(v2.1.1.9071) update veterinary SIR interpretation, add only_fungi

data-raw/ab.md5

@@ -1 +1 @@
-92d3e2f8deac335c92841d2ded974dee
+5ad894790ad048110f8eb9207b89501f

data-raw/antibiotics.txt

@@ -468,7 +468,7 @@
 "TOH"		"Tobramycin-high"	"Aminoglycosides"	"NA"			"tobra high,tobramycin high,tohl"	""					""
 "TFX"	5517	"Tosufloxacin"	"Quinolones"	"J01MA22,S01AE09"			""	"tosufloxacin"	0.45	"g"			"100061-1,76146-0"
 "TMP"	5578	"Trimethoprim"	"Trimethoprims"	"J01EA01"	"Sulfonamides and trimethoprim"	"Trimethoprim and derivatives"	"t,tmp,tr,tri,trim,w"	"abaprim,alprim,anitrim,antrima,antrimox,bacdan,bacidal,bacide,bacterial,bacticel,bactifor,bactin,bactoprim,bactramin,bactrim,bencole,bethaprim,biosulten,briscotrim,chemotrin,colizole,colizole ds,conprim,cotrimel,cotrimoxizole,deprim,dosulfin,duocide,esbesul,espectrin,euctrim,exbesul,fermagex,fortrim,idotrim,ikaprim,infectotrimet,instalac,kombinax,lagatrim,lagatrim forte,lastrim,lescot,methoprim,metoprim,monoprim,monotrim,monotrimin,novotrimel,omstat,oraprim,pancidim,polytrim,priloprim,primosept,primsol,proloprim,protrin,purbal,resprim,resprim forte,roubac,roubal,salvatrim,septrin ds,septrin forte,septrin s,setprin,sinotrim,stopan,streptoplus,sugaprim,sulfamar,sulfamethoprim,sulfoxaprim,sulthrim,sultrex,syraprim,tiempe,tmp smx,toprim,trimanyl,trimethioprim,trimethopim,trimethoprim,trimethoprime,trimethoprimum,trimethopriom,trimetoprim,trimetoprima,trimexazole,trimexol,trimezol,trimogal,trimono,trimopan,trimpex,triprim,trisul,trisulcom,trisulfam,trisural,uretrim,urobactrim,utetrin,velaten,wellcoprim,wellcoprin,xeroprim,zamboprim"	0.4	"g"	0.4	"g"	"101495-0,11005-6,17747-7,18997-7,18998-5,20387-7,23614-1,23631-5,25273-4,32342-8,4079-0,4080-8,4081-6,511-6,512-4,513-2,514-0,515-7,516-5,517-3,518-1,55584-7,7056-5,7057-3,80552-3,80973-1"
-"SXT"	358641	"Trimethoprim/sulfamethoxazole"	"Trimethoprims"	"J01EE01"	"Sulfonamides and trimethoprim"	"Combinations of sulfonamides and trimethoprim, incl. derivatives"	"cot,cotrim,sxt,t/s,trsu,trsx,ts"	"abacin,abactrim,agoprim,alfatrim,aposulfatrim,bacteral,bacterial forte,bactilen,bactiver,bacton,bactoreduct,bactrim,bactrim ds,bactrim forte,bactrim pediatric,bactrimel,bactrizol,bactromin,bactropin,baktar,belcomycine,berlocid,bibacrim,biseptol,chemitrim,chemotrim,ciplin,colimycin,colimycin sulphate,colisticin,colistimethate,colistimethate sodium,colistin sulfate,colistin sulphate,colomycin,coly-mycin,cotribene,cotrim d.s.,cotrim eu rho,cotrim holsen,cotrim.l.u.t.,cotrimaxazol,cotrimazole,cotrimhexal,cotrimoxazol,cotrimoxazol al,cotrimoxazole,cotrimstada,cotriver,dibaprim,drylin,duratrimet,eltrianyl,escoprim,esteprim,eusaprim,fectrim,gantaprim,gantaprin,gantrim,groprim,helveprim,imexim,jenamoxazol,kemoprim,kepinol,kepinol forte,laratrim,linaris,maxtrim,microtrim,microtrim forte,mikrosid,momentol,oecotrim,oriprim,oxaprim,pantoprim,polymyxin e,polymyxin e. sulfate,primazole,promixin,septra,septra ds,septra grape,septrim,septrin,servitrim,sigaprim,sigaprin,sulfatrim pediatric,sulfotrim,sulfotrimin,sulmeprim pediatric,sulprim,sumetrolim,supracombin,suprim,tacumil,teleprim,teleprin,thiocuran,totazina,tribakin,trifen,trigonyl,trimesulf,trimetho comp,trimethoprimsulfa,trimetoger,trimexazol,trimforte,trimosulfa,uroplus,uroplus ds,uroplus ss"					"101495-0,18998-5,20387-7,23631-5,25273-4,32342-8,4081-6,515-7,516-5,517-3,518-1,7057-3"
+"SXT"	358641	"Trimethoprim/sulfamethoxazole"	"Trimethoprims"	"J01EE01"	"Sulfonamides and trimethoprim"	"Combinations of sulfonamides and trimethoprim, incl. derivatives"	"cot,cotrim,sxt,t/s,tms,trsu,trsx,ts"	"abacin,abactrim,agoprim,alfatrim,aposulfatrim,bacteral,bacterial forte,bactilen,bactiver,bacton,bactoreduct,bactrim,bactrim ds,bactrim forte,bactrim pediatric,bactrimel,bactrizol,bactromin,bactropin,baktar,belcomycine,berlocid,bibacrim,biseptol,chemitrim,chemotrim,ciplin,colimycin,colimycin sulphate,colisticin,colistimethate,colistimethate sodium,colistin sulfate,colistin sulphate,colomycin,coly-mycin,cotribene,cotrim d.s.,cotrim eu rho,cotrim holsen,cotrim.l.u.t.,cotrimaxazol,cotrimazole,cotrimhexal,cotrimoxazol,cotrimoxazol al,cotrimoxazole,cotrimstada,cotriver,dibaprim,drylin,duratrimet,eltrianyl,escoprim,esteprim,eusaprim,fectrim,gantaprim,gantaprin,gantrim,groprim,helveprim,imexim,jenamoxazol,kemoprim,kepinol,kepinol forte,laratrim,linaris,maxtrim,microtrim,microtrim forte,mikrosid,momentol,oecotrim,oriprim,oxaprim,pantoprim,polymyxin e,polymyxin e. sulfate,primazole,promixin,septra,septra ds,septra grape,septrim,septrin,servitrim,sigaprim,sigaprin,sulfatrim pediatric,sulfotrim,sulfotrimin,sulmeprim pediatric,sulprim,sumetrolim,supracombin,suprim,tacumil,teleprim,teleprin,thiocuran,totazina,tribakin,trifen,trigonyl,trimesulf,trimetho comp,trimethoprimsulfa,trimetoger,trimexazol,trimforte,trimosulfa,uroplus,uroplus ds,uroplus ss"					"101495-0,18998-5,20387-7,23631-5,25273-4,32342-8,4081-6,515-7,516-5,517-3,518-1,7057-3"
 "TRL"	202225	"Troleandomycin"	"Macrolides/lincosamides"	"J01FA08"	"Macrolides, lincosamides and streptogramins"	"Macrolides"	""	"acetyloleandomycin,aovine,cyclamycin,evramicina,matromicina,matromycin t,micotil,oleandocetine,oleandomycin,t.a.o.,treolmicina,tribiocillina,triocetin,triolan,troleandomicina,troleandomycin,troleandomycine,troleandomycinum,viamicina,wytrion"	1	"g"			"18999-3,519-9,520-7,521-5,522-3"
 "TRO"	55886	"Trospectomycin"	"Other antibacterials"	"NA"			""	"rubidiumnitrate,trospectinomycin,trospectomicina,trospectomycin,trospectomycine,trospectomycinum"					""
 "TVA"	62959	"Trovafloxacin"	"Quinolones"	"J01MA13"	"Quinolone antibacterials"	"Fluoroquinolones"	"trov"	"trovafloxacin,trovan"	0.2	"g"	0.2	"g"	"23642-2,23643-0,35855-6,7058-1"

data-raw/reproduction_of_microorganisms.R

@@ -736,6 +736,8 @@ taxonomy_mycobank <- taxonomy_mycobank %>%
 
 # Combine the datasets ----------------------------------------------------
 
+# TODO !! check why e.g. Clavispora lusitaniae is gotten from GBIF, not MycoBank !!
+
 taxonomy <- taxonomy_lpsn %>%
   # add fungi
   bind_rows(taxonomy_mycobank) %>% 
@@ -1237,9 +1239,9 @@ saveRDS(taxonomy, "data-raw/taxonomy2.rds")
 # taxonomy <- readRDS("data-raw/taxonomy2.rds")
 
 
-# Remove unwanted taxonomic entries from Protoza/Fungi --------------------
+# Remove unwanted taxonomic entries ---------------------------------------
 
-taxonomy <- taxonomy %>%
+part1 <- taxonomy %>%
   filter(
     # keep all we added ourselves:
     source == "manually added" |
@@ -1259,10 +1261,7 @@ taxonomy <- taxonomy %>%
       # kingdom of Protozoa:
       (phylum %in% c("Choanozoa", "Mycetozoa") & prevalence < 2) |
       # Fungi:
-      (kingdom == "Fungi" & (!rank %in% c("genus", "species", "subspecies") | prevalence < 2)) |
-      # !(phylum %in% c("Ascomycota", "Zygomycota", "Basidiomycota") & prevalence == 2 & rank %in% c("genus", "species", "subspecies")),
-      # !(genus %in% c("Leptosphaeria", "Physarum") & rank %in% c("species", "subspecies")), # keep only genus of this rare fungus, with resp. 850 and 500 species
-      # # (leave Alternaria in there, part of human mycobiome and opportunistic pathogen)
+      (kingdom == "Fungi" & (!rank %in% c("genus", "species", "subspecies") | prevalence < 2 | class == "Pichiomycetes")) |
       # Animalia:
       genus %in% c("Lucilia", "Lumbricus") |
       (class == "Insecta" & !rank %in% c("species", "subspecies")) | # keep only genus of insects, not all of their (sub)species
@@ -1272,6 +1271,29 @@ taxonomy <- taxonomy %>%
   filter(kingdom != "Plantae",
          !(genus %in% c("Aedes", "Anopheles") & rank %in% c("species", "subspecies")))
 
+# now get the parents and old names
+part2 <- taxonomy %>% 
+  filter(gbif %in% c(part1$gbif_parent[!is.na(part1$gbif_parent)], part1$gbif_renamed_to[!is.na(part1$gbif_renamed_to)]) |
+           mycobank %in% c(part1$mycobank_parent[!is.na(part1$mycobank_parent)], part1$mycobank_renamed_to[!is.na(part1$mycobank_renamed_to)]) |
+           lpsn %in% c(part1$lpsn_parent[!is.na(part1$lpsn_parent)], part1$lpsn_renamed_to[!is.na(part1$lpsn_renamed_to)]))
+parts <- bind_rows(part1, part2)
+
+part3 <- taxonomy %>% 
+  filter(gbif %in% c(parts$gbif_parent[!is.na(parts$gbif_parent)], parts$gbif_renamed_to[!is.na(parts$gbif_renamed_to)]) |
+           mycobank %in% c(parts$mycobank_parent[!is.na(parts$mycobank_parent)], parts$mycobank_renamed_to[!is.na(parts$mycobank_renamed_to)]) |
+           lpsn %in% c(parts$lpsn_parent[!is.na(parts$lpsn_parent)], parts$lpsn_renamed_to[!is.na(parts$lpsn_renamed_to)]))
+parts <- bind_rows(part1, part2, part3)
+
+part4 <- taxonomy %>% 
+  filter(gbif %in% c(parts$gbif_parent[!is.na(parts$gbif_parent)], parts$gbif_renamed_to[!is.na(parts$gbif_renamed_to)]) |
+           mycobank %in% c(parts$mycobank_parent[!is.na(parts$mycobank_parent)], parts$mycobank_renamed_to[!is.na(parts$mycobank_renamed_to)]) |
+           lpsn %in% c(parts$lpsn_parent[!is.na(parts$lpsn_parent)], parts$lpsn_renamed_to[!is.na(parts$lpsn_renamed_to)]))
+parts <- bind_rows(part1, part2, part3, part4)
+
+taxonomy <- bind_rows(part1, part2, part3, part4) %>% 
+  arrange(fullname) %>% 
+  distinct(fullname, .keep_all = TRUE)
+
 # no ghost families, orders classes, phyla
 taxonomy <- taxonomy %>%
   group_by(kingdom, family) %>%
@@ -2005,7 +2027,7 @@ taxonomy <- taxonomy %>%
   filter(!mo %in% groups$mo) %>% 
   bind_rows(groups)
 
-# we added MO code, so make sure everything is still unique
+# we added an MO code, so make sure everything is still unique
 any(duplicated(taxonomy$mo))
 any(duplicated(taxonomy$fullname))
 
@@ -2018,13 +2040,17 @@ AMR::clinical_breakpoints %>% filter(!mo %in% taxonomy$mo)
 AMR::example_isolates %>% filter(!mo %in% taxonomy$mo)
 AMR::intrinsic_resistant %>% filter(!mo %in% taxonomy$mo)
 
+# all our previously manually added names should be in it
+all(microorganisms$fullname[microorganisms$source == "manually added"] %in% taxonomy$fullname)
+microorganisms$fullname[!microorganisms$fullname[microorganisms$source == "manually added"] %in% taxonomy$fullname]
+
 # put this one back
 taxonomy <- taxonomy %>% 
   bind_rows(microorganisms %>%
               filter(fullname == "Blastocystis hominis") %>% 
               mutate(mo = as.character(mo)))
 
-# we added MO code, so make sure everything is still unique
+# we added an MO code, so make sure everything is still unique
 any(duplicated(taxonomy$mo))
 any(duplicated(taxonomy$fullname))