comparison tximport.R @ 0:2f5e9c0fe367 draft default tip

"planemo upload for repository https://github.com/ieguinoa/tximport-galaxy-wrapper commit 2bb25471c1320fb1206afa2c4daf536b6d6e275f-dirty"
author ieguinoa
date Wed, 09 Oct 2019 15:38:21 -0400
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:2f5e9c0fe367
1 # setup R error handling to go to stderr
2 options( show.error.messages=F, error = function () { cat( geterrmessage(), file=stderr() ); q( "no", 1, F ) } )
3
4 # we need that to not crash galaxy with an UTF8 error on German LC settings.
5 loc <- Sys.setlocale("LC_MESSAGES", "en_US.UTF-8")
6
7 library("getopt")
8 #library("tools")
9 options(stringAsFactors = FALSE, useFancyQuotes = FALSE)
10 args <- commandArgs(trailingOnly = TRUE)
11
12 # get options, using the spec as defined by the enclosed list.
13 # we read the options from the default: commandArgs(TRUE).
14 spec <- matrix(c(
15 "help", "h", 0, "logical",
16 "base_dir", "w", 1, "character",
17 "out_file", "o", 1, "character",
18 "countsFiles", "n", 1, "character",
19 "countsFromAbundance", "r", 1, "character",
20 "format", "v", 1, "character",
21 "gff_file", "H", 0, "character",
22 "tx2gene", "f", 0, "character",
23 "geneIdCol", "l", 0, "character",
24 "txIdCol" , "p", 1, "character",
25 "abundanceCol", "i", 0, "character",
26 "countsCol", "y", 1, "character",
27 "lengthCol", "x", 1, "character"),
28 byrow=TRUE, ncol=4)
29
30 opt <- getopt(spec)
31
32
33
34
35 # if help was asked for print a friendly message
36 # and exit with a non-zero error code
37 if (!is.null(opt$help)) {
38 cat(getopt(spec, usage=TRUE))
39 q(status=1)
40 }
41
42 if (is.null(opt$gff_file) & is.null(opt$tx2gene)) {
43 cat("A GFF/GTF file or a tx2gene table is required\n")
44 q(status=1)
45 }
46
47 if (opt$format == 'none'){ #custom format
48 if (is.null(opt$txIdCol) | is.null(opt$abundanceCol) | is.null(opt$countsCol) | is.null(opt$lengthCol)) {
49 cat("If you select a custom format for the input files you need to specify the column names\n")
50 q(status=1)
51 }
52 }
53
54 if (is.null(opt$countsFiles)) {
55 cat("'countsFiles' is required\n")
56 q(status=1)
57 }
58
59 ## parse counts files
60 library(rjson)
61 dat <- fromJSON(opt$countsFiles)
62 samples_df <- lapply(dat, function(samples) # Loop through each "sample"
63 {
64 # Convert each group to a data frame.
65 # This assumes you have 6 elements each time
66 data.frame(matrix(unlist(samples), ncol=2, byrow=T))
67 })
68 samples_df <- do.call(rbind, samples_df)
69 colnames(samples_df) <- c("path","id")
70 rownames(samples_df) <- NULL
71
72 # Prepare char vector with files and sample names
73 files <- file.path(samples_df[,"path"])
74 names(files) <- samples_df[,"id"]
75 #files
76 #all(file.exists(files))
77
78
79
80 library(tximport)
81
82
83
84
85 ### if the input is a gff/gtf file first need to create the tx2gene table
86 if (!is.null(opt$gff_file)) {
87 suppressPackageStartupMessages({
88 library("GenomicFeatures")
89 })
90 txdb <- makeTxDbFromGFF(opt$gff_file)
91 k <- keys(txdb, keytype = "TXNAME")
92 tx2gene <- select(txdb, keys=k, columns="GENEID", keytype="TXNAME")
93 # Remove 'transcript:' from transcript IDs (when gffFile is a GFF3 from Ensembl and the transcript does not have a Name)
94 tx2gene$TXNAME <- sub('^transcript:', '', tx2gene$TXNAME)
95
96 } else {
97 tx2gene <- read.table(opt$tx2gene,header=FALSE)
98 }
99
100
101
102 ##
103 if (is.null(opt$geneIdCol)) { ## there is a tx2gene table
104 if (opt$format == 'none'){ #predefined format
105 cat("here i am too\n")
106 txi_out <- tximport(files, type="none",txIdCol=opt$txIdCol,abundanceCol=opt$abundanceCol,countsCol=opt$countsCol,lengthCol=opt$lengthCol,tx2gene=tx2gene,countsFromAbundance=opt$countsFromAbundance)
107 } else {
108 txi_out <- tximport(files, type=opt$format, tx2gene=tx2gene,countsFromAbundance=opt$countsFromAbundance)
109 }
110 } else { # the gene_ID is a column in the counts table
111 if (opt$format == 'none'){ #predefined format
112 txi_out <- tximport(files, type="none",geneIdCol=opt$geneIdCol,txIdCol=opt$txIdCol,abundanceCol=opt$abundanceCol,countsCol=opt$countsCol,lengthCol=opt$lengthCol,tx2gene=tx2gene,countsFromAbundance=opt$countsFromAbundance)
113 } else {
114 txi_out <- tximport(files, type=opt$format, geneIdCol=opt$geneIdCol,countsFromAbundance=opt$countsFromAbundance)
115 }
116
117 }
118 # write count as table
119 write.table(txi_out$counts, file=opt$out_file, row.names = TRUE, col.names = TRUE, quote = FALSE, sep = "\t")