Mercurial > repos > davidvanzessen > mutation_analysis

comparison pattern_plots.r @ 0:8a5a2abbb870 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author davidvanzessen
date Mon, 29 Aug 2016 05:36:10 -0400
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:8a5a2abbb870
1 library(ggplot2)
2 library(reshape2)
3 library(scales)
4
5 args <- commandArgs(trailingOnly = TRUE)
6
7 input.file = args[1] #the data that's get turned into the "SHM overview" table in the html report "data_sum.txt"
8
9 plot1.path = args[2]
10 plot1.png = paste(plot1.path, ".png", sep="")
11 plot1.txt = paste(plot1.path, ".txt", sep="")
12
13 plot2.path = args[3]
14 plot2.png = paste(plot2.path, ".png", sep="")
15 plot2.txt = paste(plot2.path, ".txt", sep="")
16
17 plot3.path = args[4]
18 plot3.png = paste(plot3.path, ".png", sep="")
19 plot3.txt = paste(plot3.path, ".txt", sep="")
20
21 dat = read.table(input.file, header=F, sep=",", quote="", stringsAsFactors=F, fill=T, row.names=1)
22
23
24
25 classes = c("ca", "ca1", "ca2", "cg", "cg1", "cg2", "cg3", "cg4", "cm")
26 xyz = c("x", "y", "z")
27 new.names = c(paste(rep(classes, each=3), xyz, sep="."), paste("un", xyz, sep="."), paste("all", xyz, sep="."))
28
29 names(dat) = new.names
30
31 dat["RGYW.WRCY",] = colSums(dat[c(13,14),])
32 dat["TW.WA",] = colSums(dat[c(15,16),])
33
34 data1 = dat[c("RGYW.WRCY", "TW.WA"),]
35
36 data1 = data1[,names(data1)[grepl(".z", names(data1))]]
37 names(data1) = gsub("\\..*", "", names(data1))
38
39 data1 = melt(t(data1))
40
41 names(data1) = c("Class", "Type", "value")
42
43 write.table(data1, plot1.txt, quote=F, sep="\t", na="", row.names=F, col.names=T)
44
45 p = ggplot(data1, aes(Class, value)) + geom_bar(aes(fill=Type), stat="identity", position="dodge") + ylab("% of mutations") + guides(fill=guide_legend(title=NULL))
46 png(filename=plot1.png)
47 print(p)
48 dev.off()
49
50 data2 = dat[5:8,]
51
52 data2["sum",] = colSums(data2)
53
54 data2 = data2[,names(data2)[grepl("\\.x", names(data2))]]
55 names(data2) = gsub(".x", "", names(data2))
56
57 data2["A/T",] = round(colSums(data2[3:4,]) / data2["sum",] * 100, 1)
58 data2["A/T",is.nan(unlist(data2["A/T",]))] = 0
59
60 data2["G/C transversions",] = round(data2[2,] / data2["sum",] * 100, 1)
61 data2["G/C transitions",] = round(data2[1,] / data2["sum",] * 100, 1)
62
63
64 data2["G/C transversions",is.nan(unlist(data2["G/C transversions",]))] = 0
65 data2["G/C transversions",is.infinite(unlist(data2["G/C transversions",]))] = 0
66 data2["G/C transitions",is.nan(unlist(data2["G/C transitions",]))] = 0
67 data2["G/C transitions",is.infinite(unlist(data2["G/C transitions",]))] = 0
68
69 data2 = melt(t(data2[6:8,]))
70
71 names(data2) = c("Class", "Type", "value")
72
73 write.table(data2, plot2.txt, quote=F, sep="\t", na="", row.names=F, col.names=T)
74
75 p = ggplot(data2, aes(x=Class, y=value, fill=Type)) + geom_bar(position="fill", stat="identity") + scale_y_continuous(labels=percent_format()) + guides(fill=guide_legend(title=NULL)) + ylab("% of mutations")
76 png(filename=plot2.png)
77 print(p)
78 dev.off()
79
80 data3 = dat[c(5, 6, 8, 17:20),]
81 data3 = data3[,names(data3)[grepl("\\.x", names(data3))]]
82 names(data3) = gsub(".x", "", names(data3))
83
84 data3["G/C transitions",] = round(data3[1,] / (data3[5,] + data3[7,]) * 100, 1)
85
86 data3["G/C transversions",] = round(data3[2,] / (data3[5,] + data3[7,]) * 100, 1)
87
88 data3["A/T",] = round(data3[3,] / (data3[4,] + data3[6,]) * 100, 1)
89
90 data3["G/C transitions",is.nan(unlist(data3["G/C transitions",]))] = 0
91 data3["G/C transitions",is.infinite(unlist(data3["G/C transitions",]))] = 0
92
93 data3["G/C transversions",is.nan(unlist(data3["G/C transversions",]))] = 0
94 data3["G/C transversions",is.infinite(unlist(data3["G/C transversions",]))] = 0
95
96 data3["A/T",is.nan(unlist(data3["A/T",]))] = 0
97 data3["A/T",is.infinite(unlist(data3["A/T",]))] = 0
98
99 data3 = melt(t(data3[8:10,]))
100 names(data3) = c("Class", "Type", "value")
101
102 write.table(data3, plot3.txt, quote=F, sep="\t", na="", row.names=F, col.names=T)
103
104 p = ggplot(data3, aes(Class, value)) + geom_bar(aes(fill=Type), stat="identity", position="dodge") + ylab("% of nucleotides") + guides(fill=guide_legend(title=NULL))
105 png(filename=plot3.png)
106 print(p)
107 dev.off()
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139