Mercurial > repos > tomnl > create_msp

comparison create_sqlite_db.R @ 0:4b417094bf71 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/mspurity-galaxy commit 0aa10df0ec1ed71601f932cfb11d7d4d4f620d80-dirty
author tomnl
date Wed, 02 May 2018 13:09:23 -0400
parents
children 7ae48818a229
comparison
equal deleted inserted replaced
-1:000000000000 0:4b417094bf71
1 library(msPurity)
2 library(optparse)
3 library(xcms)
4 library(CAMERA)
5 print(sessionInfo())
6 print('CREATING DATABASE')
7
8
9 xset_pa_filename_fix <- function(opt, pa, xset){
10
11
12 if (!is.null(opt$mzML_files) && !is.null(opt$galaxy_names)){
13 # NOTE: Relies on the pa@fileList having the names of files given as 'names' of the variables
14 # needs to be done due to Galaxy moving the files around and screwing up any links to files
15
16 filepaths <- trimws(strsplit(opt$mzML_files, ',')[[1]])
17 filepaths <- filepaths[filepaths != ""]
18 new_names <- basename(filepaths)
19
20 galaxy_names <- trimws(strsplit(opt$galaxy_names, ',')[[1]])
21 galaxy_names <- galaxy_names[galaxy_names != ""]
22
23 nsave <- names(pa@fileList)
24 old_filenames <- basename(pa@fileList)
25 pa@fileList <- filepaths[match(names(pa@fileList), galaxy_names)]
26 names(pa@fileList) <- nsave
27
28 pa@puritydf$filename <- basename(pa@fileList[match(pa@puritydf$filename, old_filenames)])
29 pa@grped_df$filename <- basename(pa@fileList[match(pa@grped_df$filename, old_filenames)])
30 }
31
32
33 if(!all(basename(pa@fileList)==basename(xset@filepaths))){
34 if(!all(names(pa@fileList)==basename(xset@filepaths))){
35 print('FILELISTS DO NOT MATCH')
36 message('FILELISTS DO NOT MATCH')
37 quit(status = 1)
38 }else{
39 xset@filepaths <- unname(pa@fileList)
40 }
41 }
42
43
44 return(list(pa, xset))
45 }
46
47
48
49 option_list <- list(
50 make_option(c("-o", "--out_dir"), type="character"),
51 make_option("--pa", type="character"),
52 make_option("--xset_xa", type="character"),
53 make_option("--xcms_camera_option", type="character"),
54 make_option("--eic", action="store_true"),
55 make_option("--cores", default=4),
56 make_option("--mzML_files", type="character"),
57 make_option("--galaxy_names", type="character"),
58 make_option("--grp_peaklist", type="character"),
59 make_option("--db_name", type="character", default='lcms_data.sqlite'),
60 make_option("--raw_rt_columns", action="store_true")
61 )
62
63 # store options
64 opt<- parse_args(OptionParser(option_list=option_list))
65
66 loadRData <- function(rdata_path, name){
67 #loads an RData file, and returns the named xset object if it is there
68 load(rdata_path)
69 return(get(ls()[ls() == name]))
70 }
71
72 print(paste('pa', opt$pa))
73 print(opt$xset)
74 print(opt$xcms_camera_option)
75 # Requires
76 pa <- loadRData(opt$pa, 'pa')
77
78 print('TESTETSTESTETE')
79 print(pa@fileList)
80
81
82 if (opt$xcms_camera_option=='xcms'){
83 xset <- loadRData(opt$xset, 'xset')
84 fix <- xset_pa_filename_fix(opt, pa, xset)
85 pa <- fix[[1]]
86 xset <- fix[[2]]
87 xa <- NULL
88 }else{
89
90 xa <- loadRData(opt$xset, 'xa')
91 fix <- xset_pa_filename_fix(opt, pa, xa@xcmsSet)
92 pa <- fix[[1]]
93 xa@xcmsSet <- fix[[2]]
94 xset <- NULL
95 }
96
97
98
99
100
101 if(is.null(opt$grp_peaklist)){
102 grp_peaklist = NA
103 }else{
104 grp_peaklist = opt$grp_peaklist
105 }
106
107
108 print(pa@fileList)
109 print(xset@filepaths)
110
111
112 db_pth <- msPurity::create_database(pa, xset=xset, xsa=xa, out_dir=opt$out_dir,
113 grp_peaklist=grp_peaklist, db_name=opt$db_name)
114
115
116 if (!is.null(opt$eic)){
117 if (is.null(opt$raw_rt_columns)){
118 rtrawColumns <- FALSE
119 }else{
120 rtrawColumns <- TRUE
121 }
122
123 # Saves the EICS into the previously created database
124 px <- msPurity::purityX(xset, saveEIC = TRUE,
125 cores=opt$cores, sqlitePth=db_pth,
126 rtrawColumns = rtrawColumns)
127 }
128
129
130
131 con <- DBI::dbConnect(RSQLite::SQLite(), db_pth)
132
133 cmd <- paste('SELECT cpg.grpid, cpg.mz, cpg.mzmin, cpg.mzmax, cpg.rt, cpg.rtmin, cpg.rtmax, c_peaks.cid, ',
134 'c_peaks.mzmin AS c_peak_mzmin, c_peaks.mzmax AS c_peak_mzmax, ',
135 'c_peaks.rtmin AS c_peak_rtmin, c_peaks.rtmax AS c_peak_rtmax, s_peak_meta.*, fileinfo.filename, fileinfo.nm_save ',
136 'FROM c_peak_groups AS cpg ',
137 'LEFT JOIN c_peak_X_c_peak_group AS cXg ON cXg.grpid=cpg.grpid ',
138 'LEFT JOIN c_peaks on c_peaks.cid=cXg.cid ',
139 'LEFT JOIN c_peak_X_s_peak_meta AS cXs ON cXs.cid=c_peaks.cid ',
140 'LEFT JOIN s_peak_meta ON cXs.pid=s_peak_meta.pid ',
141 'LEFT JOIN fileinfo ON s_peak_meta.fileid=fileinfo.fileid')
142
143 print(cmd)
144 cpeakgroup_msms <- DBI::dbGetQuery(con, cmd)
145
146 write.table(cpeakgroup_msms, file.path(opt$out_dir, 'cpeakgroup_msms.tsv'), row.names=FALSE, sep='\t')