improved taxonomy parsing

jbisanz · Oct 19, 2018 · 337067a · 337067a
1 parent ba443df
commit 337067a
Show file tree

Hide file tree

Showing 2 changed files with 42 additions and 1 deletion.
diff --git a/R/qza_to_phyloseq.R b/R/qza_to_phyloseq.R
@@ -34,8 +34,12 @@ qza_to_phyloseq<-function(features,tree,taxonomy,metadata, tmp){
 
   if(!missing(taxonomy)){
     taxonomy<-read_qza(taxonomy, tmp=tmp)$data
-    taxt<-suppressWarnings(do.call(rbind, strsplit(as.character(taxonomy$Taxon),"\\; ")))
+    taxt<-strsplit(as.character(taxonomy$Taxon),"\\; ")
+    taxt<-lapply(taxt, function(x){length(x)=7;return(x)})
+    taxt<-do.call(rbind, taxt)
+    taxt<-apply(taxt,2, function(x) replace(x, grepl("^[kpcofgs]__$", x), "Not_Assigned"))
     rownames(taxt)<-taxonomy$Feature.ID
+    colnames(taxt)<-c("Kingdom","Phylum","Class","Order","Family","Genus","Species")
     argstring<-paste(argstring, "tax_table(taxt),")
   }
 
@@ -56,3 +60,4 @@ qza_to_phyloseq<-function(features,tree,taxonomy,metadata, tmp){
 
   return(physeq)
 }
+
diff --git a/inst/sample-metadata.tsv b/inst/sample-metadata.tsv
@@ -0,0 +1,36 @@
+#SampleID	BarcodeSequence	LinkerPrimerSequence	BodySite	Year	Month	Day	Subject	ReportedAntibioticUsage	DaysSinceExperimentStart	Description
+#q2:types	categorical	categorical	categorical	numeric	numeric	numeric	categorical	categorical	numeric	categorical
+L1S8	AGCTGACTAGTC	GTGCCAGCMGCCGCGGTAA	gut	2008	10	28	subject-1	Yes	0	subject-1.gut.2008-10-28
+L1S57	ACACACTATGGC	GTGCCAGCMGCCGCGGTAA	gut	2009	1	20	subject-1	No	84	subject-1.gut.2009-1-20
+L1S76	ACTACGTGTGGT	GTGCCAGCMGCCGCGGTAA	gut	2009	2	17	subject-1	No	112	subject-1.gut.2009-2-17
+L1S105	AGTGCGATGCGT	GTGCCAGCMGCCGCGGTAA	gut	2009	3	17	subject-1	No	140	subject-1.gut.2009-3-17
+L2S155	ACGATGCGACCA	GTGCCAGCMGCCGCGGTAA	left palm	2009	1	20	subject-1	No	84	subject-1.left-palm.2009-1-20
+L2S175	AGCTATCCACGA	GTGCCAGCMGCCGCGGTAA	left palm	2009	2	17	subject-1	No	112	subject-1.left-palm.2009-2-17
+L2S204	ATGCAGCTCAGT	GTGCCAGCMGCCGCGGTAA	left palm	2009	3	17	subject-1	No	140	subject-1.left-palm.2009-3-17
+L2S222	CACGTGACATGT	GTGCCAGCMGCCGCGGTAA	left palm	2009	4	14	subject-1	No	168	subject-1.left-palm.2009-4-14
+L3S242	ACAGTTGCGCGA	GTGCCAGCMGCCGCGGTAA	right palm	2008	10	28	subject-1	Yes	0	subject-1.right-palm.2008-10-28
+L3S294	CACGACAGGCTA	GTGCCAGCMGCCGCGGTAA	right palm	2009	1	20	subject-1	No	84	subject-1.right-palm.2009-1-20
+L3S313	AGTGTCACGGTG	GTGCCAGCMGCCGCGGTAA	right palm	2009	2	17	subject-1	No	112	subject-1.right-palm.2009-2-17
+L3S341	CAAGTGAGAGAG	GTGCCAGCMGCCGCGGTAA	right palm	2009	3	17	subject-1	No	140	subject-1.right-palm.2009-3-17
+L3S360	CATCGTATCAAC	GTGCCAGCMGCCGCGGTAA	right palm	2009	4	14	subject-1	No	168	subject-1.right-palm.2009-4-14
+L5S104	CAGTGTCAGGAC	GTGCCAGCMGCCGCGGTAA	tongue	2008	10	28	subject-1	Yes	0	subject-1.tongue.2008-10-28
+L5S155	ATCTTAGACTGC	GTGCCAGCMGCCGCGGTAA	tongue	2009	1	20	subject-1	No	84	subject-1.tongue.2009-1-20
+L5S174	CAGACATTGCGT	GTGCCAGCMGCCGCGGTAA	tongue	2009	2	17	subject-1	No	112	subject-1.tongue.2009-2-17
+L5S203	CGATGCACCAGA	GTGCCAGCMGCCGCGGTAA	tongue	2009	3	17	subject-1	No	140	subject-1.tongue.2009-3-17
+L5S222	CTAGAGACTCTT	GTGCCAGCMGCCGCGGTAA	tongue	2009	4	14	subject-1	No	168	subject-1.tongue.2009-4-14
+L1S140	ATGGCAGCTCTA	GTGCCAGCMGCCGCGGTAA	gut	2008	10	28	subject-2	Yes	0	subject-2.gut.2008-10-28
+L1S208	CTGAGATACGCG	GTGCCAGCMGCCGCGGTAA	gut	2009	1	20	subject-2	No	84	subject-2.gut.2009-1-20
+L1S257	CCGACTGAGATG	GTGCCAGCMGCCGCGGTAA	gut	2009	3	17	subject-2	No	140	subject-2.gut.2009-3-17
+L1S281	CCTCTCGTGATC	GTGCCAGCMGCCGCGGTAA	gut	2009	4	14	subject-2	No	168	subject-2.gut.2009-4-14
+L2S240	CATATCGCAGTT	GTGCCAGCMGCCGCGGTAA	left palm	2008	10	28	subject-2	Yes	0	subject-2.left-palm.2008-10-28
+L2S309	CGTGCATTATCA	GTGCCAGCMGCCGCGGTAA	left palm	2009	1	20	subject-2	No	84	subject-2.left-palm.2009-1-20
+L2S357	CTAACGCAGTCA	GTGCCAGCMGCCGCGGTAA	left palm	2009	3	17	subject-2	No	140	subject-2.left-palm.2009-3-17
+L2S382	CTCAATGACTCA	GTGCCAGCMGCCGCGGTAA	left palm	2009	4	14	subject-2	No	168	subject-2.left-palm.2009-4-14
+L3S378	ATCGATCTGTGG	GTGCCAGCMGCCGCGGTAA	right palm	2008	10	28	subject-2	Yes	0	subject-2.right-palm.2008-10-28
+L4S63	CTCGTGGAGTAG	GTGCCAGCMGCCGCGGTAA	right palm	2009	1	20	subject-2	No	84	subject-2.right-palm.2009-1-20
+L4S112	GCGTTACACACA	GTGCCAGCMGCCGCGGTAA	right palm	2009	3	17	subject-2	No	140	subject-2.right-palm.2009-3-17
+L4S137	GAACTGTATCTC	GTGCCAGCMGCCGCGGTAA	right palm	2009	4	14	subject-2	No	168	subject-2.right-palm.2009-4-14
+L5S240	CTGGACTCATAG	GTGCCAGCMGCCGCGGTAA	tongue	2008	10	28	subject-2	Yes	0	subject-2.tongue.2008-10-28
+L6S20	GAGGCTCATCAT	GTGCCAGCMGCCGCGGTAA	tongue	2009	1	20	subject-2	No	84	subject-2.tongue.2009-1-20
+L6S68	GATACGTCCTGA	GTGCCAGCMGCCGCGGTAA	tongue	2009	3	17	subject-2	No	140	subject-2.tongue.2009-3-17
+L6S93	GATTAGCACTCT	GTGCCAGCMGCCGCGGTAA	tongue	2009	4	14	subject-2	No	168	subject-2.tongue.2009-4-14