\t24\nIGHV3-33\t25\nIGHV4-31\t26\nIGHV3-30-5\t27\nIGHV4-30-4\t28\nIGHV3-30-3\t29\nIGHV4-30-2\t30\nIGHV4-30-1\t31\nIGHV3-30\t32\nIGHV4-28\t33\nIGHV2-26\t34\nIGHV1-24\t35\nIGHV3-23D\t36\nIGHV3-23\t37\nIGHV3-21\t38\nIGHV3-20\t39\nIGHV1-18\t40\nIGHV3-15\t41\nIGHV3-13\t42\nIGHV3-11\t43\nIGHV5-10-1\t44\nIGHV3-9\t45\nIGHV1-8\t46\nIGHV3-64D\t47\nIGHV3-7\t48\nIGHV2-5\t49\nIGHV7-4-1\t50\nIGHV4-4\t51\nIGHV1-3\t52\nIGHV1-2\t53\nIGHV6-1\t54")
188 D = c("v.name\tchr.orderD\nIGHD1-7\t1\nIGHD2-8\t2\nIGHD3-9\t3\nIGHD3-10\t4\nIGHD5-12\t5\nIGHD6-13\t6\nIGHD2-15\t7\nIGHD3-16\t8\nIGHD4-17\t9\nIGHD5-18\t10\nIGHD6-19\t11\nIGHD1-20\t12\nIGHD2-21\t13\nIGHD3-22\t14\nIGHD5-24\t15\nIGHD6-25\t16\nIGHD1-26\t17\nIGHD7-27\t18")
189 J = c("v.name\tchr.orderJ\nIGHJ1\t1\nIGHJ2\t2\nIGHJ3\t3\nIGHJ4\t4\nIGHJ5\t5\nIGHJ6\t6")
190 } else if (locus == "igk"){
191 V = c("v.name\tchr.orderV\nIGKV3D-7\t1\nIGKV1D-8\t2\nIGKV1D-43\t3\nIGKV3D-11\t4\nIGKV1D-12\t5\nIGKV1D-13\t6\nIGKV3D-15\t7\nIGKV1D-16\t8\nIGKV1D-17\t9\nIGKV3D-20\t10\nIGKV2D-26\t11\nIGKV2D-28\t12\nIGKV2D-29\t13\nIGKV2D-30\t14\nIGKV1D-33\t15\nIGKV1D-39\t16\nIGKV2D-40\t17\nIGKV2-40\t18\nIGKV1-39\t19\nIGKV1-33\t20\nIGKV2-30\t21\nIGKV2-29\t22\nIGKV2-28\t23\nIGKV1-27\t24\nIGKV2-24\t25\nIGKV3-20\t26\nIGKV1-17\t27\nIGKV1-16\t28\nIGKV3-15\t29\nIGKV1-13\t30\nIGKV1-12\t31\nIGKV3-11\t32\nIGKV1-9\t33\nIGKV1-8\t34\nIGKV1-6\t35\nIGKV1-5\t36\nIGKV5-2\t37\nIGKV4-1\t38")
192 D = c("v.name\tchr.orderD\n")
193 J = c("v.name\tchr.orderJ\nIGKJ1\t1\nIGKJ2\t2\nIGKJ3\t3\nIGKJ4\t4\nIGKJ5\t5")
194 } else if (locus == "igl"){
195 V = c("v.name\tchr.orderV\nIGLV4-69\t1\nIGLV8-61\t2\nIGLV4-60\t3\nIGLV6-57\t4\nIGLV5-52\t5\nIGLV1-51\t6\nIGLV9-49\t7\nIGLV1-47\t8\nIGLV7-46\t9\nIGLV5-45\t10\nIGLV1-44\t11\nIGLV7-43\t12\nIGLV1-41\t13\nIGLV1-40\t14\nIGLV5-39\t15\nIGLV5-37\t16\nIGLV1-36\t17\nIGLV3-27\t18\nIGLV3-25\t19\nIGLV2-23\t20\nIGLV3-22\t21\nIGLV3-21\t22\nIGLV3-19\t23\nIGLV2-18\t24\nIGLV3-16\t25\nIGLV2-14\t26\nIGLV3-12\t27\nIGLV2-11\t28\nIGLV3-10\t29\nIGLV3-9\t30\nIGLV2-8\t31\nIGLV4-3\t32\nIGLV3-1\t33")
196 D = c("v.name\tchr.orderD\n")
197 J = c("v.name\tchr.orderJ\nIGLJ1\t1\nIGLJ2\t2\nIGLJ3\t3\nIGLJ6\t4\nIGLJ7\t5")
198 }
199 } else if (species == "mouse"){
200 if(locus == "trb"){
201 V = c("v.name\tchr.orderV\nTRBV1\t1\nTRBV2\t2\nTRBV3\t3\nTRBV4\t4\nTRBV5\t5\nTRBV12-1\t6\nTRBV13-1\t7\nTRBV12-2\t8\nTRBV13-2\t9\nTRBV13-3\t10\nTRBV14\t11\nTRBV15\t12\nTRBV16\t13\nTRBV17\t14\nTRBV19\t15\nTRBV20\t16\nTRBV23\t17\nTRBV24\t18\nTRBV26\t19\nTRBV29\t20\nTRBV30\t21\nTRBV31\t22")
202 D = c("v.name\tchr.orderD\nTRBD1\t1\nTRBD2\t2")
203 J = c("v.name\tchr.orderJ\nTRBJ1-1\t1\nTRBJ1-2\t2\nTRBJ1-3\t3\nTRBJ1-4\t4\nTRBJ1-5\t5\nTRBJ2-1\t6\nTRBJ2-2\t7\nTRBJ2-3\t8\nTRBJ2-4\t9\nTRBJ2-5\t10\nTRBJ2-6\t11\nTRBJ2-7\t12")
204 } else if (locus == "tra"){
205 cat("mouse tra not yet implemented")
206 } else if (locus == "trg"){
207 cat("mouse trg not yet implemented")
208 } else if (locus == "trd"){
209 cat("mouse trd not yet implemented")
210 } else if(locus == "igh"){
211 cat("mouse igh not yet implemented")
212 } else if (locus == "igk"){
213 cat("mouse igk not yet implemented")
214 } else if (locus == "igl"){
215 cat("mouse igl not yet implemented")
216 }
217 }
218
219 useD = TRUE
220 if(species == "human" && locus == "tra"){
221 useD = FALSE
222 cat("No D Genes in this species/locus")
223 }
224
225 # ---------------------- load the gene names into a data.frame and merge with the frequency count ----------------------
226
227 tcV = textConnection(V)
228 Vchain = read.table(tcV, sep="\t", header=TRUE)
229 PRODFV = merge(PRODFV, Vchain, by.x='Top.V.Gene', by.y='v.name', all.x=TRUE)
230 close(tcV)
231
232 tcD = textConnection(D)
233 Dchain = read.table(tcD, sep="\t", header=TRUE)
234 PRODFD = merge(PRODFD, Dchain, by.x='Top.D.Gene', by.y='v.name', all.x=TRUE)
235 close(tcD)
236
237 tcJ = textConnection(J)
238 Jchain = read.table(tcJ, sep="\t", header=TRUE)
239 PRODFJ = merge(PRODFJ, Jchain, by.x='Top.J.Gene', by.y='v.name', all.x=TRUE)
240 close(tcJ)
241
242 # ---------------------- Create the V, D and J frequency plots and write the data.frame for every plot to a file ----------------------
243
244 pV = ggplot(PRODFV)
245 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))
246 pV = pV + xlab("Summary of V gene") + ylab("Frequency") + ggtitle("Relative frequency of V gene usage")
247 write.table(x=PRODFV, file="VFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
248
249 png("VPlot.png",width = 1280, height = 720)
250 pV
251 dev.off();
252
253 if(useD){
254 pD = ggplot(PRODFD)
255 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))
256 pD = pD + xlab("Summary of D gene") + ylab("Frequency") + ggtitle("Relative frequency of D gene usage")
257 write.table(x=PRODFD, file="DFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
258
259 png("DPlot.png",width = 800, height = 600)
260 print(pD)
261 dev.off();
262 }
263
264 pJ = ggplot(PRODFJ)
265 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))
266 pJ = pJ + xlab("Summary of J gene") + ylab("Frequency") + ggtitle("Relative frequency of J gene usage")
267 write.table(x=PRODFJ, file="JFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
268
269 png("JPlot.png",width = 800, height = 600)
270 pJ
271 dev.off();
272
273 pJ = ggplot(PRODFJ)
274 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))
275 pJ = pJ + xlab("Summary of J gene") + ylab("Frequency") + ggtitle("Relative frequency of J gene usage")
276 write.table(x=PRODFJ, file="JFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
277
278 png("JPlot.png",width = 800, height = 600)
279 pJ
280 dev.off();
281
282 # ---------------------- Now the frequency plots of the V, D and J families ----------------------
283
284 VGenes = PRODF[,c("Sample", "Top.V.Gene")]
285 VGenes$Top.V.Gene = gsub("-.*", "", VGenes$Top.V.Gene)
286 VGenes = data.frame(data.table(VGenes)[, list(Count=.N), by=c("Sample", "Top.V.Gene")])
287 TotalPerSample = data.frame(data.table(VGenes)[, list(total=sum(.SD$Count)), by=Sample])
288 VGenes = merge(VGenes, TotalPerSample, by="Sample")
289 VGenes$Frequency = VGenes$Count * 100 / VGenes$total
290 VPlot = ggplot(VGenes)
291 VPlot = VPlot + geom_bar(aes( x = Top.V.Gene, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
292 ggtitle("Distribution of V gene families") +
293 ylab("Percentage of sequences")
294 png("VFPlot.png")
295 VPlot
296 dev.off();
297 write.table(x=VGenes, file="VFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
298
299 if(useD){
300 DGenes = PRODF[,c("Sample", "Top.D.Gene")]
301 DGenes$Top.D.Gene = gsub("-.*", "", DGenes$Top.D.Gene)
302 DGenes = data.frame(data.table(DGenes)[, list(Count=.N), by=c("Sample", "Top.D.Gene")])
303 TotalPerSample = data.frame(data.table(DGenes)[, list(total=sum(.SD$Count)), by=Sample])
304 DGenes = merge(DGenes, TotalPerSample, by="Sample")
305 DGenes$Frequency = DGenes$Count * 100 / DGenes$total
306 DPlot = ggplot(DGenes)
307 DPlot = DPlot + geom_bar(aes( x = Top.D.Gene, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
308 ggtitle("Distribution of D gene families") +
309 ylab("Percentage of sequences")
310 png("DFPlot.png")
311 print(DPlot)
312 dev.off();
313 write.table(x=DGenes, file="DFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
314 }
315
316 JGenes = PRODF[,c("Sample", "Top.J.Gene")]
317 JGenes$Top.J.Gene = gsub("-.*", "", JGenes$Top.J.Gene)
318 JGenes = data.frame(data.table(JGenes)[, list(Count=.N), by=c("Sample", "Top.J.Gene")])
319 TotalPerSample = data.frame(data.table(JGenes)[, list(total=sum(.SD$Count)), by=Sample])
320 JGenes = merge(JGenes, TotalPerSample, by="Sample")
321 JGenes$Frequency = JGenes$Count * 100 / JGenes$total
322 JPlot = ggplot(JGenes)
323 JPlot = JPlot + geom_bar(aes( x = Top.J.Gene, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
324 ggtitle("Distribution of J gene families") +
325 ylab("Percentage of sequences")
326 png("JFPlot.png")
327 JPlot
328 dev.off();
329 write.table(x=JGenes, file="JFFrequency.csv", sep=",",quote=F,row.names=F,col.names=T)
330
331 # ---------------------- Plotting the cdr3 length ----------------------
332
333 CDR3Length = data.frame(data.table(PRODF)[, list(Count=.N), by=c("Sample", "CDR3.Length.DNA")])
334 TotalPerSample = data.frame(data.table(CDR3Length)[, list(total=sum(.SD$Count)), by=Sample])
335 CDR3Length = merge(CDR3Length, TotalPerSample, by="Sample")
336 CDR3Length$Frequency = CDR3Length$Count * 100 / CDR3Length$total
337 CDR3LengthPlot = ggplot(CDR3Length)
338 CDR3LengthPlot = CDR3LengthPlot + geom_bar(aes( x = CDR3.Length.DNA, y = Frequency, fill = Sample), stat='identity', position='dodge' ) + theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
339 ggtitle("Length distribution of CDR3") +
340 xlab("CDR3 Length") +
341 ylab("Percentage of sequences")
342 png("CDR3LengthPlot.png",width = 1280, height = 720)
343 CDR3LengthPlot
344 dev.off()
345 write.table(x=CDR3Length, file="CDR3LengthPlot.csv", sep=",",quote=F,row.names=F,col.names=T)
346
347 # ---------------------- Plot the heatmaps ----------------------
348
349
350 #get the reverse order for the V and D genes
351 revVchain = Vchain
352 revDchain = Dchain
353 revVchain$chr.orderV = rev(revVchain$chr.orderV)
354 revDchain$chr.orderD = rev(revDchain$chr.orderD)
355
356 if(useD){
357 plotVD <- function(dat){
358 if(length(dat[,1]) == 0){
359 return()
360 }
361 img = ggplot() +
362 geom_tile(data=dat, aes(x=factor(reorder(Top.D.Gene, chr.orderD)), y=factor(reorder(Top.V.Gene, chr.orderV)), fill=relLength)) +
363 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
364 scale_fill_gradient(low="gold", high="blue", na.value="white") +
365 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
366 xlab("D genes") +
367 ylab("V Genes")
368
369 png(paste("HeatmapVD_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Dchain$v.name)), height=100+(15*length(Vchain$v.name)))
370 print(img)
371 dev.off()
372 write.table(x=acast(dat, Top.V.Gene~Top.D.Gene, value.var="Length"), file=paste("HeatmapVD_", unique(dat[3])[1,1], ".csv", sep=""), sep=",",quote=F,row.names=T,col.names=NA)
373 }
374
375 VandDCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.V.Gene", "Top.D.Gene", "Sample")])
376
377 VandDCount$l = log(VandDCount$Length)
378 maxVD = data.frame(data.table(VandDCount)[, list(max=max(l)), by=c("Sample")])
379 VandDCount = merge(VandDCount, maxVD, by.x="Sample", by.y="Sample", all.x=T)
380 VandDCount$relLength = VandDCount$l / VandDCount$max
381
382 cartegianProductVD = expand.grid(Top.V.Gene = Vchain$v.name, Top.D.Gene = Dchain$v.name, Sample = unique(inputdata$Sample))
383
384 completeVD = merge(VandDCount, cartegianProductVD, all.y=TRUE)
385 completeVD = merge(completeVD, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE)
386 completeVD = merge(completeVD, Dchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE)
387 VDList = split(completeVD, f=completeVD[,"Sample"])
388
389 lapply(VDList, FUN=plotVD)
390 }
391
392 plotVJ <- function(dat){
393 if(length(dat[,1]) == 0){
394 return()
395 }
396 cat(paste(unique(dat[3])[1,1]))
397 img = ggplot() +
398 geom_tile(data=dat, aes(x=factor(reorder(Top.J.Gene, chr.orderJ)), y=factor(reorder(Top.V.Gene, chr.orderV)), fill=relLength)) +
399 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
400 scale_fill_gradient(low="gold", high="blue", na.value="white") +
401 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
402 xlab("J genes") +
403 ylab("V Genes")
404
405 png(paste("HeatmapVJ_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Jchain$v.name)), height=100+(15*length(Vchain$v.name)))
406 print(img)
407 dev.off()
408 write.table(x=acast(dat, Top.V.Gene~Top.J.Gene, value.var="Length"), file=paste("HeatmapVJ_", unique(dat[3])[1,1], ".csv", sep=""), sep=",",quote=F,row.names=T,col.names=NA)
409 }
410
411 VandJCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.V.Gene", "Top.J.Gene", "Sample")])
412
413 VandJCount$l = log(VandJCount$Length)
414 maxVJ = data.frame(data.table(VandJCount)[, list(max=max(l)), by=c("Sample")])
415 VandJCount = merge(VandJCount, maxVJ, by.x="Sample", by.y="Sample", all.x=T)
416 VandJCount$relLength = VandJCount$l / VandJCount$max
417
418 cartegianProductVJ = expand.grid(Top.V.Gene = Vchain$v.name, Top.J.Gene = Jchain$v.name, Sample = unique(inputdata$Sample))
419
420 completeVJ = merge(VandJCount, cartegianProductVJ, all.y=TRUE)
421 completeVJ = merge(completeVJ, revVchain, by.x="Top.V.Gene", by.y="v.name", all.x=TRUE)
422 completeVJ = merge(completeVJ, Jchain, by.x="Top.J.Gene", by.y="v.name", all.x=TRUE)
423 VJList = split(completeVJ, f=completeVJ[,"Sample"])
424 lapply(VJList, FUN=plotVJ)
425
426 if(useD){
427 plotDJ <- function(dat){
428 if(length(dat[,1]) == 0){
429 return()
430 }
431 img = ggplot() +
432 geom_tile(data=dat, aes(x=factor(reorder(Top.J.Gene, chr.orderJ)), y=factor(reorder(Top.D.Gene, chr.orderD)), fill=relLength)) +
433 theme(axis.text.x = element_text(angle = 90, hjust = 1)) +
434 scale_fill_gradient(low="gold", high="blue", na.value="white") +
435 ggtitle(paste(unique(dat$Sample), " (N=" , sum(dat$Length, na.rm=T) ,")", sep="")) +
436 xlab("J genes") +
437 ylab("D Genes")
438
439 png(paste("HeatmapDJ_", unique(dat[3])[1,1] , ".png", sep=""), width=150+(15*length(Jchain$v.name)), height=100+(15*length(Dchain$v.name)))
440 print(img)
441 dev.off()
442 write.table(x=acast(dat, Top.D.Gene~Top.J.Gene, value.var="Length"), file=paste("HeatmapDJ_", unique(dat[3])[1,1], ".csv", sep=""), sep=",",quote=F,row.names=T,col.names=NA)
443 }
444
445
446 DandJCount = data.frame(data.table(PRODF)[, list(Length=.N), by=c("Top.D.Gene", "Top.J.Gene", "Sample")])
447
448 DandJCount$l = log(DandJCount$Length)
449 maxDJ = data.frame(data.table(DandJCount)[, list(max=max(l)), by=c("Sample")])
450 DandJCount = merge(DandJCount, maxDJ, by.x="Sample", by.y="Sample", all.x=T)
451 DandJCount$relLength = DandJCount$l / DandJCount$max
452
453 cartegianProductDJ = expand.grid(Top.D.Gene = Dchain$v.name, Top.J.Gene = Jchain$v.name, Sample = unique(inputdata$Sample))
454
455 completeDJ = merge(DandJCount, cartegianProductDJ, all.y=TRUE)
456 completeDJ = merge(completeDJ, revDchain, by.x="Top.D.Gene", by.y="v.name", all.x=TRUE)
457 completeDJ = merge(completeDJ, Jchain, by.x="Top.J.Gene", by.y="v.name", all.x=TRUE)
458 DJList = split(completeDJ, f=completeDJ[,"Sample"])
459 lapply(DJList, FUN=plotDJ)
460 }
461
462
463 # ---------------------- calculating the clonality score ----------------------
464
465 if("Replicate" %in% colnames(inputdata)) #can only calculate clonality score when replicate information is available
466 {
467 clonalityFrame = inputdata
468 if(clonaltype != "none"){
469 clonalityFrame$ReplicateConcat = paste(clonalityFrame$clonaltype, clonalityFrame$Sample, clonalityFrame$Replicate, sep = ":")
470 clonalityFrame = clonalityFrame[!duplicated(clonalityFrame$ReplicateConcat), ]
471 }
472 write.table(clonalityFrame, "clonalityComplete.csv", sep=",",quote=F,row.names=F,col.names=T)
473
474 ClonalitySampleReplicatePrint <- function(dat){
475 write.table(dat, paste("clonality_", unique(inputdata$Sample) , "_", unique(dat$Replicate), ".csv", sep=""), sep=",",quote=F,row.names=F,col.names=T)
476 }
477
478 clonalityFrameSplit = split(clonalityFrame, f=clonalityFrame[,c("Sample", "Replicate")])
479 #lapply(clonalityFrameSplit, FUN=ClonalitySampleReplicatePrint)
480
481 ClonalitySamplePrint <- function(dat){
482 write.table(dat, paste("clonality_", unique(inputdata$Sample) , ".csv", sep=""), sep=",",quote=F,row.names=F,col.names=T)
483 }
484
485 clonalityFrameSplit = split(clonalityFrame, f=clonalityFrame[,"Sample"])
486 #lapply(clonalityFrameSplit, FUN=ClonalitySamplePrint)
487
488 clonalFreq = data.frame(data.table(clonalityFrame)[, list(Type=.N), by=c("Sample", "clonaltype")])
489 clonalFreqCount = data.frame(data.table(clonalFreq)[, list(Count=.N), by=c("Sample", "Type")])
490 clonalFreqCount$realCount = clonalFreqCount$Type * clonalFreqCount$Count
491 clonalSum = data.frame(data.table(clonalFreqCount)[, list(Reads=sum(realCount)), by=c("Sample")])
492 clonalFreqCount = merge(clonalFreqCount, clonalSum, by.x="Sample", by.y="Sample")
493
494 ct = c('Type\tWeight\n2\t1\n3\t3\n4\t6\n5\t10\n6\t15')
495 tcct = textConnection(ct)
496 CT = read.table(tcct, sep="\t", header=TRUE)
497 close(tcct)
498 clonalFreqCount = merge(clonalFreqCount, CT, by.x="Type", by.y="Type", all.x=T)
499 clonalFreqCount$WeightedCount = clonalFreqCount$Count * clonalFreqCount$Weight
500
501 ReplicateReads = data.frame(data.table(clonalityFrame)[, list(Type=.N), by=c("Sample", "Replicate", "clonaltype")])
502 ReplicateReads = data.frame(data.table(ReplicateReads)[, list(Reads=.N), by=c("Sample", "Replicate")])
503 clonalFreqCount$Reads = as.numeric(clonalFreqCount$Reads)
504 ReplicateReads$squared = ReplicateReads$Reads * ReplicateReads$Reads
505
506 ReplicatePrint <- function(dat){
507 write.table(dat[-1], paste("ReplicateReads_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
508 }
509
510 ReplicateSplit = split(ReplicateReads, f=ReplicateReads[,"Sample"])
511 lapply(ReplicateSplit, FUN=ReplicatePrint)
512
513 ReplicateReads = data.frame(data.table(ReplicateReads)[, list(ReadsSum=sum(Reads), ReadsSquaredSum=sum(squared)), by=c("Sample")])
514 clonalFreqCount = merge(clonalFreqCount, ReplicateReads, by.x="Sample", by.y="Sample", all.x=T)
515
516
517 ReplicateSumPrint <- function(dat){
518 write.table(dat[-1], paste("ReplicateSumReads_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
519 }
520
521 ReplicateSumSplit = split(ReplicateReads, f=ReplicateReads[,"Sample"])
522 lapply(ReplicateSumSplit, FUN=ReplicateSumPrint)
523
524 clonalFreqCountSum = data.frame(data.table(clonalFreqCount)[, list(Numerator=sum(WeightedCount, na.rm=T)), by=c("Sample")])
525 clonalFreqCount = merge(clonalFreqCount, clonalFreqCountSum, by.x="Sample", by.y="Sample", all.x=T)
526 clonalFreqCount$ReadsSum = as.numeric(clonalFreqCount$ReadsSum) #prevent integer overflow
527 clonalFreqCount$Denominator = (((clonalFreqCount$ReadsSum * clonalFreqCount$ReadsSum) - clonalFreqCount$ReadsSquaredSum) / 2)
528 clonalFreqCount$Result = (clonalFreqCount$Numerator + 1) / (clonalFreqCount$Denominator + 1)
529
530 ClonalityScorePrint <- function(dat){
531 write.table(dat$Result, paste("ClonalityScore_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
532 }
533
534 clonalityScore = clonalFreqCount[c("Sample", "Result")]
535 clonalityScore = unique(clonalityScore)
536
537 clonalityScoreSplit = split(clonalityScore, f=clonalityScore[,"Sample"])
538 lapply(clonalityScoreSplit, FUN=ClonalityScorePrint)
539
540 clonalityOverview = clonalFreqCount[c("Sample", "Type", "Count", "Weight", "WeightedCount")]
541
542
543
544 ClonalityOverviewPrint <- function(dat){
545 write.table(dat[-1], paste("ClonalityOverView_", unique(dat[1])[1,1] , ".csv", sep=""), sep=",",quote=F,na="-",row.names=F,col.names=F)
546 }
547
548 clonalityOverviewSplit = split(clonalityOverview, f=clonalityOverview$Sample)
549 lapply(clonalityOverviewSplit, FUN=ClonalityOverviewPrint)
550 }
551
552 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")
553 if(all(imgtcolumns %in% colnames(inputdata)))
554 {
555 newData = data.frame(data.table(inputdata)[,list(unique=.N,
556 VH.DEL=mean(X3V.REGION.trimmed.nt.nb, na.rm=T),
557 P1=mean(P3V.nt.nb, na.rm=T),
558 N1=mean(N1.REGION.nt.nb, na.rm=T),
559 P2=mean(P5D.nt.nb, na.rm=T),
560 DEL.DH=mean(X5D.REGION.trimmed.nt.nb, na.rm=T),
561 DH.DEL=mean(X3D.REGION.trimmed.nt.nb, na.rm=T),
562 P3=mean(P3D.nt.nb, na.rm=T),
563 N2=mean(N2.REGION.nt.nb, na.rm=T),
564 P4=mean(P5J.nt.nb, na.rm=T),
565 DEL.JH=mean(X5J.REGION.trimmed.nt.nb, na.rm=T),
566 Total.Del=( mean(X3V.REGION.trimmed.nt.nb, na.rm=T) +
567 mean(X5D.REGION.trimmed.nt.nb, na.rm=T) +
568 mean(X3D.REGION.trimmed.nt.nb, na.rm=T) +
569 mean(X5J.REGION.trimmed.nt.nb, na.rm=T)),
570
571 Total.N=( mean(N1.REGION.nt.nb, na.rm=T) +
572 mean(N2.REGION.nt.nb, na.rm=T)),
573
574 Total.P=( mean(P3V.nt.nb, na.rm=T) +
575 mean(P5D.nt.nb, na.rm=T) +
576 mean(P3D.nt.nb, na.rm=T) +
577 mean(P5J.nt.nb, na.rm=T))),
578 by=c("Sample")])
579 write.table(newData, "junctionAnalysis.csv" , sep=",",quote=F,na="-",row.names=F,col.names=F)
580 }