# for emacs: -*- mode: sh; -*- # This file describes how the browser for # Pristionchus exspectatus WS245 version is built ############################################################################## # download sequence, create UCSC sequence (DONE - 2015-06-24 - Hiram) mkdir -p /hive/data/genomes/priExs1/ws245 cd /hive/data/genomes/priExs1/ws245 wget --no-parent --timestamping -m -nH --cut-dirs=6 \ ftp://ftp.sanger.ac.uk/pub/wormbase/releases/WS245/species/PRJEB6009 mkdir /hive/data/genomes/priExs1/ucsc cd /hive/data/genomes/priExs1/ucsc # WormBase contig names are of the pattern: # >scaffold1 1 1285551 # >scaffold4 1 1105670 # >scaffold5 1 796357 # >scaffold6 1 755106 # ... # had to remove 43 duplicate contigs to get this to work # remove the start end coordinates, to maintain equivalence with WormBase: zcat ../ws245/PRJEB6009/p_exspectatus.PRJEB6009.WS245.genomic.fa.gz \ | sed -e 's/ 1 .*//;' | gzip -c > priExs1.fa.gz hgFakeAgp priExs1.fa.gz priExs1.agp # adding a 'Pristionchus' photo: mkdir /hive/data/genomes/priExs1/photo cd /hive/data/genomes/priExs1/photo wget --timestamping \ http://www.pristionchus-sp.de/sites/default/files/styles/large/public/pristionchus.jpg?itok=si96btaL convert -quality 80 -geometry 400x300 "pristionchus.jpg?itok=si96btaL" \ Pristionchus_exspectatus.jpg identify Pristionchus_exspectatus.jpg # Pristionchus_exspectatus.jpg JPEG 400x278 400x278+0+0 ############################################################################# # Initial database build (DONE - 2015-07-14 - Hiram) cd /hive/data/genomes/priExs1 cat << '_EOF_' > priExs1.config.ra db priExs1 clade worm genomeCladePriority 68 scientificName Pristionchus exspectatus commonName P. exspectatus assemblyDate Mar. 2014 assemblyLabel Max Planck Institite for Developmental Genetics assemblyShortLabel WS243/P_exspectatus_v1 orderKey 16000 mitoAcc none fastaFiles /hive/data/genomes/priExs1/ucsc/priExs1.fa.gz agpFiles /hive/data/genomes/priExs1/ucsc/priExs1.agp # qualFiles none dbDbSpeciesDir worm photoCreditURL http://www.molgen.mpg.de/2168/en photoCreditName Max Planck Institite for Developmental Genetics ncbiGenomeId n/a ncbiAssemblyId n/a ncbiAssemblyName WS243/P_exspectatus_v1 ncbiBioProject 243339 genBankAccessionID n/a taxId 1195656 '_EOF_' # << happy emacs # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp priExs1.config.ra) > agp.log 2>&1 # *** All done! (through the 'agp' step) # real 0m19.593s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db priExs1.config.ra) > db.log 2>&1 # real 1m34.864s # check in the trackDb files created and add to trackDb/makefile ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2015-07-14 - Hiram) mkdir /hive/data/genomes/priExs1/bed/cpgIslandsUnmasked cd /hive/data/genomes/priExs1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/priExs1/priExs1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku priExs1) > do.log 2>&1 # real 48m47.434s cat fb.priExs1.cpgIslandExtUnmasked.txt # 8803700 bases of 166917800 (5.274%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2015-07-14 - Hiram) mkdir /hive/data/genomes/priExs1/bed/cytoBand cd /hive/data/genomes/priExs1/bed/cytoBand makeCytoBandIdeo.csh priExs1 ######################################################################### # ucscToINSDC table/track (TBD - 2015-03-20 - Hiram) # this can be run up only on genbank assemblies, not on WormBase assemblies mkdir /hive/data/genomes/priExs1/bed/ucscToINSDC cd /hive/data/genomes/priExs1/bed/ucscToINSDC ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../genbank/GCA_*assembly_structure/Primary_Assembly awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > name.coordinate.tab join name.coordinate.tab ucscToINSDC.txt | tr '[ ]' '[\t]' \ > ucscToINSDC.bed # verify all names are coming through, should be same line count: wc -l * # 25187 name.coordinate.tab # 25187 ucscToINSDC.bed # 25187 ucscToINSDC.txt cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1 # 14 # use the 14 in this sed sed -e "s/21/14/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab priExs1 ucscToINSDC stdin ucscToINSDC.bed checkTableCoords priExs1 # should cover %100 entirely: featureBits -countGaps priExs1 ucscToINSDC # 2053849526 bases of 2053849526 (100.000%) in intersection ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2015-07-14 - Hiram) hgsql -N -e "select frag from gold;" priExs1 | sort | head -3 scaffold1000_1 scaffold1000_2 scaffold1001_1 [JP][Fp][a4][1_][C4][o1][n1][t7][i\.](g[0-9]*)?(\_[0-9]*)? hgsql -N -e "select frag from gold;" priExs1 | sort | tail -2 scaffold99_6 scaffold99_7 # verify this rule will find them all or eliminate them all: hgsql -N -e "select frag from gold;" priExs1 | wc -l # 9775 hgsql -N -e "select frag from gold;" priExs1 \ | egrep -e 'scaffold[0-9]*(\_[0-9]*)?' | wc -l # 9775 hgsql -N -e "select frag from gold;" priExs1 \ | egrep -v -e 'scaffold[0-9]*(\_[0-9]*)?' | wc -l # 0 # hence, add to trackDb/worm/priExs1/trackDb.ra searchTable gold shortCircuit 1 termRegex scaffold[0-9]*(\_[0-9]*)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 ########################################################################## # running repeat masker (DONE - 2015-07-14 - Hiram) mkdir /hive/data/genomes/priExs1/bed/repeatMasker cd /hive/data/genomes/priExs1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku priExs1) > do.log 2>&1 # real 39m51.044s cat faSize.rmsk.txt # 177536987 bases (10621485 N's 166915502 real 163462154 upper 3453348 # lower) in 4369 sequences in 1 files # Total size: mean 40635.6 sd 79495.5 min 945 (scaffold4418) max 1285551 (scaffold1) median 9535 # %1.95 masked total, %2.07 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 priExs1 rmsk # 3453434 bases of 177536987 (1.945%) in intersection # real 0m3.481s # why is it different than the faSize above ? # because rmsk masks out some N's as well as bases, the count above # separates out the N's from the bases, it doesn't show lower case N's ########################################################################## # running simple repeat (DONE - 2015-07-14 - Hiram) mkdir /hive/data/genomes/priExs1/bed/simpleRepeat cd /hive/data/genomes/priExs1/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ priExs1) > do.log 2>&1 # real 12m10.380s cat fb.simpleRepeat # 5329576 bases of 166917800 (3.193%) in intersection # using the Window Masker result as indicated below ########################################################################## # CREATE MICROSAT TRACK (DONE - 2015-07-15 - Hiram) ssh hgwdev mkdir /cluster/data/priExs1/bed/microsat cd /cluster/data/priExs1/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 priExs1 microsat microsat.bed # Read 411 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/priExs1/bed/windowMasker cd /hive/data/genomes/priExs1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev priExs1) > do.log 2>&1 # real 9m16.195s # Masking statistics cat faSize.priExs1.cleanWMSdust.txt # 177536987 bases (10621485 N's 166915502 real 128074383 upper # 38841119 lower) in 4369 sequences in 1 files # Total size: mean 40635.6 sd 79495.5 min 945 (scaffold4418) # max 1285551 (scaffold1) median 9535 # %21.88 masked total, %23.27 masked real cat fb.priExs1.rmsk.windowmaskerSdust.txt # 2940274 bases of 177536987 (1.656%) in intersection # using this Window Masker result for final masking:: cd /hive/data/genomes/priExs1 # you can safely ignore the warning about fields >= 13 twoBitMask bed/windowMasker/priExs1.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed priExs1.2bit # measure the final masking: twoBitToFa priExs1.2bit stdout | faSize stdin > faSize.priExs1.2bit.txt cat faSize.priExs1.2bit.txt # 177536987 bases (10621485 N's 166915502 real 128017169 upper # 38898333 lower) in 4369 sequences in 1 files # Total size: mean 40635.6 sd 79495.5 min 945 (scaffold4418) # max 1285551 (scaffold1) median 9535 # %21.91 masked total, %23.30 masked real # and reset the symlink rm /gbdb/priExs1/priExs1.2bit ln -s /hive/data/genomes/priExs1/priExs1.2bit /gbdb/priExs1/priExs1.2bit ########################################################################## # cpgIslands - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/priExs1/bed/cpgIslands cd /hive/data/genomes/priExs1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku priExs1) > do.log 2>&1 & # real 26m4.376s cat fb.priExs1.cpgIslandExt.txt # 2848910 bases of 166917800 (1.707%) in intersection ######################################################################### # augustus - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/priExs1/bed/augustus cd /hive/data/genomes/priExs1/bed/augustus # XXX this is not specifically correct, the species caenorhabditis # is not necessarily accurate here time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=caenorhabditis -dbHost=hgwdev \ -workhorse=hgwdev priExs1) > do.log 2>&1 # real 25m11.115s cat fb.priExs1.augustusGene.txt # 35136230 bases of 166917800 (21.050%) in intersection ######################################################################### # genscan - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/priExs1/bed/genscan cd /hive/data/genomes/priExs1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku priExs1) > do.log 2>&1 # real 39m26.736s cat fb.priExs1.genscan.txt # 10432088 bases of 166917800 (6.250%) in intersection cat fb.priExs1.genscanSubopt.txt # 9498870 bases of 166917800 (5.691%) in intersection ######################################################################## # Create kluster run files (TBD - 2015-03-24 - Hiram) cd /hive/data/genomes/priExs1 # numerator is priExs1 gapless bases "real" as reported by: head -1 faSize.priExs1.2bit.txt # 177536987 bases (10621485 N's 166915502 real 128017169 upper 38898333 lower) # in 4369 sequences in 1 files # 172510819 bases (19302620 N's 153208199 real 116540993 upper 36667206 lower) # in 18084 sequences in 1 files # numerator is 'real' base count # 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 \( 166915502 / 2861349177 \) \* 1024 # ( 166915502 / 2861349177 ) * 1024 = 59.734574 # ==> use -repMatch=100 according to size scaled down from 1024 for human. # not rounded down to 50 since that produces too many cd /hive/data/genomes/priExs1 time blat priExs1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/priExs1.11.ooc \ -repMatch=100 # Wrote 5972 overused 11-mers to jkStuff/priExs1.11.ooc # real 0m2.812s # there are a few non-bridged gaps # check non-bridged gaps to see what the typical size is: hgsql -N -e 'select * from gap where bridge="no" order by size;' priExs1 \ | ave -tableOut -col=7 stdin # min Q1 median Q3 max mean N sum stddev # 50982 62629.5 70410 83825 97150 72037 33 2.37722e+06 13087. # note the minimum non-bridged gap size is 50,982 gapToLift -verbose=2 -minGap=50000 priExs1 jkStuff/priExs1.nonBridged.lft \ -bedFile=jkStuff/priExs1.nonBridged.bed # survey sizes: n50.pl chrom.sizes # reading: chrom.sizes # contig count: 4369, total size: 177536987, one half size: 88768493 # cumulative N50 count contig contig size # 88630386 341 scaffold300 142436 # 88768493 one half size # 88772631 342 scaffold305 142245 ############################################################################# # GENBANK AUTO UPDATE (DONE - 2015-08-20 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows nothing # for Pristionchus exspectatus: # #organism mrnaCnt estCnt refSeqCnt # Pristionchus exspectatus # edit etc/genbank.conf to add priExs1 just before priPac3 # priExs1 (P. exspectatus/Pristionchus exspectatus) priExs1.serverGenome = /hive/data/genomes/priExs1/priExs1.2bit priExs1.clusterGenome = /hive/data/genomes/priExs1/priExs1.2bit priExs1.ooc = /hive/data/genomes/priExs1/jkStuff/priExs1.11.ooc priExs1.lift = /hive/data/genomes/priExs1/jkStuff/priExs1.nonBridged.lft # the nematodes have small introns priExs1.maxIntron = 100000 priExs1.perChromTables = no priExs1.refseq.mrna.native.pslCDnaFilter = ${lowCover.refseq.mrna.native.pslCDnaFilter} priExs1.refseq.mrna.xeno.pslCDnaFilter = ${lowCover.refseq.mrna.xeno.pslCDnaFilter} priExs1.genbank.mrna.native.pslCDnaFilter = ${lowCover.genbank.mrna.native.pslCDnaFilter} priExs1.genbank.mrna.xeno.pslCDnaFilter = ${lowCover.genbank.mrna.xeno.pslCDnaFilter} priExs1.genbank.est.native.pslCDnaFilter = ${lowCover.genbank.est.native.pslCDnaFilter} # refseq.mrna native and xeno are default yes # genbank.mrna and genbank.est native are default yes, the xeno is default no priExs1.downloadDir = priExs1 git pull git commit -m "Added priExs1 - P. exspectatus refs #15209" etc/genbank.conf git push # update /cluster/data/genbank/etc/: make etc-update # adding priExsNames to src/lib/gbGenome.c git pull git commit -m "Added priExsNames - P. exspectatus refs #15209" \ src/lib/gbGenome.c git push make install-server screen # control this business with a screen since it takes a while cd /cluster/data/genbank time ./bin/gbAlignStep -initial priExs1 # logFile: var/build/logs/2016.08.08-15:06:11.priExs1.initalign.log # real 1428m15.320s # hgwdev 2016.08.09-14:54:12 priExs1.initalign: Succeeded: priExs1 # hgwdev 2016.08.09-14:54:26 priExs1.initalign: finish # To re-do, rm these two sets first: # /cluster/data/genbank/work/initial.priExs1 # /cluster/data/genbank/data/aligned/*/priExs1 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad priExs1 XXX - running Wed Aug 31 09:36:23 PDT 2016 # logFile: var/dbload/hgwdev/logs/2016.08.31-09:36:16.priExs1.dbload.log # real 12m13.931s XXX - failed loading Fri Aug 21 15:01:26 PDT 2015 # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add priExs1 to: # vi etc/align.dbs etc/hgwdev.dbs git commit -m "Added priExs1 - Pristionchus pacificus refs #15209" \ etc/align.dbs etc/hgwdev.dbs git push make etc-update ######################################################################### # all.joiner update, downloads and in pushQ - (TBD - 2015-06-22 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output joinerCheck -database=priExs1 -tableCoverage all.joiner joinerCheck -database=priExs1 -times all.joiner joinerCheck -database=priExs1 -keys all.joiner cd /hive/data/genomes/priExs1 time makeDownloads.pl priExs1 > downloads.log 2>&1 # real 13m42.027s # now ready for pushQ entry mkdir /hive/data/genomes/priExs1/pushQ cd /hive/data/genomes/priExs1/pushQ makePushQSql.pl priExs1 > priExs1.pushQ.sql 2> stderr.out # check for errors in stderr.out, some are OK, e.g.: # WARNING: hgwdev does not have /gbdb/priExs1/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/priExs1/wib/quality.wib # WARNING: hgwdev does not have /gbdb/priExs1/bbi/qualityBw/quality.bw # WARNING: priExs1 does not have seq # WARNING: priExs1 does not have extFile # WARNING: priExs1 does not have estOrientInfo # WARNING: priExs1 does not have mrnaOrientInfo # copy it to hgwbeta scp -p priExs1.pushQ.sql qateam@hgwbeta:/tmp ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/priExs1.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly #############################################################################