Mercurial > repos > prog > lcmsmatching

comparison UniProtCompound.R @ 0:e66bb061af06 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
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('UniprotCompound')) { # Do not load again if already loaded
2
3 source('BiodbEntry.R')
4
5 #####################
6 # CLASS DECLARATION #
7 #####################
8
9 UniprotCompound <- setRefClass("UniprotCompound", contains = "BiodbEntry")
10
11 ###########
12 # FACTORY #
13 ###########
14
15 createUniprotCompoundFromXml <- function(contents, drop = FALSE) {
16
17 library(XML)
18
19 # Set XML namespace
20 ns <- c(uniprot = "http://uniprot.org/uniprot")
21
22 compounds <- list()
23
24 # Define xpath expressions
25 xpath.values <- character()
26 xpath.values[[RBIODB.NAME]] <- "/uniprot:uniprot/uniprot:compound/uniprot:name"
27 xpath.values[[RBIODB.GENE.SYMBOLS]] <- "//uniprot:gene/uniprot:name"
28 xpath.values[[RBIODB.FULLNAMES]] <- "//uniprot:protein//uniprot:fullName"
29 xpath.values[[RBIODB.SEQUENCE]] <- "//uniprot:entry/uniprot:sequence"
30 xpath.values[[RBIODB.ACCESSION]] <- "//uniprot:accession[1]"
31 xpath.attr <- list()
32 xpath.attr[[RBIODB.KEGG.ID]] <- list(path = "//uniprot:dbReference[@type='KEGG']", attr = 'id')
33 xpath.attr[[RBIODB.NCBI.GENE.ID]] <- list(path = "//uniprot:dbReference[@type='GeneID']", attr = 'id')
34 xpath.attr[[RBIODB.ENZYME.ID]] <- list(path = "//uniprot:dbReference[@type='EC']", attr = 'id')
35 xpath.attr[[RBIODB.MASS]] <- list(path = "//uniprot:entry/uniprot:sequence", attr = 'mass')
36 xpath.attr[[RBIODB.LENGTH]] <- list(path = "//uniprot:entry/uniprot:sequence", attr = 'length')
37
38 for (content in contents) {
39
40 # Create instance
41 compound <- HmdbCompound$new()
42
43 # If the entity doesn't exist (i.e.: no <id>.xml page), then it returns an HTML page
44 if ( ! grepl("^<!DOCTYPE html ", content, perl = TRUE)) {
45
46 # Parse XML
47 xml <- xmlInternalTreeParse(content, asText = TRUE)
48
49 # Test value xpath
50 for (field in names(xpath.values)) {
51 v <- xpathSApply(xml, xpath.values[[field]], xmlValue, namespaces = ns)
52 if (length(v) > 0)
53 compound$setField(field, v)
54 }
55
56 # Test attribute xpath
57 for (field in names(xpath.attr)) {
58 v <- xpathSApply(xml, xpath.attr[[field]]$path, xmlGetAttr, xpath.attr[[field]]$attr, namespaces = ns)
59 if (length(v) > 0)
60 compound$setField(field, v)
61 }
62
63 # Remove new lines from sequence string
64 seq <- compound$getField(RBIODB.SEQUENCE)
65 if ( ! is.na(seq))
66 compound$setField(RBIODB.SEQUENCE, gsub("\\n", "", seq))
67 }
68
69 compounds <- c(compounds, compound)
70 }
71
72 # Replace elements with no accession id by NULL
73 compounds <- lapply(compounds, function(x) if (is.na(x$getField(RBIODB.ACCESSION))) NULL else x)
74
75 # If the input was a single element, then output a single object
76 if (drop && length(contents) == 1)
77 compounds <- compounds[[1]]
78
79 return(compounds)
80 }
81 }