Mercurial > repos > tomnl > msp_split

comparison flag-remove-peaks.R @ 0:aa55ff1d76d4 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/computational-metabolomics/dma-tools-galaxy commit dcfc95273101a7ef0405c2efb8d83f5d456ccd15
author tomnl
date Thu, 10 May 2018 07:20:04 -0400
parents
children 89f33758ad22
comparison
equal deleted inserted replaced
-1:000000000000 0:aa55ff1d76d4
1 library(XCMSwrapper)
2 library(optparse)
3
4 option_list <- list(
5 make_option(c("-o", "--out_dir"), type="character", default=getwd(),
6 help="Output folder for resulting files [default = %default]"
7 ),
8 make_option(c("-x", "--xset_path"), type="character", default=file.path(getwd(),"xset.rds"),
9 help="The path to the xcmsSet object [default = %default]"
10 ),
11 make_option("--polarity", default='NA',
12 help="polarity (just used for naming purpose for files being saved) [positive, negative, NA] [default %default]"
13 ),
14 make_option("--rsd_i_blank", default=NA,
15 help="RSD threshold for the blank [default = %default]"
16 ),
17 make_option("--minfrac_blank", default=0.5,
18 help="minimum fraction of files for features needed for the blank [default = %default]"
19 ),
20 make_option("--rsd_rt_blank", default=NA,
21 help="RSD threshold for the RT of the blank [default = %default]"
22 ),
23
24 make_option("--ithres_blank", default=NA,
25 help="Intensity threshold for the blank [default = %default]"
26 ),
27 make_option("--s2b", default=10,
28 help="fold change (sample/blank) needed for sample peak to be allowed. e.g.
29 if s2b set to 10 and the recorded sample 'intensity' value was 100 and blank was 10.
30 1000/10 = 100, so sample has fold change higher than the threshold and the peak
31 is not considered a blank [default = %default]"
32 ),
33 make_option("--blank_class", default='blank', type="character",
34 help="A string representing the class that will be used for the blank.[default = %default]"
35 ),
36 make_option("--egauss_thr", default=NA,
37 help="Threshold for filtering out non gaussian shaped peaks. Note this only works
38 if the 'verbose columns' and 'fit gauss' was used with xcms
39 [default = %default]"
40 ),
41 make_option("--rsd_i_sample", default=NA,
42 help="RSD threshold for the samples [default = %default]"
43 ),
44 make_option("--minfrac_sample", default=0.8,
45 help="minimum fraction of files for features needed for the samples [default = %default]"
46 ),
47 make_option("--rsd_rt_sample", default=NA,
48 help="RSD threshold for the RT of the samples [default = %default]"
49 ),
50 make_option("--ithres_sample", default=5000,
51 help="Intensity threshold for the sample [default = %default]"
52 ),
53 make_option("--grp_rm_ids", default=NA,
54 help="vector of grouped_xcms peaks to remove (corresponds to the row from xcms::group output)
55 [default = %default]"
56 ),
57 make_option("--remove_spectra", action="store_true",
58 help=" TRUE if flagged spectra is to be removed [default = %default]"
59 ),
60 make_option("--minfrac_xcms", default=0.5,
61 help="minfrac for xcms grouping [default = %default]"
62 ),
63 make_option("--mzwid", default=0.001,
64 help="mzwid for xcms grouping [default = %default]"
65 ),
66 make_option("--bw", default=5,
67 help="bw for xcms grouping [default = %default]"
68 ),
69
70 make_option("--temp_save", action="store_true",
71 help="Assign True if files for each step saved (for testing purposes) [default = %default]"
72 ),
73
74 make_option("--samplelist", type="character", help="Sample list to determine the blank class"),
75
76 make_option("--xset_name", default="xset",
77 help="Name of the xcmsSet object within the RData file [default = %default]"
78 )
79
80
81
82 )
83
84 #make_option("--multilist", action="store_true"
85 # help="NOT CURRENTLY IMPLEMENTED: If paired blank removal is to be performed a - multilist - sample list file has to be provided"
86 #),
87
88 # store options
89 opt<- parse_args(OptionParser(option_list=option_list))
90
91 if (is.null(opt$temp_save)){
92 temp_save<-FALSE
93 }else{
94 temp_save<-TRUE
95 }
96
97 if (is.null(opt$remove_spectra)){
98 remove_spectra<-FALSE
99 }else{
100 remove_spectra<-TRUE
101 }
102
103
104 print(opt)
105
106 loadRData <- function(rdata_path, xset_name){
107 #loads an RData file, and returns the named xset object if it is there
108 load(rdata_path)
109 return(get(ls()[ls() == xset_name]))
110 }
111
112 xset<-loadRData(opt$xset_path, opt$xset_name)
113 print(xset)
114 if (is.null(opt$samplelist)){
115 blank_class <- opt$blank_class
116 }else{
117 samplelist <- read.table(opt$samplelist, sep='\t', header=TRUE)
118 samplelist_blank <- unique(samplelist$sample_class[samplelist$blank=='yes'])
119
120 chosen_blank <- samplelist_blank[samplelist_blank %in% xset@phenoData$class]
121 if (length(chosen_blank)>1){
122 print('ERROR: only 1 blank is currently allowed to be used with this tool')
123 exit()
124 }
125 blank_class <- as.character(chosen_blank)
126 print(blank_class)
127 }
128
129
130 if (is.null(opt$multilist)){
131 ffrm_out <- XCMSwrapper::flag_remove(xset,
132 pol=opt$polarity,
133 rsd_i_blank=opt$rsd_i_blank,
134 minfrac_blank=opt$minfrac_blank,
135 rsd_rt_blank=opt$rsd_rt_blank,
136 ithres_blank=opt$ithres_blank,
137 s2b=opt$s2b,
138 ref.class=blank_class,
139 egauss_thr=opt$egauss_thr,
140 rsd_i_sample=opt$rsd_i_sample,
141 minfrac_sample=opt$minfrac_sample,
142 rsd_rt_sample=opt$rsd_rt_sample,
143 ithres_sample=opt$ithres_sample,
144 minfrac_xcms=opt$minfrac_xcms,
145 mzwid=opt$mzwid,
146 bw=opt$bw,
147 out_dir=opt$out_dir,
148 temp_save=temp_save,
149 remove_spectra=remove_spectra,
150 grp_rm_ids=unlist(strsplit(as.character(opt$grp_rm_ids), split=", "))[[1]])
151
152 xset <- ffrm_out[[1]]
153 grp_peaklist <- ffrm_out[[2]]
154 removed_peaks <- ffrm_out[[3]]
155
156 save.image(file=file.path(opt$out_dir, 'xset_filtered.RData'))
157
158 # grpid needed for mspurity ID needed for deconrank... (will clean up at some up)
159 write.table(data.frame('grpid'=rownames(grp_peaklist), 'ID'=rownames(grp_peaklist), grp_peaklist),
160 file.path(opt$out_dir, 'peaklist_filtered.txt'), row.names=FALSE, sep='\t')
161
162 removed_peaks <- data.frame(removed_peaks)
163 write.table(data.frame('ID'=rownames(removed_peaks),removed_peaks),
164 file.path(opt$out_dir, 'removed_peaks.txt'), row.names=FALSE, sep='\t')
165
166 }else{
167
168
169 # TODO
170 #xsets <- split(xset, multilist_df$multlist)
171 #
172 #mult_grps <- unique(multilist_df$multlist)
173 #
174 #for (mgrp in mult_grps){
175 # xset_i <- xsets[mgrp]
176 # xcms::group(xset_i,
177 #
178 # }
179
180
181
182 }
183
184