cd2eb699ee42cfe22aa6d1587ed3ac444953d8e9
hiram
  Fri Apr 17 13:00:43 2026 -0700
fixup to get geneName2 column correctly populated refs #14450

diff --git src/hg/utils/automation/doXenoRefGene.pl src/hg/utils/automation/doXenoRefGene.pl
index a7b8e41912f..9b0fafe91f6 100755
--- src/hg/utils/automation/doXenoRefGene.pl
+++ src/hg/utils/automation/doXenoRefGene.pl
@@ -302,32 +302,38 @@
 
 if [ -s "\$db.xenoRefGene.psl" ]; then
   grep NR_ \$db.xenoRefGene.psl > NR.psl
   grep NM_ \$db.xenoRefGene.psl > NM.psl
   mrnaToGene -cdsDb=hgFixed NM.psl NM.gp
   mrnaToGene -noCds NR.psl NR.gp
   cat NM.gp NR.gp | genePredSingleCover stdin \$db.xenoRefGene.gp
   genePredCheck -db=\$db -chromSizes=\$db.chrom.sizes \$db.xenoRefGene.gp
   genePredToBed \$db.xenoRefGene.gp stdout \\
     | bedToExons stdin stdout | bedSingleCover.pl stdin > \$db.exons.bed
   export baseCount=`awk '{sum+=\$3-\$2}END{printf "%d", sum}' \$db.exons.bed`
   export asmSizeNoGaps=`grep sequences ../../\$db.faSize.txt | awk '{print \$5}'`
   export perCent=`echo \$baseCount \$asmSizeNoGaps | awk '{printf "%.3f", 100.0*\$1/\$2}'`
   printf "%d bases of %d (%s%%) in intersection\\n" "\$baseCount" "\$asmSizeNoGaps" "\$perCent" > fb.\$db.xenoRefGene.txt
   rm -f \$db.exons.bed
-  genePredToBigGenePred -geneNames=$mrnas/geneOrgXref.txt \$db.xenoRefGene.gp \\
+  # The genePred names have version suffixes (e.g. NM_000014.6) but
+  # geneOrgXref.txt keys are unversioned (NM_000014).  Build a versioned
+  # xref so genePredToBigGenePred can match them and populate geneName2.
+  awk -F'\\t' 'NR==FNR{key[\$1]=\$2"\\t"\$3; next} {split(\$1,a,"."); if(a[1] in key) print \$1"\\t"key[a[1]]}' \\
+    $mrnas/geneOrgXref.txt \$db.xenoRefGene.gp > \$db.versionedXref.txt
+  genePredToBigGenePred -geneNames=\$db.versionedXref.txt \$db.xenoRefGene.gp \\
      stdout | sort -k1,1 -k2,2n > \$db.bgpInput
+  rm -f \$db.versionedXref.txt
   sed -e 's#Alternative/human readable gene name#species of origin of the mRNA#; s#Name or ID of item, ideally both human readable and unique#RefSeq accession id#; s#Primary identifier for gene#gene name#;' \\
     \$HOME/kent/src/hg/lib/bigGenePred.as > xenoRefGene.as
   bedToBigBed -extraIndex=name,geneName -type=bed12+8 -tab -as=xenoRefGene.as \\
      \$db.bgpInput \$db.chrom.sizes \$db.xenoRefGene.bb
   \$HOME/kent/src/hg/utils/automation/xenoRefGeneIx.pl \$db.bgpInput | sort -u > \$db.ix.txt
   ixIxx \$db.ix.txt \$db.xenoRefGene.ix \$db.xenoRefGene.ixx
   mkdir -p /dev/shm/\$db
   cp -p \$db.xenoRefGene.gp /dev/shm/\$db/xenoRefGene.\$db
   cd /dev/shm/\$db
   genePredToGtf -utr file xenoRefGene.\$db stdout | gzip -c \\
     > \$buildDir/\$db.xenoRefGene.gtf.gz
   cd \$buildDir
   rm -f /dev/shm/\$db/xenoRefGene.\$db
   rmdir /dev/shm/\$db
 fi