# for emacs: -*- mode: sh; -*- # This file describes how the browser for C. angaria WS245 version is built ############################################################################## # download sequence, create UCSC sequence (DONE - 2015-06-24 - Hiram) mkdir -p /hive/data/genomes/caeSp51/ws245 cd /hive/data/genomes/caeSp51/ws245 wget --no-parent --timestamping -m -nH --cut-dirs=6 \ ftp://ftp.sanger.ac.uk/pub/wormbase/releases/WS245/species/PRJNA194557 mkdir /hive/data/genomes/caeSp51/ucsc cd /hive/data/genomes/caeSp51/ucsc # WormBase contig names are of the pattern: # >Csp5_scaffold_00001 # >Csp5_scaffold_00002 # >Csp5_scaffold_00003 # ... # using the same names as wormBase, no change: zcat ../ws245/PRJNA194557/c_sp5.PRJNA194557.WS245.genomic.fa.gz \ | gzip -c > caeSp51.fa.gz hgFakeAgp caeSp51.fa.gz caeSp51.agp ############################################################################# # Initial database build (DONE - 2015-07-08 - Hiram) cd /hive/data/genomes/caeSp51 cat << '_EOF_' > caeSp51.config.ra # Config parameters for makeGenomeDb.pl: db caeSp51 clade worm genomeCladePriority 68 scientificName Caenorhabditis sp5 ju800 commonName C. sp. 5 ju800 assemblyDate Jan. 2012 assemblyLabel University of Edinburgh Caenorhabditis sp. 5 genome project assemblyShortLabel WS230/Caenorhabditis_sp_5-JU800-1.0 orderKey 3082 mitoAcc none fastaFiles /hive/data/genomes/caeSp51/ucsc/caeSp51.fa.gz agpFiles /hive/data/genomes/caeSp51/ucsc/caeSp51.agp # qualFiles none dbDbSpeciesDir worm photoCreditURL https://www.wormbase.org/species/c_sp5 photoCreditName no photo found ncbiGenomeId 1219476 ncbiAssemblyId 1219476 ncbiAssemblyName WS230/Caenorhabditis_sp_5-JU800-1.0 ncbiBioProject 194557 genBankAccessionID n/a taxId 1219476 '_EOF_' # << happy emacs # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp caeSp51.config.ra) > agp.log 2>&1 # *** All done! (through the 'agp' step) # real 0m23.494s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db caeSp51.config.ra) > db.log 2>&1 # real 1m30.545s # check in the trackDb files created and add to trackDb/makefile ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2015-07-08 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/cpgIslandsUnmasked cd /hive/data/genomes/caeSp51/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/caeSp51/caeSp51.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku caeSp51) > do.log 2>&1 # real 11m21.377s cat fb.caeSp51.cpgIslandExtUnmasked.txt # 3359804 bases of 130562982 (2.573%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2015-07-08 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/cytoBand cd /hive/data/genomes/caeSp51/bed/cytoBand makeCytoBandIdeo.csh caeSp51 ######################################################################### # ucscToINSDC table/track (TBD - 2015-03-20 - Hiram) # do this on assemblies from genbank, not WormBase mkdir /hive/data/genomes/caeSp51/bed/ucscToINSDC cd /hive/data/genomes/caeSp51/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 caeSp51 ucscToINSDC stdin ucscToINSDC.bed checkTableCoords caeSp51 # should cover %100 entirely: featureBits -countGaps caeSp51 ucscToINSDC # 2053849526 bases of 2053849526 (100.000%) in intersection ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2015-07-09 - Hiram) hgsql -N -e "select frag from gold;" caeSp51 | sort | head -1 Csp5_scaffold_00001_1 hgsql -N -e "select frag from gold;" caeSp51 | sort | tail -2 Csp5_scaffold_15260_1 Csp5_scaffold_15261_1 # verify this rule will find them all or eliminate them all: hgsql -N -e "select frag from gold;" caeSp51 | wc -l # 26360 hgsql -N -e "select frag from gold;" caeSp51 \ | egrep -e 'Csp5_scaffold_[0-9]*(_[0-9]*)?' | wc -l # 26360 hgsql -N -e "select frag from gold;" caeSp51 \ | egrep -v -e 'Csp5_scaffold_[0-9]*(_[0-9]*)?' | wc -l # 0 # hence, add to trackDb/worm/caeSp51/trackDb.ra searchTable gold shortCircuit 1 termRegex Csp5_scaffold_[0-9]*(_[0-9]*)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 ########################################################################## # running repeat masker (DONE - 2015-07-08 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/repeatMasker cd /hive/data/genomes/caeSp51/bed/repeatMasker # RM can't recognize this non-standard species name, it does know this one: # "caenorhabditis sp. 5 sensu sudhaus & kiontke 1996" # however, the & is causing trouble, so, using caenorhabditis elegans time (doRepeatMasker.pl -buildDir=`pwd` \ -species="caenorhabditis elegans" \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku caeSp51) > do.log 2>&1 # real 223m36.259s cat faSize.rmsk.txt # 131797386 bases (1406227 N's 130391159 real 128274686 upper # 2116473 lower) in 15261 sequences in 1 files # Total size: mean 8636.2 sd 17173.4 min 200 (Csp5_scaffold_15261) # max 383975 (Csp5_scaffold_00001) median 2600 # %1.61 masked total, %1.62 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 caeSp51 rmsk # 2123601 bases of 131797386 (1.611%) in intersection # real 0m5.909s # 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 - 2015-07-08 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/simpleRepeat cd /hive/data/genomes/caeSp51/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ caeSp51) > do.log 2>&1 # real 12m10.769s cat fb.simpleRepeat # 2716869 bases of 130562982 (2.081%) in intersection # using the Window Masker result as indicated below ########################################################################## # CREATE MICROSAT TRACK (DONE - 2015-07-08 - Hiram) ssh hgwdev mkdir /cluster/data/caeSp51/bed/microsat cd /cluster/data/caeSp51/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 caeSp51 microsat microsat.bed # Read 36 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2015-07-08 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/windowMasker cd /hive/data/genomes/caeSp51/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev caeSp51) > do.log 2>&1 # real 7m45.735s # Masking statistics cat faSize.caeSp51.cleanWMSdust.txt # 131797386 bases (1406227 N's 130391159 real 93741941 upper # 36649218 lower) in 15261 sequences in 1 files # Total size: mean 8636.2 sd 17173.4 min 200 (Csp5_scaffold_15261) # max 383975 (Csp5_scaffold_00001) median 2600 # %27.81 masked total, %28.11 masked real cat fb.caeSp51.rmsk.windowmaskerSdust.txt # 1310574 bases of 131797386 (0.994%) in intersection # using this Window Masker result for final masking:: cd /hive/data/genomes/caeSp51 # you can safely ignore the warning about fields >= 13 twoBitMask bed/windowMasker/caeSp51.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed caeSp51.2bit # measure the final masking: twoBitToFa caeSp51.2bit stdout | faSize stdin > faSize.caeSp51.2bit.txt cat faSize.caeSp51.2bit.txt | sed -e 's/^/ # /;' # 131797386 bases (1406227 N's 130391159 real 93674152 upper # 36717007 lower) in 15261 sequences in 1 files # Total size: mean 8636.2 sd 17173.4 min 200 (Csp5_scaffold_15261) # max 383975 (Csp5_scaffold_00001) median 2600 # %27.86 masked total, %28.16 masked real # and reset the symlink rm /gbdb/caeSp51/caeSp51.2bit ln -s /hive/data/genomes/caeSp51/caeSp51.2bit /gbdb/caeSp51/caeSp51.2bit ########################################################################## # cpgIslands - (DONE - 2015-07-09 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/cpgIslands cd /hive/data/genomes/caeSp51/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku caeSp51) > do.log 2>&1 & # real 19m6.982s cat fb.caeSp51.cpgIslandExt.txt # 5267650 bases of 95237926 (5.531%) in intersection ######################################################################### # augustus - (DONE - 2015-07-09 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/augustus cd /hive/data/genomes/caeSp51/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=caenorhabditis -dbHost=hgwdev \ -workhorse=hgwdev caeSp51) > do.log 2>&1 # real 19m2.915s cat fb.caeSp51.augustusGene.txt # 35056559 bases of 130562982 (26.850%) in intersection ######################################################################### # genscan - (DONE - 2015-07-09 - Hiram) mkdir /hive/data/genomes/caeSp51/bed/genscan cd /hive/data/genomes/caeSp51/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku caeSp51) > do.log 2>&1 # real 17m46.765s cat fb.caeSp51.genscan.txt # 12235030 bases of 130562982 (9.371%) in intersection cat fb.caeSp51.genscanSubopt.txt # 9906961 bases of 130562982 (7.588%) in intersection ######################################################################## # Create kluster run files (TBD - 2015-03-24 - Hiram) cd /hive/data/genomes/caeSp51 # numerator is caeSp51 gapless bases "real" as reported by: head -1 faSize.caeSp51.2bit.txt # 105997628 bases (11442593 N's 94555035 real 56820897 upper 37734138 lower) # in 34621 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 \( 94555035 / 2861349177 \) \* 1024 # ( 94555035 / 2861349177 ) * 1024 = 33.838707 # ==> use -repMatch=100 because 30 or 50 masks too many cd /hive/data/genomes/caeSp51 time blat caeSp51.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/caeSp51.11.ooc \ -repMatch=100 # Wrote 10073 overused 11-mers to jkStuff/caeSp51.11.ooc # real 0m1.898s # there are no non-bridged gaps, do not need to do this # check non-bridged gaps to see what the typical size is: # hgsql -N -e 'select * from gap where bridge="no" order by size;' caeSp51 # | ave -tableOut -col=7 stdin # # min Q1 median Q3 max mean N sum stddev # 50076 58368.8 70128 100495 1.07816e+07 178173 670 1.19376e+08 672006 # note the minimum non-bridged gap size is 50,076 # gapToLift -verbose=2 -minGap=50000 caeSp51 jkStuff/caeSp51.nonBridged.lft \ # -bedFile=jkStuff/caeSp51.nonBridged.bed # hgsql -N \ # -e 'select * from gap where bridge="no" order by size;' caeSp51 \ # | ave -col=7 stdin # not needed: # gapToLift -verbose=2 -minGap=100 bosTau7 jkStuff/nonBridged.lft \ # -bedFile=jkStuff/nonBridged.bed # survey sizes: n50.pl chrom.sizes # reading: chrom.sizes # contig count: 34621, total size: 105997628, one half size: 52998814 # cumulative N50 count contig contig size # 52925057 353 Cang_2012_03_13_00353 80006 # 52998814 one half size # 53004915 354 Cang_2012_03_13_00354 79858 ############################################################################# # GENBANK AUTO UPDATE (TBD - 2015-06-09 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # #organism mrnaCnt estCnt refSeqCnt # Nanorana parkeri 3 0 0 # Nanorana pleskei 3 0 0 # Nanorana yunnanensis 1 0 0 # edit etc/genbank.conf to add caeSp51 just before xenTro7 # caeSp51 (tibetan frog) caeSp51.serverGenome = /hive/data/genomes/caeSp51/caeSp51.2bit caeSp51.clusterGenome = /hive/data/genomes/caeSp51/caeSp51.2bit caeSp51.ooc = /hive/data/genomes/caeSp51/jkStuff/caeSp51.11.ooc caeSp51.lift = no caeSp51.perChromTables = no caeSp51.refseq.mrna.xeno.pslCDnaFilter = ${lowCover.refseq.mrna.xeno.pslCDnaFilter} caeSp51.genbank.mrna.native.pslCDnaFilter = ${lowCover.genbank.mrna.native.pslCDnaFilter} caeSp51.genbank.mrna.xeno.pslCDnaFilter = ${lowCover.genbank.mrna.xeno.pslCDnaFilter} caeSp51.genbank.est.native.pslCDnaFilter = ${lowCover.genbank.est.native.pslCDnaFilter} caeSp51.genbank.est.xeno.pslCDnaFilter = ${lowCover.genbank.est.xeno.pslCDnaFilter} caeSp51.downloadDir = caeSp51 caeSp51.refseq.mrna.native.load = no caeSp51.refseq.mrna.xeno.load = yes caeSp51.refseq.mrna.xeno.loadDesc = yes caeSp51.genbank.mrna.native.load = no caeSp51.genbank.mrna.native.loadDesc = no caeSp51.genbank.mrna.xeno.load = yes caeSp51.genbank.mrna.xeno.loadDesc = yes caeSp51.genbank.est.native.load = no caeSp51.genbank.est.native.loadDesc = no # caeSp51.upstreamGeneTbl = ensGene # Edit src/lib/gbGenome.c to add new species. git commit -m "Added nanPar Nanorana parkeri refs #15056" \ etc/genbank.conf src/lib/gbGenome.c git push # update /cluster/data/genbank/etc/: make etc-update # update /cluster/data/genbank/bin/: make install-server screen # control this business with a screen since it takes a while cd /cluster/data/genbank time ./bin/gbAlignStep -initial caeSp51 # logFile: var/build/logs/2015.05.27-15:45:35.caeSp51.initalign.log # real 1651m57.350s # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.caeSp51 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad caeSp51 # logFile: var/dbload/hgwdev/logs/2015.06.09-14:34:12.caeSp51.dbload.log # real 128m32.009s # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add caeSp51 to: # vi etc/align.dbs etc/hgwdev.dbs git commit -m "Added caeSp51 - Tibetan frog refs #15056" \ 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=caeSp51 -tableCoverage all.joiner joinerCheck -database=caeSp51 -times all.joiner joinerCheck -database=caeSp51 -keys all.joiner cd /hive/data/genomes/caeSp51 time makeDownloads.pl caeSp51 > downloads.log 2>&1 # real 13m42.027s # now ready for pushQ entry mkdir /hive/data/genomes/caeSp51/pushQ cd /hive/data/genomes/caeSp51/pushQ makePushQSql.pl caeSp51 > caeSp51.pushQ.sql 2> stderr.out # check for errors in stderr.out, some are OK, e.g.: # WARNING: hgwdev does not have /gbdb/caeSp51/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/caeSp51/wib/quality.wib # WARNING: hgwdev does not have /gbdb/caeSp51/bbi/qualityBw/quality.bw # WARNING: caeSp51 does not have seq # WARNING: caeSp51 does not have extFile # WARNING: caeSp51 does not have estOrientInfo # WARNING: caeSp51 does not have mrnaOrientInfo # copy it to hgwbeta scp -p caeSp51.pushQ.sql qateam@hgwbeta:/tmp ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/caeSp51.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly #########################################################################