3683a482f97f26e6c08290f6c8acccb764d16eac
lrnassar
  Thu Mar 2 17:58:16 2023 -0800
Removing the use of refseq select because the annotations are too sparse. Refs #30320

diff --git src/hg/makeDb/doc/evaSnp4.txt src/hg/makeDb/doc/evaSnp4.txt
index aaae0a9..0bdde85 100644
--- src/hg/makeDb/doc/evaSnp4.txt
+++ src/hg/makeDb/doc/evaSnp4.txt
@@ -207,32 +207,32 @@
     chromDic = {}
     for chrom in chromSizes[1:-1]:
         chrom = chrom.split("\t")
         chromDic[chrom[0]] = chrom[1]
     return(chromDic)
     
 def splitChromsAndRunHgVai(workDir,dbs):
     """Split all the chroms in tenths in order to be able to run hgVai without running out of memory"""
     chromSizes = bash("fetchChromSizes %s" % (dbs)).split("\n")
     inputFile = workDir+"evaSnps.ucscChroms.vcf.gz"
     outputFile = workDir+"evaSnps"+dbs+"VaiResults.vep"
     n=0
     allTables = bash("\hgsql -e \"show tables\" "+dbs+"")
     if dbs in ['galGal6','oviAri4','ponAbe3']: #Special exception to use refGene because ncbiRefSeq has incorrect protein sequence in Link table see #29262
         geneTableToUse = "refGene"
-    elif "ncbiRefSeqSelect" in allTables:
-        geneTableToUse = "ncbiRefSeqSelect"
+#     elif "ncbiRefSeqSelect" in allTables: #Removing refseq select as the annotations are too sparse
+#         geneTableToUse = "ncbiRefSeqSelect"
     elif "ncbiRefSeqCurated" in allTables:
         geneTableToUse = "ncbiRefSeqCurated"
     elif "ensGene" in allTables:
         geneTableToUse = "ensGene"
     elif "refGene" in allTables:
         geneTableToUse = "refGene"
     elif "ncbiGene" in allTables:
         geneTableToUse = "ncbiGene"
     else:
         print(dbs)
         sys.exit("Could not find any tables to use for the following database: "+dbs) 
     
     chromDic = buildChromSizesDic(workDir,dbs)
     #For function below, only bother with the chromosomes in the VCF
     chromsInVcf = bash("zcat "+inputFile+" | grep -v '^#' | cut -f1 | uniq").rstrip().split("\n")