Restructured RDF, so that the author ordinal is now pathway specific

Author: egonw

make-authors-rdf.R

@@ -61,7 +61,10 @@ pathway.mds <- list.files("wikipathways.github.io/_pathways",".md")
 df <- lapply(pathway.mds, function(md){
   dat <- yaml::read_yaml(file.path("wikipathways.github.io/_pathways",md))
   dat$authors = gsub("%\\(\\^", "PctOpenHat", dat$authors)
+  rev = dat$revision
+  if (is.null(dat$revision)) rev = "r1"
   data.frame(wpid=dat$wpid,
+             revision=rev,
              username=dat$authors)
 
 })
@@ -86,20 +89,36 @@ wpa.ttl <- wpa.df %>% dplyr::ungroup() %>%
   dplyr::mutate(s1 = paste0("https://identifiers.org/wikipathways/",wpid)) %>%
   dplyr::mutate(p1 = "dc:creator") %>%
   dplyr::mutate(o1 = paste0('<http://rdf.wikipathways.org/User/',gsub(" ","_",username),'>')) %>%
+  dplyr::mutate(p3 = "pav:hasVersion") %>%
+  dplyr::mutate(o3 = paste0('<https://identifiers.org/wikipathways/',wpid,'_',revision,'>')) %>%
   dplyr::mutate(p2a = "a") %>%
   dplyr::mutate(o2a = paste0('foaf:Person')) %>%
   dplyr::mutate(p2b = "foaf:name") %>%
   dplyr::mutate(o2b = paste0('"',realname,'"')) %>%
   dplyr::mutate(p2c = "foaf:homepage") %>%
   dplyr::mutate(o2c = paste0('<https://wikipathways.org/authors/',gsub(" ","%20",username),'>')) %>%
-  dplyr::mutate(p2d = "pq:series_ordinal") %>%
-  dplyr::mutate(o2d = paste0('"',ordinalRank,'"')) %>%
-  dplyr::select(s1,p1,o1,p2a,o2a,p2b,o2b,p2c,o2c,p2d,o2d)
+  dplyr::mutate(p2d = "wp:hasAuthorship") %>%
+  dplyr::mutate(o2d = paste0('<http://rdf.wikipathways.org/User/',gsub(" ","_",username),'/',wpid,'_',revision,'>')) %>%
+  dplyr::select(s1,p1,o1,p3,o3,p2a,o2a,p2b,o2b,p2c,o2c,p2d,o2d)
 
 wpa.ttl2 <- wpa.ttl %>%
   dplyr::group_by(s1,p1) %>% 
   dplyr::summarise(o1 = paste(gsub("\"","\\\\\"",unique(o1)), sep = "", collapse = ' , '))
 
+wpa.ttl3 <- wpa.df %>% dplyr::ungroup() %>%
+  dplyr::mutate(s3 = paste0('<http://rdf.wikipathways.org/User/',gsub(" ","_",username),'/',wpid,'_',revision,'>')) %>%
+  dplyr::mutate(p4 = "a") %>%
+  dplyr::mutate(o4 = paste0('wp:Authorship')) %>%
+  dplyr::mutate(p3a = "dc:creator") %>%
+  dplyr::mutate(o3a = paste0('<http://rdf.wikipathways.org/User/',gsub(" ","_",username),'>')) %>%
+  dplyr::mutate(p3b = "pq:series_ordinal") %>%
+  dplyr::mutate(o3b = paste0('"',ordinalRank,'"')) %>%
+  dplyr::select(s3,p4,o4,p3a,o3a,p3b,o3b)
+
+wpa.ttl4 <- wpa.ttl %>%
+  dplyr::group_by(s1,p2d) %>%
+  dplyr::summarise(o2d = paste(gsub("\"","\\\\\"",unique(o2d)), sep = "", collapse = ' , '))
+
 ## dplyr::mutate orcid data for RDF
 orcid.ttl <- orcid_wd.df %>%
   dplyr::mutate(o1 = paste0('<http://rdf.wikipathways.org/User/',gsub(" ","_",username),'>')) %>%
@@ -119,19 +138,27 @@ sapply(wpa.ttl2$s1, function(x){
                "@prefix dc:    <http://purl.org/dc/elements/1.1/> .",
                "@prefix owl: <http://www.w3.org/2002/07/owl#> .",
                "@prefix foaf:  <http://xmlns.com/foaf/0.1/> .",
+               "@prefix pav:  <http://purl.org/pav/> .",
                "@prefix pq: <http://www.wikidata.org/prop/qualifier/> .",
+               "@prefix wp:  <http://vocabularies.wikipathways.org/wp#> .",
                "",
                paste0("<",x,">")
   ),filename)
   apply(wpa.ttl2[which(wpa.ttl2$s1==x),], 1, function(y){
-    write(paste0("        ",y['p1'],"            ",y['o1'], " ."), filename, append = TRUE)
+    write(paste0("        ",y['p1'],"           ",y['o1'], " ;"), filename, append = TRUE)
+  })
+  apply(wpa.ttl4[which(wpa.ttl4$s1==x),], 1, function(y){
+    write(paste0("        ",y['p2d'],"     ",y['o2d'], " ;"), filename, append = TRUE)
+  })
+  apply(wpa.ttl[which(wpa.ttl$s1==x)[1],], 1, function(z){
+    write(paste0("        ",z['p3'],"       ",z['o3'], " ."), filename, append = TRUE)
   })
   apply(wpa.ttl[which(wpa.ttl$s1==x),], 1, function(z){
     write("\n", filename, append = TRUE)
     write(z['o1'], filename, append = TRUE)
     write(paste0("        ",z['p2a'],"                    ",z['o2a']," ;"), filename, append = TRUE)
     write(paste0("        ",z['p2b'],"            ",z['o2b']," ;"), filename, append = TRUE)
-    write(paste0("        ",z['p2d'],"            ",z['o2d']," ;"), filename, append = TRUE)
+    write(paste0("        ",z['p2d'],"     ",z['o2d']," ;"), filename, append = TRUE)
     apply(orcid.ttl[which(orcid.ttl$o1==z['o1']),], 1, function(w){
       if (!is.na(w['p2a'])){
         write(paste0("        ",w['p2a'],"           ",w['o2a']," ;"), filename, append = TRUE)
@@ -140,6 +167,14 @@ sapply(wpa.ttl2$s1, function(x){
     })
     write(paste0("        ",z['p2c'],"        ",z['o2c']," ."), filename, append = TRUE)
   })
+  x_rev = wpa.ttl[which(wpa.ttl$s1==x),]$o2d
+  apply(wpa.ttl3[which(wpa.ttl3$s3 %in% x_rev),], 1, function(z){
+    write("\n", filename, append = TRUE)
+    write(z['s3'], filename, append = TRUE)
+    write(paste0("        ",z['p4'],"                    ",z['o4']," ;"), filename, append = TRUE)
+    write(paste0("        ",z['p3a'],"           ",z['o3a']," ;"), filename, append = TRUE)
+    write(paste0("        ",z['p3b'],"    ",z['o3b']," ."), filename, append = TRUE)
+  })
 
 })