3cc8e060d62a38f811bfe8b66f3a33c1ca5049fd hiram Sat Sep 13 09:00:04 2025 -0700 temporary work around bugs in RepeatMasker code diff --git src/hg/utils/automation/asmHubRepeatMasker.sh src/hg/utils/automation/asmHubRepeatMasker.sh index 92e0225f4c5..878f116a7d6 100755 --- src/hg/utils/automation/asmHubRepeatMasker.sh +++ src/hg/utils/automation/asmHubRepeatMasker.sh @@ -39,32 +39,35 @@ export RepeatMaskerPath="/hive/data/staging/data/RepeatMasker221107" if [ -d "${destDir}" ]; then cd "${destDir}" # might already be gzipped if [ ! -s "${faAlign}" ]; then faAlign="${faAlign}.gz" fi # align file only exists when RM has been run locally, not for NCBI version # it is OK if it is missing, can do this anyway without it if [ -s "${faAlign}" ]; then printf "# using faAlign file: %s\n" "${faAlign}" 1>&2 printf "time $RepeatMaskerPath/util/rmToTrackHub.pl -sizes \"${chrSizes}\" -genome \"${asmId}\" -hubname \"${asmId}\" -out \"${rmOutFile}\" -align \"${faAlign}\"\n" 1>&2 time $RepeatMaskerPath/util/rmToTrackHub.pl -sizes "${chrSizes}" -genome "${asmId}" -hubname "${asmId}" -out "${rmOutFile}" -align "${faAlign}" + awk -F$'\t' '$15 > -1 && $13 > -1' "$asmId.fa.align.tsv" | sort -k1,1 -k2,2n > t.tsv + rm -f "$asmId.fa.align.tsv" + mv t.tsv "$asmId.fa.align.tsv" # in place same file sort using the -o output option - sort -k1,1 -k2,2n -o "${asmId}.fa.align.tsv" "${asmId}.fa.align.tsv" & +# sort -k1,1 -k2,2n -o "${asmId}.fa.align.tsv" "${asmId}.fa.align.tsv" & else printf "# there is no faAlign file\n" 1>&2 printf "time $RepeatMaskerPath/util/rmToTrackHub.pl -sizes \"${chrSizes}\" -genome \"${asmId}\" -hubname \"${asmId}\" -out \"${rmOutFile}\"\n" 1>&2 time $RepeatMaskerPath/util/rmToTrackHub.pl -sizes "${chrSizes}" -genome "${asmId}" -hubname "${asmId}" -out "${rmOutFile}" fi sort -k1,1 -k2,2n -o "${asmId}.sorted.fa.join.tsv" "${asmId}.sorted.fa.join.tsv" & wait printf "bedToBigBed -tab -as=$HOME/kent/src/hg/lib/bigRmskBed.as -type=bed9+5 \"${asmId}.sorted.fa.join.tsv\" \"${chrSizes}\" \"${asmId}.rmsk.bb\" &\n" 1>&2 bedToBigBed -tab -as=$HOME/kent/src/hg/lib/bigRmskBed.as -type=bed9+5 \ "${asmId}.sorted.fa.join.tsv" "${chrSizes}" \ "${asmId}.rmsk.bb" & if [ -s "${asmId}.fa.align.tsv" ]; then printf "bedToBigBed -tab -as=$HOME/kent/src/hg/lib/bigRmskAlignBed.as