90552b46adcb6b1641a0e027e1cf722e411c9664 hiram Mon Aug 17 10:23:57 2026 -0700 now running the oneLftp.sh to fetch the NCBI files for the build of the assembly refs #32596 diff --git src/hg/makeDb/doc/asmHubs/runBuild src/hg/makeDb/doc/asmHubs/runBuild index 9d8c917efe0..241c6e7f305 100755 --- src/hg/makeDb/doc/asmHubs/runBuild +++ src/hg/makeDb/doc/asmHubs/runBuild @@ -1,23 +1,23 @@ #!/bin/bash set -beEu -o pipefail export LANG=C export asmId=$1 export clade=$2 -export sciName=$3 +export sciName="${3}" ### the 'clade' is used to establish specific options for the build ### start seconds export startT=`date "+%s"` export bigHub="hgwdev" export smallHub="hgwdev" export fileServer="hgwdev" export workHorse="hgwdev" # export bigHub="ku" # export smallHub="ku" export rmskSpecies="xy" ### This ucscNames decision needs to be an automatic process since @@ -25,30 +25,34 @@ ### This is important for track updates, such as ncbiRefSeq ### export ucscNames="-ucscNames" export ucscNames="" export augustusSpecies="-augustusSpecies=human" export ncbiRmsk="-ncbiRmsk" export noRmsk="" export RMSK="" export subGroup="vertebrate_other" # sleep 5 to 25 seconds to avoid the ssh overload of a bunch of jobs # all starting at the same time. export sleepTime=$((RANDOM % 21 + 5)) sleep "${sleepTime}" +export fetchLog="/dev/shm/fetch.$$.log" +# make sure the NCBI source is here: +/hive/data/outside/ncbi/genomes/oneLftp.sh "${asmId}" > "${fetchLog}" + if [ "${clade}" = "primates" ]; then subGroup="vertebrate_mammalian" rmskSpecies=`/hive/data/genomes/asmHubs/allBuild/rmCheck/findTaxon.pl "${sciName}" 2> /dev/null` elif [ "${clade}" = "mammals" ]; then subGroup="vertebrate_mammalian" rmskSpecies=`/hive/data/genomes/asmHubs/allBuild/rmCheck/findTaxon.pl "${sciName}" 2> /dev/null` elif [ "${clade}" = "fishes" ]; then subGroup="vertebrate_other" augustusSpecies="-augustusSpecies=zebrafish" rmskSpecies=`/hive/data/genomes/asmHubs/allBuild/rmCheck/findTaxon.pl "${sciName}" 2> /dev/null` elif [ "${clade}" = "fish" ]; then subGroup="vertebrate_other" augustusSpecies="-augustusSpecies=zebrafish" rmskSpecies=`/hive/data/genomes/asmHubs/allBuild/rmCheck/findTaxon.pl "${sciName}" 2> /dev/null` elif [ "${clade}" = "birds" ]; then @@ -106,33 +110,33 @@ noRmsk="-noRmsk" augustusSpecies="-noAugustus -noXenoRefSeq" elif [ "${clade}" = "bacteria" ]; then subGroup="bacteria" noRmsk="-noRmsk" augustusSpecies="-noAugustus -noXenoRefSeq" else printf "ERROR: unrecognized clade: '%s'\n" "${clade}" 1>&2 exit 255 fi if [ "${rmskSpecies}" = "xy" ]; then noRmsk="-noRmsk" elif [ "x${rmskSpecies}y" = "xy" ]; then export RMSK="-runRepeatModeler" - printf "# rmskSpecies: $sciName NOT FOUND, running RepeatModeler\n" 1>&2 + printf "# rmskSpecies: ${sciName} NOT FOUND, running RepeatModeler\n" 1>&2 else - printf "# rmskSpecies: $sciName -> $rmskSpecies\n" 1>&2 + printf "# rmskSpecies: ${sciName} -> $rmskSpecies\n" 1>&2 RMSK='-rmskSpecies="'"${rmskSpecies}"'"' # RMSK='-runRepeatModeler -rmskSpecies="'"${rmskSpecies}"'"' fi export stepStart="download" ## export stepStart="sequence" ## export stepStart="gatewayPage" ## export stepStart="chromAlias" ## export stepStart="cytoBand" ## export stepStart="repeatModeler" ## export stepStart="repeatMasker" ## export stepStart="simpleRepeat" ## export stepStart="allGaps" ## export stepStart="windowMasker" ## export stepStart="addMask" @@ -200,31 +204,35 @@ export logFile="${TOP0}/${topBuild}/buildLogs/${subGroup}/${asmId}.log" if [ ! -d "${buildDir}" ]; then mkdir -p "${buildDir}" fi if [ ! -d "${linkDir}" ]; then mkdir -p "${linkDir}" fi if [ ! -L "${linkDir}/${asmId}" ]; then printf "@ ln -s \"${buildDir}\" \"${linkDir}\"\n" 1>&2 ln -s "${buildDir}" "${linkDir}" fi ls -d "${buildDir}" "${linkDir}/${asmId}" echo "========================= "`date "+%F %T"` >> "${logFile}" +echo "=== NCBI fetch: ====" >> "${logFile}" +echo "/hive/data/outside/ncbi/genomes/oneLftp.sh \"${asmId}\"" >> "${logFile}" +cat "${fetchLog}" >> "${logFile}" +echo "=== doIdKeys: ==== "`date "+%F %T"` >> "${logFile}" echo "/hive/data/genomes/asmHubs/${genbankRefseq}Build/doIdKeys \"${asmId}\" &" >> "${logFile}" /hive/data/genomes/asmHubs/${genbankRefseq}Build/doIdKeys "${asmId}" >> "${logFile}" 2>&1 & echo "### \$HOME/kent/src/hg/utils/automation/doAssemblyHub.pl \ -continue=\"${stepStart}\" -stop=\"${stepEnd}\" \ ${RMSK} -bigClusterHub="${bigHub}" -buildDir=\`pwd\` \ -fileServer=${fileServer} -smallClusterHub=${smallHub} \ ${noRmsk} ${ncbiRmsk} ${ucscNames} ${augustusSpecies} -workhorse=${workHorse} \"${asmId}\"" >> "${logFile}" cd "${buildDir}" if [ ! -s "${buildDir}/build.log" ]; then symLinkPath=`readlink "${buildDir}/build.log" || true` if [ "x${symLinkPath}y" != "xy" ]; then # empty symLink rm -f "${buildDir}/build.log"