Mercurial > repos > mvdbeek > r_goseq_1_22_0

comparison goseq.r @ 0:a6427f7893b0 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/ARTbio/tools-artbio/tree/master/tools/goseq_1_22_0 commit fdd0811efc61c31f88ff17096fbe8ee8cfacd766-dirty
author mvdbeek
date Fri, 26 Feb 2016 06:57:47 -0500
parents
children 81283f3d65c7
comparison
equal deleted inserted replaced
-1:000000000000 0:a6427f7893b0
1 sink(stdout(), type = "message")
2 library(goseq)
3 library(optparse)
4
5 option_list <- list(
6 make_option(c("-d", "--dge_file"), type="character", help="Path to file with differential gene expression result"),
7 make_option(c("-w","--wallenius_tab"), type="character", help="Path to output file with P-values estimated using wallenius distribution."),
8 make_option(c("-s","--sampling_tab"), type="character", default=FALSE, help="Path to output file with P-values estimated using wallenius distribution."),
9 make_option(c("-n","--nobias_tab"), type="character", default=FALSE, help="Path to output file with P-values estimated using wallenius distribution and no correction for gene length bias."),
10 make_option(c("-l","--length_bias_plot"), type="character", default=FALSE, help="Path to length-bias plot."),
11 make_option(c("-sw","--sample_vs_wallenius_plot"), type="character", default=FALSE, help="Path to plot comparing sampling with wallenius p-values."),
12 make_option(c("-padj", "--p_adj_column"), type="integer",help="Column that contains p. adjust values"),
13 make_option(c("-c", "--cutoff"), type="double",dest="p_adj_cutoff",
14 help="Genes with p.adjust below cutoff are considered not differentially expressed and serve as control genes"),
15 make_option(c("-r", "--repcnt"), type="integer", default=100, help="Number of repeats for sampling"),
16 make_option(c("-lf", "--length_file"), default=FALSE, help = "Path to "),
17 make_option(c("-g", "--genome"), type="character", help = "Genome [used for looking up correct gene length]"),
18 make_option(c("-i", "--gene_id"), type="character", help="Gene ID of gene column in DGE file")
19 )
20 parser <- OptionParser(usage = "%prog [options] file", option_list=option_list)
21 args = parse_args(parser)
22
23 # Vars:
24 dge_file = args$dge_file
25 p_adj_column = args$p_adj_colum
26 p_adj_cutoff = args$p_adj_cutoff
27 length_file = args$length_file
28 genome = args$genome
29 gene_id = args$gene_id
30 wallenius_tab = args$wallenius_tab
31 sampling_tab = args$sampling_tab
32 nobias_tab = args$nobias_tab
33 length_bias_plot = args$length_bias_plot
34 sample_vs_wallenius_plot = args$sample_vs_wallenius_plot
35 repcnt = args$repcnt
36
37
38 # format DE genes into vector suitable for use with goseq
39 dge_table = read.delim(dge_file, header = TRUE, sep="\t", check.names = FALSE)
40 genes = as.integer(dge_table[,p_adj_column]<p_adj_cutoff)
41 names(genes) = dge_table[,1] # Assuming first row contains gene names
42
43 # Get gene lengths
44
45 if (length_file) {
46 length_table = read.delim(length_file, header=TRUE, sep="\t", check.names=FALSE, row.names=TRUE)
47 gene_lengths = length_table[names(genes),]$length
48 } else {
49 gene_lengths = getlength(names(genes), genome, gene_id)
50 }
51
52 # Estimate PWF
53
54 pdf(length_bias_plot)
55 pwf=nullp(genes, genome, gene_id, gene_lengths)
56 dev.off()
57 # wallenius approximation of p-values
58 GO.wall=goseq(pwf, genome, gene_id)
59
60 GO.nobias=goseq(pwf, genome, gene_id, method="Hypergeometric")
61
62 # Sampling distribution
63 if (repcnt > 0) {
64 GO.samp=goseq(pwf,genome, gene_id, method="Sampling", repcnt=repcnt)
65 # Compare sampling with wallenius
66 pdf(sample_vs_wallenius_plot)
67 plot(log10(GO.wall[,2]), log10(GO.samp[match(GO.samp[,1],GO.wall[,1]),2]),
68 xlab="log10(Wallenius p-values)",ylab="log10(Sampling p-values)",
69 xlim=c(-3,0))
70 abline(0,1,col=3,lty=2)
71 dev.off()
72 write.table(GO.samp, sampling_tab, sep="\t", row.names = FALSE, quote = FALSE)
73 }
74
75
76 write.table(GO.wall, wallenius_tab, sep="\t", row.names = FALSE, quote = FALSE)
77 write.table(GO.nobias, nobias_tab, sep="\t", row.names = FALSE, quote = FALSE)
78
79 sessionInfo()
80
81 # Use the following to get a list of supported genomes / gene ids
82
83 # write.table(supportedGenomes(), "available_genomes.tab", row.names = FALSE, quote=FALSE)
84 # write.table(supportedGeneIDs(), "supported_gene_ids.tab", row.name = FALSE, quote = FALSE)
85 # write.table(table.summary, "input_gene_count_matrix.tab", row.names = FALSE, quote = FALSE)