# for emacs: -*- mode: sh; -*- # This file describes the browser build for Loa loa/eye worm WS245 version ############################################################################## # download sequence, create UCSC sequence (DONE - 2015-06-24 - Hiram) # both the genbank and the WormBase releases are identical mkdir -p /hive/data/genomes/loaLoa1/genbank cd /hive/data/genomes/loaLoa1/genbank time rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/genbank/invertebrate/Loa_loa/all_assembly_versions/GCA_000183805.3_Loa_loa_V3.1/ ./ # sent 394 bytes received 112378513 bytes 6810842.85 bytes/sec # total size is 112362653 speedup is 1.00 # real 0m15.630s mkdir -p /hive/data/genomes/loaLoa1/ws245 cd /hive/data/genomes/loaLoa1/ws245 wget --no-parent --timestamping -m -nH --cut-dirs=6 \ ftp://ftp.sanger.ac.uk/pub/wormbase/releases/WS245/species/PRJNA60051 mkdir /hive/data/genomes/loaLoa1/ucsc cd /hive/data/genomes/loaLoa1/ucsc zcat ../genbank/GCA_000183805.3_Loa_loa_V3.1_assembly_structure/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fna.gz \ | sed -e 's/.1 Loa .*/v1/;' | gzip -c > loaLoa1.fa.gz zcat ../genbank/GCA_000183805.3_Loa_loa_V3.1_assembly_structure/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \ | sed -e 's/\.1/v1/;' > loaLoa1.agp # obtain photo, CDC public domain photo mkdir /hive/data/genomes/loaLoa1/photo cd /hive/data/genomes/loaLoa1/photo wget --timestamping \ http://www.cdc.gov/dpdx/images/loiasis/L_loa_adult_GA1.jpg # this image is already 381x300 so it can go as is # check into source tree # src/hg/htdocs/images/Loa_loa.jpg # and copy to /usr/local/apache/htdocs/images/Loa_loa.jpg ############################################################################# # Initial database build (DONE - 2015-07-10 - Hiram) cd /hive/data/genomes/loaLoa1 cat << '_EOF_' > loaLoa1.config.ra # Config parameters for makeGenomeDb.pl: db loaLoa1 clade worm genomeCladePriority 68 scientificName Loa loa commonName Eye worm assemblyDate Jul. 2012 assemblyLabel Broad Institute L. loa genome project assemblyShortLabel WS235/L_loa_Cameroon_isolate orderKey 6000 mitoAcc HQ186250.1 fastaFiles /hive/data/genomes/loaLoa1/ucsc/loaLoa1.fa.gz agpFiles /hive/data/genomes/loaLoa1/ucsc/loaLoa1.agp # qualFiles none dbDbSpeciesDir worm photoCreditURL http://www.cdc.gov/parasites/loiasis/disease.html photoCreditName Centers for Disease Control and Prevention ncbiGenomeId 2686 ncbiAssemblyId 195281 ncbiAssemblyName WS235/Loa_loa_V3.1 ncbiBioProject 60051 genBankAccessionID GCA_000183805.3 taxId 7209 '_EOF_' # << happy emacs # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp loaLoa1.config.ra) > agp.log 2>&1 # *** All done! (through the 'agp' step) # real 0m21.029s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db loaLoa1.config.ra) > db.log 2>&1 # real 1m3.894s # check in the trackDb files created and add to trackDb/makefile ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2015-07-10 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/cpgIslandsUnmasked cd /hive/data/genomes/loaLoa1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/loaLoa1/loaLoa1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku loaLoa1) > do.log 2>&1 # real 5m4.522s cat fb.loaLoa1.cpgIslandExtUnmasked.txt # 50875 bases of 87533141 (0.058%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2015-07-10 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/cytoBand cd /hive/data/genomes/loaLoa1/bed/cytoBand makeCytoBandIdeo.csh loaLoa1 ######################################################################### # ucscToINSDC table/track (DONE - 2015-07-10 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/ucscToINSDC cd /hive/data/genomes/loaLoa1/bed/ucscToINSDC # find the chrM accession: grep chrM ../../*.agp # chrM 1 13590 2 F HQ186250.1 1 13590 + # use that as the second argument here: ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../genbank/GCA_*_assembly_structure/Primary_Assembly HQ186250.1 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 * # 5765 name.coordinate.tab # 5765 ucscToINSDC.bed # 5765 ucscToINSDC.txt cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1 # 10 # use the 10 in this sed sed -e "s/21/10/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab loaLoa1 ucscToINSDC stdin ucscToINSDC.bed checkTableCoords loaLoa1 # should cover %100 entirely: featureBits -countGaps loaLoa1 ucscToINSDC # 91379422 bases of 91379422 (100.000%) in intersection ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2015-07-10 - Hiram) hgsql -N -e "select frag from gold;" loaLoa1 | sort | head -3 ADBU02000001.1 ADBU02000002.1 ADBU02000003.1 hgsql -N -e "select frag from gold;" loaLoa1 | sort | tail -2 ADBU02014333.1 HQ186250.1 # verify this rule will find them all or eliminate them all: hgsql -N -e "select frag from gold;" loaLoa1 | wc -l # 14324 hgsql -N -e "select frag from gold;" loaLoa1 \ | egrep -e '(ADBU020|HQ186250)[0-9]*(\.1)?' | wc -l # 14324 hgsql -N -e "select frag from gold;" loaLoa1 \ | egrep -v -e '(ADBU020|HQ186250)[0-9]*(\.1)?' | wc -l # 0 # hence, add to trackDb/worm/loaLoa1/trackDb.ra searchTable gold shortCircuit 1 termRegex (ADBU020|HQ186250)[0-9]*(\.1)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 ########################################################################## # running repeat masker (DONE - 2015-07-10 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/repeatMasker cd /hive/data/genomes/loaLoa1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku loaLoa1) > do.log 2>&1 # real 83m49.429s cat faSize.rmsk.txt # 91379422 bases (3847239 N's 87532183 real 83624537 upper 3907646 lower) # in 5765 sequences in 1 files # Total size: mean 15850.7 sd 64640.7 min 265 (JH717805v1) # max 1325655 (JH712066v1) median 1186 # %4.28 masked total, %4.46 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 loaLoa1 rmsk # 3908609 bases of 91379422 (4.277%) in intersection # real 0m3.289s ########################################################################## # running simple repeat (DONE - 2015-07-10 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/simpleRepeat cd /hive/data/genomes/loaLoa1/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ loaLoa1) > do.log 2>&1 # real 10m57.576s cat fb.simpleRepeat # 2964793 bases of 87533141 (3.387%) in intersection # using the Window Masker result as indicated below ########################################################################## # CREATE MICROSAT TRACK (DONE - 2015-07-15 - Hiram) ssh hgwdev mkdir /cluster/data/loaLoa1/bed/microsat cd /cluster/data/loaLoa1/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 loaLoa1 microsat microsat.bed # Read 2111 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/windowMasker cd /hive/data/genomes/loaLoa1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev loaLoa1) > do.log 2>&1 # real 5m3.906s # Masking statistics cat faSize.loaLoa1.cleanWMSdust.txt # 91379422 bases (3847239 N's 87532183 real 59418740 upper 28113443 lower) # in 5765 sequences in 1 files # Total size: mean 15850.7 sd 64640.7 min 265 (JH717805v1) # max 1325655 (JH712066v1) median 1186 # %30.77 masked total, %32.12 masked real cat fb.loaLoa1.rmsk.windowmaskerSdust.txt # 3520248 bases of 91379422 (3.852%) in intersection # using this Window Masker result for final masking:: cd /hive/data/genomes/loaLoa1 # you can safely ignore the warning about fields >= 13 twoBitMask bed/windowMasker/loaLoa1.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed loaLoa1.2bit # measure the final masking: twoBitToFa loaLoa1.2bit stdout | faSize stdin > faSize.loaLoa1.2bit.txt cat faSize.loaLoa1.2bit.txt # 91379422 bases (3847239 N's 87532183 real 59394205 upper 28137978 lower) # in 5765 sequences in 1 files # Total size: mean 15850.7 sd 64640.7 min 265 (JH717805v1) # max 1325655 (JH712066v1) median 1186 # %30.79 masked total, %32.15 masked real # and reset the symlink rm /gbdb/loaLoa1/loaLoa1.2bit ln -s /hive/data/genomes/loaLoa1/loaLoa1.2bit /gbdb/loaLoa1/loaLoa1.2bit ########################################################################## # cpgIslands - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/cpgIslands cd /hive/data/genomes/loaLoa1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku loaLoa1) > do.log 2>&1 & # real 56m0.652s cat fb.loaLoa1.cpgIslandExt.txt # 40816 bases of 87533141 (0.047%) in intersection ######################################################################### # augustus - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/augustus cd /hive/data/genomes/loaLoa1/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 loaLoa1) > do.log 2>&1 & # real 35m34.472s cat fb.loaLoa1.augustusGene.txt # 6440572 bases of 87533141 (7.358%) in intersection ######################################################################### # genscan - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/loaLoa1/bed/genscan cd /hive/data/genomes/loaLoa1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku loaLoa1) > do.log 2>&1 # real 7m2.652s cat fb.loaLoa1.genscan.txt # 3545367 bases of 87533141 (4.050%) in intersection cat fb.loaLoa1.genscanSubopt.txt # 3368899 bases of 87533141 (3.849%) in intersection ######################################################################## # Create kluster run files (TBD - 2015-07-08 - Hiram) cd /hive/data/genomes/loaLoa1 # numerator is loaLoa1 gapless bases "real" as reported by: head -1 faSize.loaLoa1.2bit.txt # 94076581 bases (7759220 N's 86317361 real 53341517 upper 32975844 lower) # in 9780 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 \( 86317361 / 2861349177 \) \* 1024 # ( 86317361 / 2861349177 ) * 1024 = 30.890665 # ==> use -repMatch=100 since 30 or 50 masks too much cd /hive/data/genomes/loaLoa1 time blat loaLoa1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/loaLoa1.11.ooc \ -repMatch=100 # Wrote 8527 overused 11-mers to jkStuff/loaLoa1.11.ooc # real 0m1.846s # 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;' loaLoa1 \ | ave -tableOut -col=7 stdin # min Q1 median Q3 max mean N sum stddev # 78831 79056 79362 79503 79503 79162.5 4 316650 320.991 # note the minimum non-bridged gap size is 78,831 gapToLift -verbose=2 -minGap=50000 loaLoa1 jkStuff/loaLoa1.nonBridged.lft \ -bedFile=jkStuff/loaLoa1.nonBridged.bed # survey sizes: n50.pl chrom.sizes # reading: chrom.sizes # contig count: 9780, total size: 94076581, one half size: 47038290 # cumulative N50 count contig contig size # 46960459 61 Bmal_v3_scaffold61 194773 # 47038290 one half size # 47151548 62 Bmal_v3_scaffold62 191089 ############################################################################# # 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 # Pristionchus pacificus 97 37470 0 # edit etc/genbank.conf to add loaLoa1 just before priPac1 # loaLoa1 (P. pacificus) loaLoa1.serverGenome = /hive/data/genomes/loaLoa1/loaLoa1.2bit loaLoa1.clusterGenome = /hive/data/genomes/loaLoa1/loaLoa1.2bit loaLoa1.ooc = /hive/data/genomes/loaLoa1/jkStuff/loaLoa1.11.ooc loaLoa1.lift = /hive/data/genomes/loaLoa1/jkStuff/loaLoa1.nonBridged.lft loaLoa1.refseq.mrna.native.pslCDnaFilter = ${ordered.refseq.mrna.native.pslCDnaFilter} loaLoa1.refseq.mrna.xeno.pslCDnaFilter = ${ordered.refseq.mrna.xeno.pslCDnaFilter} loaLoa1.genbank.mrna.native.pslCDnaFilter = ${ordered.genbank.mrna.native.pslCDnaFilter} loaLoa1.genbank.mrna.xeno.pslCDnaFilter = ${ordered.genbank.mrna.xeno.pslCDnaFilter} loaLoa1.genbank.est.native.pslCDnaFilter = ${ordered.genbank.est.native.pslCDnaFilter} loaLoa1.refseq.mrna.native.load = yes loaLoa1.refseq.mrna.xeno.load = yes loaLoa1.refseq.mrna.xeno.loadDesc = yes # DO NOT NEED genbank.mrna.xeno except for human, mouse loaLoa1.genbank.mrna.xeno.load = no loaLoa1.genbank.est.native.load = yes loaLoa1.genbank.est.native.loadDesc = no loaLoa1.downloadDir = loaLoa1 loaLoa1.perChromTables = no git commit -m "Added loaLoa1 - P. pacificus refs #15209" etc/genbank.conf git push # update /cluster/data/genbank/etc/: make etc-update screen # control this business with a screen since it takes a while cd /cluster/data/genbank time ./bin/gbAlignStep -initial loaLoa1 # logFile: var/build/logs/2015.07.02-11:39:01.loaLoa1.initalign.log # real 91m20.362s # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.loaLoa1 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad loaLoa1 # logFile: var/dbload/hgwdev/logs/2015.07.06-09:51:31.loaLoa1.dbload.log # real 18m34.687s # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add loaLoa1 to: # vi etc/align.dbs etc/hgwdev.dbs git commit -m "Added loaLoa1 - 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=loaLoa1 -tableCoverage all.joiner joinerCheck -database=loaLoa1 -times all.joiner joinerCheck -database=loaLoa1 -keys all.joiner cd /hive/data/genomes/loaLoa1 time makeDownloads.pl loaLoa1 > downloads.log 2>&1 # real 13m42.027s # now ready for pushQ entry mkdir /hive/data/genomes/loaLoa1/pushQ cd /hive/data/genomes/loaLoa1/pushQ makePushQSql.pl loaLoa1 > loaLoa1.pushQ.sql 2> stderr.out # check for errors in stderr.out, some are OK, e.g.: # WARNING: hgwdev does not have /gbdb/loaLoa1/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/loaLoa1/wib/quality.wib # WARNING: hgwdev does not have /gbdb/loaLoa1/bbi/qualityBw/quality.bw # WARNING: loaLoa1 does not have seq # WARNING: loaLoa1 does not have extFile # WARNING: loaLoa1 does not have estOrientInfo # WARNING: loaLoa1 does not have mrnaOrientInfo # copy it to hgwbeta scp -p loaLoa1.pushQ.sql qateam@hgwbeta:/tmp ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/loaLoa1.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly #############################################################################