Merge pull request #310 from egonw/patches/BioHackEU20

ropensci · Feb 5, 2021 · 4c63e29 · 4c63e29
2 parents 7857d26 + 5937bc8
commit 4c63e29
Show file tree

Hide file tree

Showing 4 changed files with 28 additions and 12 deletions.
diff --git a/NEWS.md b/NEWS.md
@@ -3,6 +3,8 @@
 ## BUG FIXES
 
 * Fix get_csid() so it doesn't break when a query is invalid.
+* Typo fix in Wikidata.R
+* Added support for fetching LIPID MAPS and SwissLipids identifiers from Wikidata.
 
 # webchem 1.1.0.
 

diff --git a/R/wikidata.R b/R/wikidata.R
@@ -118,7 +118,7 @@ get_wdid <-
 
 
 
-#' Retrieve Indentifiers from Wikidata
+#' Retrieve identifiers from Wikidata
 #'
 #' @import jsonlite
 #' @import httr
@@ -127,7 +127,7 @@ get_wdid <-
 #' @param verbose logical; print message during processing to console?
 #'
 #' @return A data.frame of identifiers. Currently these are 'smiles', 'cas', 'cid', 'einecs', 'csid', 'inchi', 'inchikey',
-#' 'drugbank', 'zvg', 'chebi', 'chembl', 'unii' and source_url.
+#' 'drugbank', 'zvg', 'chebi', 'chembl', 'unii', 'lipidmaps', 'swisslipids' and source_url.
 #'
 #' @note Only matches in labels are returned. If more than one unique hit is found,
 #' only the first is returned.
@@ -153,19 +153,19 @@ wd_ident <- function(id, verbose = TRUE){
   # id <- c( "Q163648", "Q18216")
   # id <- 'Q408646'
   foo <- function(id, verbose){
-    empty <- as.list(rep(NA, 13))
+    empty <- as.list(rep(NA, 15))
     names(empty) <- c("smiles", "cas", "cid", "einecs", "csid", "inchi",
                       "inchikey", "drugbank", "zvg", "chebi", "chembl", "unii",
-                      "source_url")
+                      'lipidmaps', 'swisslipids', "source_url")
     if (is.na(id)) {
       if (verbose) webchem_message("na")
       return(empty)
     }
     baseurl <- 'https://query.wikidata.org/sparql?format=json&query='
     props <- c('P233', 'P231', 'P662', 'P232', 'P661', 'P234', 'P235', 'P715', 'P679',
-               'P683', 'P592', 'P652')
+               'P683', 'P592', 'P652', 'P2063', 'P8691')
     names <- c('smiles', 'cas', 'cid', 'einecs', 'csid', 'inchi', 'inchikey',
-               'drugbank', 'zvg', 'chebi', 'chembl', 'unii')
+               'drugbank', 'zvg', 'chebi', 'chembl', 'unii', 'lipidmaps', 'swisslipids')
 
     sparql_head <- paste('PREFIX wd: <http://www.wikidata.org/entity/>',
       'PREFIX wdt: <http://www.wikidata.org/prop/direct/>',
@@ -195,7 +195,7 @@ wd_ident <- function(id, verbose = TRUE){
 
       if (length(out) == 0) {
         if (verbose) webchem_message("not_found")
-        out <- as.list(rep(NA, 13))
+        out <- as.list(rep(NA, 15))
         names(out) <- c(vars_out, 'source_url')
         return(out)
       }

diff --git a/man/wd_ident.Rd b/man/wd_ident.Rd
diff --git a/tests/testthat/test-wikidata.R b/tests/testthat/test-wikidata.R
@@ -39,7 +39,21 @@ test_that("wd_ident returns correct results", {
   expect_true(is.na(o1$smiles[4]))
   expect_equal(o1$cas[1], '50-29-3')
   expect_equal(names(o1), c('smiles', 'cas', 'cid', 'einecs', 'csid', 'inchi', 'inchikey',
-                           'drugbank', 'zvg', 'chebi', 'chembl', 'unii', 'source_url', 'query'))
+                           'drugbank', 'zvg', 'chebi', 'chembl', 'unii', 'lipidmaps', 'swisslipids', 'source_url', 'query'))
+})
+
+test_that("wd_ident returns correct results for two lipids", {
+  skip_on_cran()
+  skip_if_not(up, "Wikidata service is down")
+
+  id <- c( "Q27089367", "Q410888")
+  o1 <- wd_ident(id)
+  expect_s3_class(o1, 'data.frame')
+  expect_equal(nrow(o1), 2)
+  expect_equal(names(o1), c('smiles', 'cas', 'cid', 'einecs', 'csid', 'inchi', 'inchikey',
+                           'drugbank', 'zvg', 'chebi', 'chembl', 'unii', 'lipidmaps', 'swisslipids', 'source_url', 'query'))
+  expect_equal(o1$swisslipids[1], 'SLM:000000510')
+  expect_equal(o1$lipidmaps[2], 'LMPR0102010003')
 })
 
 
@@ -51,7 +65,7 @@ test_that("wd integration test", {
   f <- wd_ident(get_wdid('xxxxxxxAX', language = 'en', match = 'best')$wdid)
 
   expect_equal(d$cas, "110-54-3")
-  expect_equal(ncol(d), 14)
+  expect_equal(ncol(d), 16)
   expect_s3_class(d, 'data.frame')
   expect_true(all(is.na(f[1, ])))
 })