# for emacs: -*- mode: sh; -*- # This file describes browser build for the galGal5 # New starting procedure, 2016-04-12 # # Can use existing photograph (otherwise find one before starting here) ######################################################################### # Initial steps, find photograph (DONE - 2016-04-12 - Hiram) # To start this initialBuild.txt document, from a previous assembly document: mkdir ~/kent/src/hg/makeDb/doc/galGal5 cd ~/kent/src/hg/makeDb/doc/galGal5 sed -e 's/rouAeg1/galGal5/g; s/RouAeg1/GalGal5/g; s/DONE/TBD/g;' \ ../rouAeg1/initialBuild.txt > initialBuild.txt # the files required are probably already here, take a look into: # /hive/data/outside/ncbi/genomes/refseq///all_assembly_versions # and merely symlink them in: mkdir -p /hive/data/genomes/galGal5/refseq cd -p /hive/data/genomes/galGal5/refseq ln -s /hive/data/outside/ncbi/genomes/refseq/vertebrate_other/Gallus_gallus/all_assembly_versions/GCF_000002315.4_Gallus_gallus-5.0/* ./ # Can use existing photograph # construct the required photoReference.txt cd /hive/data/genomes/galGal5 printf "photoCreditURL %s\nphotoCreditName %s\n" \ 'http://www.genome.gov/dmd/img.cfm?node=Photos/Animals/Red%20Jungle%20Fowl&id=2120' \ "NHGRI Press Photos" > photoReference.txt # this information is from the top of # galGal5/refseq/*_assembly_report.txt # (aka: galGal5/refseq/GCF_001466805.2_Raegyp2.0_assembly_report.txt) # Assembly name: Gallus_gallus-5.0 # Organism name: Gallus gallus (chicken) # Infraspecific name: breed=Red Jungle fowl, inbred line UCD001 # Isolate: RJF #256 # Sex: female # Taxid: 9031 # BioSample: SAMN02981218 # BioProject: PRJNA10808 # Submitter: International Chicken Genome Consortium # Date: 2015-12-16 # Assembly type: haploid # Release type: major # Assembly level: Chromosome # Genome representation: full # WGS project: AADN04 # Assembly method: MHAP/PBcR v. 8.2beta # Genome coverage: 70x # Sequencing technology: Sanger; 454; Illumina; PacBio # RefSeq category: Representative Genome # GenBank assembly accession: GCA_000002315.3 (latest) # RefSeq assembly accession: GCF_000002315.4 (species-representative latest) # RefSeq assembly and GenBank assemblies identical: no # ## Assembly-Units: ## GenBank Unit Accession RefSeq Unit Accession Assembly-Unit name ## GCA_000000185.3 GCF_000000185.3 Primary Assembly ## GCF_000184395.1 non-nuclear ############################################################################# # establish config.ra file (DONE - Hiram - 2016-04-12) cd /hive/data/genomes/galGal5 ~/kent/src/hg/utils/automation/prepConfig.pl galGal5 vertebrate chicken \ refseq/*_assembly_report.txt > galGal5.config.ra # verify it looks sane cat galGal5.config.ra # config parameters for makeGenomeDb.pl: db galGal5 clade vertebrate scientificName Gallus gallus commonName Chicken assemblyDate Dec 2015 assemblyLabel International Chicken Genome Consortium assemblyShortLabel Gallus_gallus-5.0 orderKey 3314 # mitochondrial sequence included in refseq release # mitoAcc NC_001323.1 mitoAcc none fastaFiles /hive/data/genomes/galGal5/ucsc/*.fa.gz agpFiles /hive/data/genomes/galGal5/ucsc/*.agp # qualFiles none dbDbSpeciesDir chicken photoCreditURL http://www.genome.gov/dmd/img.cfm?node=Photos/Animals/Red%20Jungle%20Fowl&id=2120 photoCreditName NHGRI Press Photos ncbiGenomeId 111 ncbiAssemblyId 595851 ncbiAssemblyName Gallus_gallus-5.0 ncbiBioProject 10808 genBankAccessionID GCF_000002315.4 taxId 9031 ############################################################################# # setup UCSC named files (DONE - 2016-04-13 - Hiram) mkdir /hive/data/genomes/galGal5/ucsc cd /hive/data/genomes/galGal5/ucsc # measure what is in the refseq release: faSize ../refseq/*genomic.fna.gz # 1230258557 bases (11766024 N's 1218492533 real 916521624 upper # 301970909 lower) in 23475 sequences in 1 files # Total size: mean 52407.2 sd 2051122.3 min 200 (NT_479312.1) # max 196202544 (NC_006088.4) median 6930 # %24.55 masked total, %24.78 masked real # check for duplicate sequences: time faToTwoBit -noMask ../refseq/*_genomic.fna.gz refseq.2bit # real 0m52.302s twoBitDup refseq.2bit # no output is a good result, otherwise, would have to eliminate duplicates # the scripts creating the fasta here will be using this refseq.2bit file # remove it later # new option required to ucscCompositeAgp.pl 2016-04-13 time ~/kent/src/hg/utils/automation/ucscCompositeAgp.pl \ ../refseq/*_genomic.fna.gz ../refseq/*_assembly_structure/Primary_Assembly # constructing refseq.2bit from ../refseq/GCF_000002315.4_Gallus_gallus-5.0_genomic.fna.gz NC_006088.4 chr1 NC_006089.4 chr2 NC_006090.4 chr3 NC_006091.4 chr4 NC_006092.4 chr5 NC_006093.4 chr6 NC_006094.4 chr7 NC_006095.4 chr8 NC_006096.4 chr9 NC_006097.4 chr10 NC_006098.4 chr11 NC_006099.4 chr12 NC_006100.4 chr13 NC_006101.4 chr14 NC_006102.4 chr15 NC_006103.4 chr16 NC_006104.4 chr17 NC_006105.4 chr18 NC_006106.4 chr19 NC_006107.4 chr20 NC_006108.4 chr21 NC_006109.4 chr22 NC_006110.4 chr23 NC_006111.4 chr24 NC_006112.3 chr25 NC_006113.4 chr26 NC_006114.4 chr27 NC_006115.4 chr28 NC_006119.3 chr32 NC_006126.4 chrW NC_006127.4 chrZ NC_008465.3 chr33 NC_008466.3 chrLGE64 NC_028739.1 chr30 NC_028740.1 chr31 real 6m42.463s time ~/kent/src/hg/utils/automation/unplacedWithChroms.pl \ ../refseq/*_assembly_structure/Primary_Assembly ./unlocalizedWithChroms.pl ../refseq/*_assembly_structure/Primary_Assembly # bash syntax here mitoAcc=`grep "^# mitoAcc" ../galGal5.config.ra | awk '{print $NF}'` printf "# mitoAcc %s\n" "$mitoAcc" # mitoAcc NC_001323.1 zcat \ ../refseq/*_assembly_structure/non-nuclear/assem*/AGP/chrMT.comp.agp.gz \ | grep -v "^#" | sed -e "s/^$mitoAcc/chrM/;" > chrM.agp printf ">chrM\n" > chrM.fa twoBitToFa -noMask refseq.2bit:NC_001323.1 stdout | grep -v "^>" >> chrM.fa # verify fasta and AGPs agree time faToTwoBit *.fa.gz test.2bit # real 0m37.886s cat *.agp | checkAgpAndFa stdin test.2bit 2>&1 | tail -4 # All AGP and FASTA entries agree - both files are valid # and no sequence lost from orginal: twoBitToFa test.2bit stdout | faSize stdin # 1230258557 bases (11766024 N's 1218492533 real 1218492533 upper 0 lower) # in 23475 sequences in 1 files # Total size: mean 52407.2 sd 2051122.3 min 200 (chrUn_NT_479312v1) # max 196202544 (chr1) median 6930 # same numbers as above (except for upper/lower masking) # no longer need these temporary 2bit files rm test.2bit refseq.2bit ############################################################################# # Initial database build (DONE - 2016-04-14 - Hiram) # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp galGal5.config.ra) > agp.log 2>&1 # about 2 minutes # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db galGal5.config.ra) > db.log 2>&1 # real 11m9.465s # check in the trackDb files created in TemporaryTrackDbCheckout/ # and add galGal5 to trackDb/makefile # temporary symlink until masked sequence is available cd /hive/data/genomes/galGal5 ln -s `pwd`/galGal5.unmasked.2bit /gbdb/galGal5/galGal5.2bit ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2016-04-14 - Hiram) mkdir /hive/data/genomes/galGal5/bed/cpgIslandsUnmasked cd /hive/data/genomes/galGal5/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/galGal5/galGal5.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku galGal5) > do.log 2>&1 # real 6m59.252s cat fb.galGal5.cpgIslandExtUnmasked.txt # 40554959 bases of 1218501075 (3.328%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2016-04-14 - Hiram) mkdir /hive/data/genomes/galGal5/bed/cytoBand cd /hive/data/genomes/galGal5/bed/cytoBand makeCytoBandIdeo.csh galGal5 ######################################################################### # ucscToINSDC table/track (DONE - 2016-04-14 - Hiram) # the sequence here is working for a 'refseq' assembly with a chrM # situation may be specific depending upon what is available in the assembly mkdir /hive/data/genomes/galGal5/bed/ucscToINSDC cd /hive/data/genomes/galGal5/bed/ucscToINSDC # find accession for chrM grep chrM ../../galGal5.agp # chrM 1 16775 1 O NC_001323.1 1 16775 + # use that accession here: ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../refseq/GCF_*structure/Primary_Assembly NC_001323.1 awk '{printf "%s\t%s\n", $2, $1}' ucscToINSDC.txt | sort > insdcToUcsc.txt # there is no name for chrM/NC_001323.1 sequence, there is no such # sequence with an INSDC name grep -v "^#" ../../refseq/GCF*_assembly_report.txt | cut -f5,7 \ | sed -e 's/na\b/notAvailable/;' | awk '{printf "%s\t%s\n", $2, $1}' \ | sort > insdc.refseq.txt # the sed \b means to match word awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > name.coordinate.tab join insdc.refseq.txt insdcToUcsc.txt | tr '[ ]' '[\t]' | sort -k3 \ | join -2 3 name.coordinate.tab - | tr '[ ]' '[\t]' | cut -f1-3,5 \ > ucscToINSDC.bed # should be same line counts throughout: wc -l * # 23475 insdc.refseq.txt # 23475 insdcToUcsc.txt # 23475 name.coordinate.tab # 23475 ucscToINSDC.bed # 23475 ucscToINSDC.txt cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1 # 27 # use the 27 in this sed sed -e "s/21/27/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab galGal5 ucscToINSDC stdin ucscToINSDC.bed checkTableCoords galGal5 # should cover %100 entirely: featureBits -countGaps galGal5 ucscToINSDC # 1230258557 bases of 1230258557 (100.000%) in intersection ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2016-04-14 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/chicken/galGal5 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" galGal5 \ | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c 24806 AADN.1 47 AC.1 310 AC.2 327 AC.3 74 AC.4 20 AC.5 1 AC.6 1 NC_.1 # implies a rule: '[AN][AC][D0-9_][N0-9][0-9]+(\.[0-9]+)?' # verify this rule will find them all and eliminate them all: hgsql -N -e "select frag from gold;" galGal5 | wc -l # 25586 hgsql -N -e "select frag from gold;" galGal5 \ | egrep -e '[AN][AC][D0-9_][N0-9][0-9]+(\.[0-9]+)?' | wc -l # 25586 hgsql -N -e "select frag from gold;" galGal5 \ | egrep -v -e '[AN][AC][D0-9_][N0-9][0-9]+(\.[0-9]+)?' | wc -l # 0 # hence, add to trackDb/chicken/galGal5/trackDb.ra searchTable gold shortCircuit 1 termRegex [AN][AC][D0-9_][N0-9][0-9]+(\.[0-9]+)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 # verify searches work in the position box ########################################################################## # running repeat masker (DONE - 2016-04-14 - Hiram) mkdir /hive/data/genomes/galGal5/bed/repeatMasker cd /hive/data/genomes/galGal5/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku galGal5) > do.log 2>&1 # real 254m18.208s cat faSize.rmsk.txt # 1230258557 bases (11766024 N's 1218492533 real 1016107420 upper # 202385113 lower) in 23475 sequences in 1 files # Total size: mean 52407.2 sd 2051122.3 min 200 (chrUn_NT_479312v1) # max 196202544 (chr1) median 6930 # %16.45 masked total, %16.61 masked real egrep -i "versi|relea" do.log # RepeatMasker version open-4.0.5 # January 31 2015 (open-4-0-5) version of RepeatMasker # CC RELEASE 20140131; * time featureBits -countGaps galGal5 rmsk # 202387399 bases of 1230258557 (16.451%) in intersection # real 0m17.365s # why is it different than the faSize above ? # because rmsk masks out some N's as well as bases, the faSize count above # separates out the N's from the bases, it doesn't show lower case N's # faster way to get the same result: time hgsql -N -e 'select genoName,genoStart,genoEnd from rmsk;' galGal5 \ | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" # total 202387399.000000 # real 0m6.394s ########################################################################## # running simple repeat (DONE - 2016-04-14 - Hiram) mkdir /hive/data/genomes/galGal5/bed/simpleRepeat cd /hive/data/genomes/galGal5/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ galGal5) > do.log 2>&1 # real 148m20.217s cat fb.simpleRepeat # 89675196 bases of 1218501075 (7.359%) in intersection cd /hive/data/genomes/galGal5 # using the Window Masker result: cd /hive/data/genomes/galGal5 twoBitMask bed/windowMasker/galGal5.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed galGal5.2bit # you can safely ignore the warning about fields >= 13 # add to rmsk after it is done: # twoBitMask galGal5.rmsk.2bit \ # -add bed/simpleRepeat/trfMask.bed galGal5.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa galGal5.2bit stdout | faSize stdin > faSize.galGal5.2bit.txt cat faSize.galGal5.2bit.txt # 1230258557 bases (11766024 N's 1218492533 real 904905740 upper # 313586793 lower) in 23475 sequences in 1 files # Total size: mean 52407.2 sd 2051122.3 min 200 (chrUn_NT_479312v1) # max 196202544 (chr1) median 6930 # %25.49 masked total, %25.74 masked real rm /gbdb/galGal5/galGal5.2bit ln -s `pwd`/galGal5.2bit /gbdb/galGal5/galGal5.2bit ######################################################################### # CREATE MICROSAT TRACK (DONE - 2016-04-14 - Hiram) ssh hgwdev mkdir /cluster/data/galGal5/bed/microsat cd /cluster/data/galGal5/bed/microsat awk '($5==2 || $5==3) && $6 >= 15 && $8 == 100 && $9 == 0 {printf("%s\t%s\t%s\t%dx%s\n", $1, $2, $3, $6, $16);}' \ ../simpleRepeat/simpleRepeat.bed > microsat.bed hgLoadBed galGal5 microsat microsat.bed # Read 2030 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2016-04-14 - Hiram) mkdir /hive/data/genomes/galGal5/bed/windowMasker cd /hive/data/genomes/galGal5/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev galGal5) > do.log 2>&1 # real 53m45.183s # Masking statistics cat faSize.galGal5.cleanWMSdust.txt # 1230258557 bases (11766024 N's 1218492533 real 905812271 upper # 312680262 lower) in 23475 sequences in 1 files # Total size: mean 52407.2 sd 2051122.3 min 200 (chrUn_NT_479312v1) # max 196202544 (chr1) median 6930 # %25.42 masked total, %25.66 masked real cat fb.galGal5.rmsk.windowmaskerSdust.txt # 1061810183 bases of 3236224332 (32.810%) in intersection ########################################################################## # cpgIslands - (DONE - 2016-04-14 - Hiram) mkdir /hive/data/genomes/galGal5/bed/cpgIslands cd /hive/data/genomes/galGal5/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku galGal5) > do.log 2>&1 # real 8m27.957s cat fb.galGal5.cpgIslandExt.txt # 18599973 bases of 1218501075 (1.526%) in intersection ############################################################################## # genscan - (DONE - 2016-04-14,05-13 - Hiram) mkdir /hive/data/genomes/galGal5/bed/genscan cd /hive/data/genomes/galGal5/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku galGal5) > do.log 2>&1 # real 84m26.041s # Completed: 23472 of 23475 jobs # Crashed: 3 jobs # CPU time in finished jobs: 42682s 711.36m 11.86h 0.49d 0.001 y # IO & Wait Time: 60750s 1012.50m 16.88h 0.70d 0.002 y # Average job time: 4s 0.07m 0.00h 0.00d # Longest finished job: 6418s 106.97m 1.78h 0.07d # Submission to last job: 7582s 126.37m 2.11h 0.09d # three jobs needed to be completed with window=2000000 # real 28m9.007s # continuing: time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -continue=makeBed -bigClusterHub=ku galGal5) > makeBed.log 2>&1 # real 3m28.728s cat fb.galGal5.genscan.txt # 27936393 bases of 1218501075 (2.293%) in intersection cat fb.galGal5.genscanSubopt.txt # 28748140 bases of 1218501075 (2.359%) in intersection ######################################################################### # Create kluster run files (DONE - 2016-04-14 - Hiram) # numerator is galGal5 gapless bases "real" as reported by: featureBits -noRandom -noHap galGal5 gap # 11060953 bases of 1010394850 (1.095%) in intersection # ^^^ # denominator is hg19 gapless bases as reported by: # featureBits -noRandom -noHap hg19 gap # 234344806 bases of 2861349177 (8.190%) in intersection # 1024 is threshold used for human -repMatch: calc \( 1010394850 / 2861349177 \) \* 1024 # ( 1010394850 / 2861349177 ) * 1024 = 361.593172 # ==> use -repMatch=350 according to size scaled down from 1024 for human. # and rounded down to nearest 50 cd /hive/data/genomes/galGal5 blat galGal5.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/galGal5.11.ooc \ -repMatch=350 # Wrote 31510 overused 11-mers to jkStuff/galGal5.11.ooc # check non-bridged gaps to see what the typical size is: hgsql -N \ -e 'select * from gap where bridge="no" order by size;' galGal5 \ | sort -k7,7nr | ave -col=7 stdin # minimum gap size is 10 and produces a reasonable number of lifts gapToLift -verbose=2 -minGap=10 galGal5 jkStuff/nonBridged.lft \ -bedFile=jkStuff/nonBridged.bed ######################################################################## # GENBANK AUTO UPDATE (DONE - 2016-04-15,19 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # #organism mrnaCnt estCnt refSeqCnt # Gallus gallus 30510 600534 6360 # edit etc/genbank.conf to add galGal5 just before rheMac2 # galGal5 (chicken) galGal5.serverGenome = /hive/data/genomes/galGal5/galGal5.2bit galGal5.clusterGenome = /hive/data/genomes/galGal5/galGal5.2bit galGal5.ooc = /hive/data/genomes/galGal5/jkStuff/galGal5.11.ooc galGal5.lift = /hive/data/genomes/galGal5/jkStuff/nonBridged.lft galGal5.perChromTables = no galGal5.refseq.mrna.native.pslCDnaFilter = ${ordered.refseq.mrna.native.pslCDnaFilter} galGal5.refseq.mrna.xeno.pslCDnaFilter = ${ordered.refseq.mrna.xeno.pslCDnaFilter} galGal5.genbank.mrna.native.pslCDnaFilter = ${ordered.genbank.mrna.native.pslCDnaFilter} galGal5.genbank.mrna.xeno.pslCDnaFilter = ${ordered.genbank.mrna.xeno.pslCDnaFilter} galGal5.genbank.est.native.pslCDnaFilter = ${ordered.genbank.est.native.pslCDnaFilter} # DO NOT NEED genbank.mrna.xeno except for human, mouse # defaults are fine: genbank.mrna.native refseq.mrna.native refseq.mrna.xeno yes # and genbank.est.native galGal5.downloadDir = galGal5 # galGal5.upstreamGeneTbl = refGene # galGal5.upstreamMaf = multiz7way # /hive/data/genomes/galGal4/bed/multiz7way/species.lst git commit -m "Added galGal5; refs #17168" etc/genbank.conf git push # update /cluster/data/genbank/: make etc-update screen # control this business with a screen since it takes a while cd /cluster/data/genbank time ./bin/gbAlignStep -initial galGal5 # logFile: var/build/logs/2016.04.15-09:58:43.galGal5.initalign.log # real 5678m9.548s tail var/build/logs/2016.04.15-09:58:43.galGal5.initalign.log # hgwdev 2016.04.19-08:29:59 galGal5.initalign: Succeeded: galGal5 # hgwdev 2016.04.19-08:36:52 galGal5.initalign: finish # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.galGal5 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad galGal5 # var/dbload/hgwdev/logs/2016.04.19-08:38:37.galGal5.dbload.log # real 195m36.119s tail -1 var/dbload/hgwdev/logs/2016.04.19-08:38:37.galGal5.dbload.log # hgwdev 2016.04.19-11:54:13 galGal5.dbload: finish # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add galGal5 to: # etc/align.dbs etc/hgwdev.dbs git add etc/align.dbs etc/hgwdev.dbs git commit -m "Added galGal5 - chicken refs #17168" etc/align.dbs etc/hgwdev.dbs git push make etc-update ############################################################################# # augustus gene track (DONE - 2016-04-14 - Hiram) mkdir /hive/data/genomes/galGal5/bed/augustus cd /hive/data/genomes/galGal5/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=chicken -dbHost=hgwdev \ -workhorse=hgwdev galGal5) > do.log 2>&1 cat fb.galGal5.augustusGene.txt # 25248650 bases of 100286401 (25.177%) in intersection ######################################################################### # ncbiRefSeq (DONE - 2016-05-13 - Hiram) mkdir /hive/data/genomes/galGal5/bed/ncbiRefSeq cd /hive/data/genomes/galGal5/bed/ncbiRefSeq # running step wise as this script is still under development time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev \ -stop=download -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.4_Gallus_gallus-5.0 galGal5) > download.log 2>&1 # real 16m29.536s time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -continue=process -bigClusterHub=ku -dbHost=hgwdev \ -stop=process -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.4_Gallus_gallus-5.0 galGal5) > process.log 2>&1 # real 3m58.858s time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -continue=load -bigClusterHub=ku -dbHost=hgwdev \ -stop=load -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.4_Gallus_gallus-5.0 galGal5) > load.log 2>&1 # real 0m33.205s cat fb.ncbiRefSeq.galGal5.txt # 82563006 bases of 1218501075 (6.776%) in intersection featureBits -enrichment galGal5 refGene ncbiRefSeq # refGene 1.181%, ncbiRefSeq 6.776%, both 1.175%, cover 99.49%, # enrich 14.68x ######################################################################### # LIFTOVER TO galGal4 (DONE - 2016-04-15 - Hiram) ssh hgwdev mkdir /hive/data/genomes/galGal5/bed/blat.galGal4.2016-04-15 cd /hive/data/genomes/galGal5/bed/blat.galGal4.2016-04-15 doSameSpeciesLiftOver.pl -verbose=2 \ -debug -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/galGal5/jkStuff/galGal5.11.ooc \ galGal5 galGal4 time (doSameSpeciesLiftOver.pl -verbose=2 \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/galGal5/jkStuff/galGal5.11.ooc \ galGal5 galGal4) > doLiftOverToGalGal4.log 2>&1 # real 86m43.038s # see if the liftOver menus function in the browser from galGal5 to galGal4 ######################################################################### # BLATSERVERS ENTRY (DONE - 2016-04-15 - Hiram) # After getting a blat server assigned by the Blat Server Gods, ssh hgwdev hgsql -e 'INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("galGal5", "blat1b", "17862", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("galGal5", "blat1b", "17863", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################ ## reset default position to MEPE gene (egg shell protein) ## (DONE - 2016-05-13 - Hiram) ssh hgwdev hgsql -e 'update dbDb set defaultPos="chr4:45985744-45991655" where name="galGal5";' hgcentraltest ######################################################################### # all.joiner update, downloads and in pushQ - (DONE - 2016-05-13 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output joinerCheck -database=galGal5 -tableCoverage all.joiner joinerCheck -database=galGal5 -times all.joiner joinerCheck -database=galGal5 -keys all.joiner cd /hive/data/genomes/galGal5 time (makeDownloads.pl galGal5) > downloads.log 2>&1 # real 27m1.207s # now ready for pushQ entry mkdir /hive/data/genomes/galGal5/pushQ cd /hive/data/genomes/galGal5/pushQ time (makePushQSql.pl galGal5) > galGal5.pushQ.sql 2> stderr.out # real 7m21.629s 71 galGal5.pushQ.sql 43 stderr.out 114 total # check for errors in stderr.out, some are OK, e.g.: # WARNING: hgwdev does not have /gbdb/galGal5/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/galGal5/wib/quality.wib # WARNING: hgwdev does not have /gbdb/galGal5/bbi/quality.bw # WARNING: galGal5 does not have seq # WARNING: galGal5 does not have extFile # copy it to hgwbeta scp -p galGal5.pushQ.sql qateam@hgwbeta:/tmp/ ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/galGal5.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly ######################################################################### # UCSC to RefSeq name correspondence (DONE - 2016-05-12 - Hiram) mkdir /hive/data/genomes/galGal5/bed/ucscToRefSeq cd /hive/data/genomes/galGal5/bed/ucscToRefSeq ln -s /hive/data/outside/ncbi/genomes/refseq/vertebrate_other/Gallus_gallus/all_assembly_versions/GCF_000002315.4_Gallus_gallus-5.0/GCF_000002315.4_Gallus_gallus-5.0_assembly_report.txt . # this assembly_report has "UCSC-style-name" in column 10 # but it does not name anything, they are all "na" # columns 5 and 7 are the INSDC and RefSeq names # chrMT fixup in the sed grep -v "^#" GCF_000002315.4_Gallus_gallus-5.0_assembly_report.txt \ | awk -F'\t' '{printf "%s\t%s\n", $5,$7}' | sed -e 's/^na/NC_001323.1/;' | sort > insdc.refSeq.tab # the sed fixes chrM since it doesn't have an INSDC name hgsql -N -e 'select name,chrom,chromStart,chromEnd from ucscToINSDC;' \ galGal5 | sed -e 's/notAvailable/NC_001323.1/;' | sort > insdc.ucsc.tab join insdc.ucsc.tab insdc.refSeq.tab | tr '[ ]' '[\t]' \ | cut -f2- > ucsc.refSeq.tab # when working perfectly, all these tab files have the same line count: wc -l *.tab # 23475 insdc.refSeq.tab # 23475 insdc.ucsc.tab # 23474 ucsc.refSeq.tab export chrSize=`cut -f1 ucsc.refSeq.tab | awk '{print length($0)}' | sort -n | tail -1` echo $chrSize # 27 sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | sed -e 's/INSDC/RefSeq/g;' > ucscToRefSeq.sql hgLoadSqlTab galGal5 ucscToRefSeq ./ucscToRefSeq.sql ucsc.refSeq.tab checkTableCoords galGal5 -table=ucscToRefSeq # should cover %100 all bases: featureBits -countGaps galGal5 ucscToRefSeq # 1230258557 bases of 1230258557 (100.000%) in intersection ######################################################################### # LIFTOVER TO galGal6 (DONE - 2016-10-11 - Hiram) ssh hgwdev mkdir /hive/data/genomes/galGal5/bed/blat.galGal6.2018-10-11 cd /hive/data/genomes/galGal5/bed/blat.galGal6.2018-10-11 doSameSpeciesLiftOver.pl -verbose=2 \ -debug -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/galGal5/jkStuff/galGal5.11.ooc \ galGal5 galGal6 time (doSameSpeciesLiftOver.pl -verbose=2 \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/galGal5/jkStuff/galGal5.11.ooc \ galGal5 galGal6) > doLiftOverToGalGal6.log 2>&1 # real 49m18.791s # see if the liftOver menus function in the browser from galGal5 to galGal6 #########################################################################