Mercurial > repos > proteore > proteore_prot_features

comparison add_protein_features.R @ 9:0b46a7aead62 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit eb7450f36863f02f036cbc52bf5525d68f22bd9e
author proteore
date Tue, 18 Dec 2018 08:26:55 -0500
parents
children 8df559ad14a7
comparison
equal deleted inserted replaced
8:2e53ba3b1697 9:0b46a7aead62
1 # Read file and return file content as data.frame
2 read_file <- function(path,header){
3 file <- try(read.table(path,header=header, sep="\t",stringsAsFactors = FALSE, quote="", check.names = F),silent=TRUE)
4 if (inherits(file,"try-error")){
5 stop("File not found !")
6 }else{
7 file <- file[!apply(is.na(file) | file == "", 1, all), , drop=FALSE]
8 return(file)
9 }
10 }
11
12 order_columns <- function (df,ncol,id_type,file){
13 if (id_type=="Uniprot_AC"){ncol=ncol(file)}
14 if (ncol==1){ #already at the right position
15 return (df)
16 } else {
17 df = df[,c(2:ncol,1,(ncol+1):dim.data.frame(df)[2])]
18 }
19 return (df)
20 }
21
22 get_list_from_cp <-function(list){
23 list = strsplit(list, "[ \t\n]+")[[1]]
24 list = gsub("NA","",list)
25 list = list[list != ""] #remove empty entry
26 list = gsub("-.+", "", list) #Remove isoform accession number (e.g. "-2")
27 return(list)
28 }
29
30 get_args <- function(){
31
32 ## Collect arguments
33 args <- commandArgs(TRUE)
34
35 ## Default setting when no arguments passed
36 if(length(args) < 1) {
37 args <- c("--help")
38 }
39
40 ## Help section
41 if("--help" %in% args) {
42 cat("Selection and Annotation HPA
43 Arguments:
44 --inputtype: type of input (list of id or filename)
45 --input: input
46 --nextprot: path to nextprot information file
47 --column: the column number which you would like to apply...
48 --header: true/false if your file contains a header
49 --type: the type of input IDs (Uniprot_AC/EntrezID)
50 --pc_features: IsoPoint,SeqLength,MW
51 --localization: Chr,SubcellLocations
52 --diseases_info: Diseases
53 --output: text output filename \n")
54
55 q(save="no")
56 }
57
58 parseArgs <- function(x) strsplit(sub("^--", "", x), "=")
59 argsDF <- as.data.frame(do.call("rbind", parseArgs(args)))
60 args <- as.list(as.character(argsDF$V2))
61 names(args) <- argsDF$V1
62
63 return(args)
64 }
65
66 str2bool <- function(x){
67 if (any(is.element(c("t","true"),tolower(x)))){
68 return (TRUE)
69 }else if (any(is.element(c("f","false"),tolower(x)))){
70 return (FALSE)
71 }else{
72 return(NULL)
73 }
74 }
75
76 #take data frame, return data frame
77 split_ids_per_line <- function(line,ncol){
78
79 #print (line)
80 header = colnames(line)
81 line[ncol] = gsub("[[:blank:]]|\u00A0","",line[ncol])
82
83 if (length(unlist(strsplit(as.character(line[ncol]),";")))>1) {
84 if (length(line)==1 ) {
85 lines = as.data.frame(unlist(strsplit(as.character(line[ncol]),";")),stringsAsFactors = F)
86 } else {
87 if (ncol==1) { #first column
88 lines = suppressWarnings(cbind(unlist(strsplit(as.character(line[ncol]),";")), line[2:length(line)]))
89 } else if (ncol==length(line)) { #last column
90 lines = suppressWarnings(cbind(line[1:ncol-1],unlist(strsplit(as.character(line[ncol]),";"))))
91 } else {
92 lines = suppressWarnings(cbind(line[1:ncol-1], unlist(strsplit(as.character(line[ncol]),";"),use.names = F), line[(ncol+1):length(line)]))
93 }
94 }
95 colnames(lines)=header
96 return(lines)
97 } else {
98 return(line)
99 }
100 }
101
102 #create new lines if there's more than one id per cell in the columns in order to have only one id per line
103 one_id_one_line <-function(tab,ncol){
104
105 if (ncol(tab)>1){
106
107 tab[,ncol] = sapply(tab[,ncol],function(x) gsub("[[:blank:]]","",x))
108 header=colnames(tab)
109 res=as.data.frame(matrix(ncol=ncol(tab),nrow=0))
110 for (i in 1:nrow(tab) ) {
111 lines = split_ids_per_line(tab[i,],ncol)
112 res = rbind(res,lines)
113 }
114 }else {
115 res = unlist(sapply(tab[,1],function(x) strsplit(x,";")),use.names = F)
116 res = data.frame(res[which(!is.na(res[res!=""]))],stringsAsFactors = F)
117 colnames(res)=colnames(tab)
118 }
119 return(res)
120 }
121
122 # Get information from neXtProt
123 get_nextprot_info <- function(nextprot,input,pc_features,localization,diseases_info){
124 if(diseases_info){
125 cols = c("NextprotID",pc_features,localization,"Diseases")
126 } else {
127 cols = c("NextprotID",pc_features,localization)
128 }
129
130 cols=cols[cols!="None"]
131 info = nextprot[match(input,nextprot$NextprotID),cols]
132 return(info)
133 }
134
135 protein_features = function() {
136
137 args <- get_args()
138
139 #save(args,file="/home/dchristiany/proteore_project/ProteoRE/tools/add_human_protein_features/args.rda")
140 #load("/home/dchristiany/proteore_project/ProteoRE/tools/add_human_protein_features/args.rda")
141
142 #setting variables
143 inputtype = args$inputtype
144 if (inputtype == "copy_paste") {
145 input = get_list_from_cp(args$input)
146 file = data.frame(input,stringsAsFactors = F)
147 ncol=1
148 } else if (inputtype == "file") {
149 filename = args$input
150 ncol = args$column
151 # Check ncol
152 if (! as.numeric(gsub("c", "", ncol)) %% 1 == 0) {
153 stop("Please enter an integer for level")
154 } else {
155 ncol = as.numeric(gsub("c", "", ncol))
156 }
157
158 header = str2bool(args$header)
159 file = read_file(filename, header) # Get file content
160 if (any(grep(";",file[,ncol]))) {file = one_id_one_line(file,ncol)}
161 if (args$type == "NextprotID" && ! "NextprotID" %in% colnames(file)) { colnames(file)[ncol] <- "NextprotID"
162 } else if (args$type == "NextprotID" && "NextprotID" %in% colnames(file) && match("NextprotID",colnames(file))!=ncol ) {
163 colnames(file)[match("NextprotID",colnames(file))] <- "old_NextprotID"
164 colnames(file)[ncol] = "NextprotID"
165 }
166 }
167
168 # Read reference file
169 nextprot = read_file(args$nextprot,T)
170
171 # Parse arguments
172 id_type = args$type
173 pc_features = strsplit(args$pc_features, ",")[[1]]
174 localization = strsplit(args$localization, ",")[[1]]
175 diseases_info = str2bool(args$diseases_info)
176 output = args$output
177
178 # Change the sample ids if they are Uniprot_AC ids to be able to match them with
179 # Nextprot data
180 if (id_type=="Uniprot_AC"){
181 NextprotID = gsub("^NX_$","",gsub("^","NX_",file[,ncol]))
182 file = cbind(file,NextprotID)
183 if (inputtype=="copy_paste") {colnames(file)[1]="Uniprot-AC"}
184 ncol=ncol(file)
185 }
186 NextprotID = file[,ncol]
187
188 #Select user input protein ids in nextprot
189 #NextprotID = unique(NextprotID[which(!is.na(NextprotID[NextprotID!=""]))])
190 if (all(!NextprotID %in% nextprot[,1])){
191 write.table("None of the input ids can be found in Nextprot",file=output,sep="\t",quote=FALSE,col.names=TRUE,row.names=FALSE)
192 } else {
193 res <- get_nextprot_info(nextprot,NextprotID,pc_features,localization,diseases_info)
194 res = res[!duplicated(res$NextprotID),]
195 output_content = merge(file, res,by.x=ncol,by.y="NextprotID",incomparables = NA,all.x=T)
196 output_content = order_columns(output_content,ncol,id_type,file)
197 output_content <- as.data.frame(apply(output_content, c(1,2), function(x) gsub("^$|^ $", NA, x))) #convert "" et " " to NA
198 write.table(output_content, output, row.names = FALSE, sep = "\t", quote = FALSE)
199 }
200
201 }
202 protein_features()