Mercurial > repos > prog > lcmsmatching
comparison KeggCompound.R @ 0:e66bb061af06 draft
planemo upload for repository https://github.com/workflow4metabolomics/lcmsmatching.git commit 3529b25417f8e1a5836474c9adec4b696d35099d-dirty
| author | prog |
|---|---|
| date | Tue, 12 Jul 2016 12:02:37 -0400 |
| parents | |
| children | 253d531a0193 |
comparison
equal
deleted
inserted
replaced
| -1:000000000000 | 0:e66bb061af06 |
|---|---|
| 1 if ( ! exists('KeggCompound')) { # Do not load again if already loaded | |
| 2 | |
| 3 source('BiodbEntry.R') | |
| 4 | |
| 5 ##################### | |
| 6 # CLASS DECLARATION # | |
| 7 ##################### | |
| 8 | |
| 9 KeggCompound <- setRefClass("KeggCompound", contains = 'BiodbEntry') | |
| 10 | |
| 11 ########### | |
| 12 # FACTORY # | |
| 13 ########### | |
| 14 | |
| 15 createKeggCompoundFromTxt <- function(contents, drop = TRUE) { | |
| 16 | |
| 17 library(stringr) | |
| 18 | |
| 19 compounds <- list() | |
| 20 | |
| 21 # Define fields regex | |
| 22 regex <- character() | |
| 23 regex[[RBIODB.NAME]] <- "^NAME\\s+([^,;]+)" | |
| 24 regex[[RBIODB.CHEBI.ID]] <- "^\\s+ChEBI:\\s+(\\S+)" | |
| 25 regex[[RBIODB.LIPIDMAPS.ID]] <- "^\\s+LIPIDMAPS:\\s+(\\S+)" | |
| 26 | |
| 27 for (text in contents) { | |
| 28 | |
| 29 # Create instance | |
| 30 compound <- KeggCompound$new() | |
| 31 | |
| 32 lines <- strsplit(text, "\n") | |
| 33 for (s in lines[[1]]) { | |
| 34 | |
| 35 # Test generic regex | |
| 36 parsed <- FALSE | |
| 37 for (field in names(regex)) { | |
| 38 g <- str_match(s, regex[[field]]) | |
| 39 if ( ! is.na(g[1,1])) { | |
| 40 compound$setField(field, g[1,2]) | |
| 41 parsed <- TRUE | |
| 42 break | |
| 43 } | |
| 44 } | |
| 45 if (parsed) | |
| 46 next | |
| 47 | |
| 48 # ACCESSION | |
| 49 { | |
| 50 # ENZYME ID | |
| 51 g <- str_match(s, "^ENTRY\\s+EC\\s+(\\S+)") | |
| 52 if ( ! is.na(g[1,1])) | |
| 53 compound$setField(RBIODB.ACCESSION, paste('ec', g[1,2], sep = ':')) | |
| 54 | |
| 55 # ENTRY ID | |
| 56 else { | |
| 57 g <- str_match(s, "^ENTRY\\s+(\\S+)\\s+Compound") | |
| 58 if ( ! is.na(g[1,1])) | |
| 59 compound$setField(RBIODB.ACCESSION, paste('cpd', g[1,2], sep = ':')) | |
| 60 | |
| 61 # OTHER ID | |
| 62 else { | |
| 63 g <- str_match(s, "^ENTRY\\s+(\\S+)") | |
| 64 if ( ! is.na(g[1,1])) | |
| 65 compound$setField(RBIODB.ACCESSION, g[1,2]) | |
| 66 } | |
| 67 } | |
| 68 | |
| 69 # ORGANISM | |
| 70 g <- str_match(s, "^ORGANISM\\s+(\\S+)") | |
| 71 if ( ! is.na(g[1,1])) | |
| 72 compound$setField(RBIODB.ACCESSION, paste(g[1,2], compound$getField(RBIODB.ACCESSION), sep = ':')) | |
| 73 } | |
| 74 } | |
| 75 | |
| 76 compounds <- c(compounds, compound) | |
| 77 } | |
| 78 | |
| 79 # Replace elements with no accession id by NULL | |
| 80 compounds <- lapply(compounds, function(x) if (is.na(x$getField(RBIODB.ACCESSION))) NULL else x) | |
| 81 | |
| 82 # If the input was a single element, then output a single object | |
| 83 if (drop && length(contents) == 1) | |
| 84 compounds <- compounds[[1]] | |
| 85 | |
| 86 return(compounds) | |
| 87 } | |
| 88 } |
