16bf6ea9e265c3d36d2485cccb9a6ad45a1646d0 hiram Wed May 6 12:05:34 2026 -0700 additional comments refs #34370 diff --git src/hg/makeDb/doc/vgp577way/vgp577way.txt src/hg/makeDb/doc/vgp577way/vgp577way.txt index 461da3c8aea..5320415f769 100644 --- src/hg/makeDb/doc/vgp577way/vgp577way.txt +++ src/hg/makeDb/doc/vgp577way/vgp577way.txt @@ -132,46 +132,47 @@ ### three of the names need a sed translation to get to UCSC database names: cd /hive/data/genomes/asmHubs/refseqBuild/GCF/037/176/765/GCF_037176765.1_rAnoSag1.mat/trackData/vgp577way echo 's/GCA_000001405.15/hg38/g; s/GCA_000001635.9/mm39/g; s/GCA_009914755.4/hs1/g;' > dbName.sed sed -f dbName.sed fromGlenn/577.species.list | sort > species.list # should always be the same: sum species.list 39306 9 ######################################################################## ### convert to UCSC names, done as a cluster run: +cd /hive/data/genomes/asmHubs/refseqBuild/GCF/037/176/765/GCF_037176765.1_rAnoSag1.mat/trackData/vgp577way + printf '#LOOP dbNameOne $(path1) {check out exists+ ucscMaf/$(path1)} #ENDLOOP ' > template printf '#!/bin/bash set -beEu -o pipefail export chrN="${1}" sed -f dbName.sed split/${chrN} > ucscMaf/${chrN} ' > dbNameOne chmod +x dbNameOne -cd /hive/data/genomes/asmHubs/refseqBuild/GCF/037/176/765/GCF_037176765.1_rAnoSag1.mat/trackData/vgp577way mkdir -p ucscMaf ls -S split > maf.list gensub2 maf.list single template jobList para -ram=3g create jobList para push ### when finished: para time > run.time.ucscMaf cat run.time.ucscMaf Completed: 29 of 29 jobs CPU time in finished jobs: 3427s 57.11m 0.95h 0.04d 0.000 y IO & Wait Time: 185s 3.09m 0.05h 0.00d 0.000 y Average job time: 125s 2.08m 0.03h 0.00d Longest finished job: 690s 11.50m 0.19h 0.01d Submission to last job: 701s 11.68m 0.19h 0.01d @@ -320,30 +321,35 @@ para -ram=6g create jobList para push ### when finished para time > run.time Completed: 29 of 29 jobs CPU time in finished jobs: 187386s 3123.11m 52.05h 2.17d 0.006 y IO & Wait Time: 1052s 17.53m 0.29h 0.01d 0.000 y Average job time: 6498s 108.30m 1.80h 0.08d Longest finished job: 40641s 677.35m 11.29h 0.47d Submission to last job: 40643s 677.38m 11.29h 0.47d ## create a summary count and order list from these measurements ~/kent/src/hg/makeDb/doc/vgp577way/queryCounts.sh \ `pwd -P | sed -e "s#.*/GC\([AF]\)_#GC\\1_#;" | cut -d"_" -f1-2` +## adjust that pwd business: use hg38, hs1, mm39 instead + +## should be 576: +wc -l order.list +576 order.list ######################################################################## ### create the maf summary bed file mkdir /hive/data/genomes/asmHubs/refseqBuild/GCF/037/176/765/GCF_037176765.1_rAnoSag1.mat/trackData/vgp577way/summary cd /hive/data/genomes/asmHubs/refseqBuild/GCF/037/176/765/GCF_037176765.1_rAnoSag1.mat/trackData/vgp577way/summary printf '#LOOP ./runOne $(path1) result/$(root1).bed #ENDLOOP ' > template ls -S ../iRows/result > maf.list gensub2 maf.list single template jobList @@ -364,56 +370,72 @@ export target=`pwd -P | sed -e "s#.*/GC\([AF]\)_#GC\\1_#;" | cut -d"_" -f1-2 | tr -d "_" | sed -e "s/\./v/;"` sed -e "s/GC\([AF]\)_\([0-9]\+\)./GC\\1\\2v/;" "${srcMaf}" \ | /cluster/bin/x86_64/hgLoadMafSummary.v483 \ -test -verbose=2 -minSize=30000 \ -mergeGap=1500 -maxSize=200000 "${target}" "${B}" stdin 2> "${B}.log" sed -e "s/GC\([AF]\)\([0-9]\+\)v/GC\\1_\\2./g;" "${B}.tab" \ > ../${result} rm -f "${B}.tab" ' > runOne chmod +x runOne ### find optimum ram setting here -para -ram=16g create jobList +para -ram=32g create jobList para push ### when finished para time > run.time ls result/*.bed | xargs cut -f2- \ | $HOME/bin/x86_64/gnusort --parallel=32 -k1,1 -k2,2n > vgp577waySummary.bed +### makes a big file: +-rw-rw-r-- 1 4157628035 Apr 24 14:41 vgp577waySummary.bed bedToBigBed -type=bed3+4 -as=$HOME/kent/src/hg/lib/mafSummary.as -tab \ vgp577waySummary.bed ../../../*.chrom.sizes \ ../`pwd -P | sed -e "s#.*/GC\([AF]\)_#GC\1_#;" | cut -d"_" -f1-2`.vgp577waySummary.bb bigBedInfo ../*.vgp577waySummary.bb +### gzip the bed file: +gzip vgp577waySummary.bed +-rw-rw-r-- 1 679264107 Apr 24 14:41 vgp577waySummary.bed.gz + + ######################################################################## ### construct the trackDb cd /hive/data/genomes/asmHubs/refseqBuild/GCF/037/176/765/GCF_037176765.1_rAnoSag1.mat/trackData/vgp577way ~/kent/src/hg/makeDb/doc/vgp577way/mkTdb.sh ## this makes the file -rw-rw-r-- 1 39887 Apr 11 23:08 vgp577way.trackDb.txt ## symlink this track into the contrib directory: +### or, for hg38, hs1, mm39 - take it over to trackDb to incorporate the tracks +### and the bb symlinks go into, for example: +### mkdir -p /gbdb/mm39/vgp577way +### ln -s `pwd`/mm39.*.bb /gbdb/mm39/vgp577way/ +### and the references in the trackDb.txt file need to change: +### bigDataUrl contrib/vgp577way/mm39.vgp577way.bb +### summary contrib/vgp577way/mm39.vgp577waySummary.bb +### bigDataUrl /gbdb/mm39/vgp577way/mm39.vgp577way.bb +### summary /gbdb/mm39/vgp577way/mm39.vgp577waySummary.bb mkdir -p ../../contrib/vgp577way/ ln -s "../../trackData/vgp577way/vgp577way.trackDb.txt" ../../contrib/vgp577way/vgp577way.trackDb.txt for F in *.vgp577way*.bb do ln -s "../../trackData/vgp577way/${F}" "../../contrib/vgp577way/${F}" done ### verify symLinks: ls -og ../../contrib/vgp577way/ total 3 lrwxrwxrwx 1 54 Apr 22 14:06 GCA_964198595.1.vgp577way.bb -> ../../trackData/vgp577way/GCA_964198595.1.vgp577way.bb lrwxrwxrwx 1 61 Apr 22 14:06 GCA_964198595.1.vgp577waySummary.bb -> ../../trackData/vgp577way/GCA_964198595.1.vgp577waySummary.bb lrwxrwxrwx 1 47 Apr 22 14:04 vgp577way.trackDb.txt -> ../../trackData/vgp577way/vgp577way.trackDb.txt