2aa890be7b759af7095a891fa999cb27a63bf5a2
hiram
  Fri Feb 28 15:32:32 2020 -0800
better detection if step already done and need to fixup spaces in gene names from NCBI refs #23891

diff --git src/hg/utils/automation/doNcbiRefSeq.pl src/hg/utils/automation/doNcbiRefSeq.pl
index 6e244a5..394369d 100755
--- src/hg/utils/automation/doNcbiRefSeq.pl
+++ src/hg/utils/automation/doNcbiRefSeq.pl
@@ -402,31 +402,31 @@
   fi
 fi
 if [ -s \$db.predicted.gp ]; then
   $genePredCheckDb \$db.predicted.gp
 fi
 if [ -s \$db.other.gp ]; then
   $genePredCheckDb \$db.other.gp
 fi
 
 # join the refLink metadata with curated+predicted names
 cut -f1 \$db.ncbiRefSeq.gp | sort -u > \$asmId.\$db.name.list
 join -t\$'\\t' \$asmId.\$db.name.list \$asmId.refLink.tab > \$asmId.\$db.ncbiRefSeqLink.tab
 
 # Make bigBed with attributes in extra columns for ncbiRefSeqOther:
 twoBitInfo $dbTwoBit stdout | sort -k2,2n > \$db.chrom.sizes
-genePredToBed \$db.other.gp stdout | sort -k1,1 -k2n,2n > \$db.other.bed
+genePredToBed -tab -fillSpace \$db.other.gp stdout | sort -k1,1 -k2n,2n > \$db.other.bed
 $ncbiRefSeqOtherAttrs \$db.other.bed \$asmId.attrs.txt > \$db.other.extras.bed
 bedToBigBed -type=bed12+13 -as=ncbiRefSeqOther.as -tab \\
   -extraIndex=name \\
   \$db.other.extras.bed \$db.chrom.sizes \$db.other.bb
 
 # Make trix index for ncbiRefSeqOther
 $ncbiRefSeqOtherIxIxx \\
   ncbiRefSeqOther.as \$db.other.extras.bed > ncbiRefSeqOther.ix.tab
 
 ixIxx ncbiRefSeqOther.ix.tab ncbiRefSeqOther.ix{,x}
 
 # PSL data will be loaded into a psl type track to show the alignments
 (zgrep "^#" \$ncbiGffGz | head || true) > gffForPsl.gff
 (zegrep -v "NG_" \$ncbiGffGz || true) \\
   | awk -F\$'\\t' '\$3 == "cDNA_match" || \$3 == "match"' >> gffForPsl.gff