Mercurial > repos > prog > lcmsmatching
comparison UniprotEntry.R @ 2:20d69a062da3 draft
planemo upload for repository https://github.com/workflow4metabolomics/lcmsmatching.git commit d4048accde6bdfd5b3e14f5394902d38991854f8
| author | prog |
|---|---|
| date | Thu, 02 Mar 2017 08:55:00 -0500 |
| parents | |
| children |
comparison
equal
deleted
inserted
replaced
| 1:253d531a0193 | 2:20d69a062da3 |
|---|---|
| 1 ##################### | |
| 2 # CLASS DECLARATION # | |
| 3 ##################### | |
| 4 | |
| 5 UniprotEntry <- methods::setRefClass("UniprotEntry", contains = "BiodbEntry") | |
| 6 | |
| 7 ########### | |
| 8 # FACTORY # | |
| 9 ########### | |
| 10 | |
| 11 createUniprotEntryFromXml <- function(contents, drop = FALSE) { | |
| 12 | |
| 13 # Set XML namespace | |
| 14 ns <- c(uniprot = "http://uniprot.org/uniprot") | |
| 15 | |
| 16 entries <- list() | |
| 17 | |
| 18 # Define xpath expressions | |
| 19 xpath.values <- character() | |
| 20 xpath.values[[BIODB.NAME]] <- "/uniprot:uniprot/uniprot:entry/uniprot:name" | |
| 21 xpath.values[[BIODB.GENE.SYMBOLS]] <- "//uniprot:gene/uniprot:name" | |
| 22 xpath.values[[BIODB.FULLNAMES]] <- "//uniprot:protein//uniprot:fullName" | |
| 23 xpath.values[[BIODB.SEQUENCE]] <- "//uniprot:entry/uniprot:sequence" | |
| 24 xpath.values[[BIODB.ACCESSION]] <- "//uniprot:accession[1]" | |
| 25 xpath.attr <- list() | |
| 26 xpath.attr[[BIODB.KEGG.ID]] <- list(path = "//uniprot:dbReference[@type='KEGG']", attr = 'id') | |
| 27 xpath.attr[[BIODB.NCBI.GENE.ID]] <- list(path = "//uniprot:dbReference[@type='GeneID']", attr = 'id') | |
| 28 xpath.attr[[BIODB.ENZYME.ID]] <- list(path = "//uniprot:dbReference[@type='EC']", attr = 'id') | |
| 29 xpath.attr[[BIODB.MASS]] <- list(path = "//uniprot:entry/uniprot:sequence", attr = 'mass') | |
| 30 xpath.attr[[BIODB.LENGTH]] <- list(path = "//uniprot:entry/uniprot:sequence", attr = 'length') | |
| 31 | |
| 32 for (content in contents) { | |
| 33 | |
| 34 # Create instance | |
| 35 entry <- UniprotEntry$new() | |
| 36 | |
| 37 # If the entity doesn't exist (i.e.: no <id>.xml page), then it returns an HTML page | |
| 38 if ( ! grepl("^<!DOCTYPE html ", content, perl = TRUE)) { | |
| 39 | |
| 40 # Parse XML | |
| 41 xml <- XML::xmlInternalTreeParse(content, asText = TRUE) | |
| 42 | |
| 43 # Test value xpath | |
| 44 for (field in names(xpath.values)) { | |
| 45 v <- XML::xpathSApply(xml, xpath.values[[field]], XML::xmlValue, namespaces = ns) | |
| 46 if (length(v) > 0) | |
| 47 entry$setField(field, v) | |
| 48 } | |
| 49 | |
| 50 # Test attribute xpath | |
| 51 for (field in names(xpath.attr)) { | |
| 52 v <- XML::xpathSApply(xml, xpath.attr[[field]]$path, XML::xmlGetAttr, xpath.attr[[field]]$attr, namespaces = ns) | |
| 53 if (length(v) > 0) | |
| 54 entry$setField(field, v) | |
| 55 } | |
| 56 | |
| 57 # Remove new lines from sequence string | |
| 58 seq <- entry$getField(BIODB.SEQUENCE) | |
| 59 if ( ! is.na(seq)) | |
| 60 entry$setField(BIODB.SEQUENCE, gsub("\\n", "", seq)) | |
| 61 } | |
| 62 | |
| 63 entries <- c(entries, entry) | |
| 64 } | |
| 65 | |
| 66 # Replace elements with no accession id by NULL | |
| 67 entries <- lapply(entries, function(x) if (is.na(x$getField(BIODB.ACCESSION))) NULL else x) | |
| 68 | |
| 69 # If the input was a single element, then output a single object | |
| 70 if (drop && length(contents) == 1) | |
| 71 entries <- entries[[1]] | |
| 72 | |
| 73 return(entries) | |
| 74 } |
