# for emacs: -*- mode: sh; -*- # This file describes how browser build for # Necator americanus/hookworm WS245 version ############################################################################## # download sequence, create UCSC sequence (DONE - 2015-06-24 - Hiram) mkdir -p /hive/data/genomes/necAme1/genbank cd /hive/data/genomes/necAme1/genbank rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/genbank/invertebrate/Necator_americanus/all_assembly_versions/GCA_000507365.1_N_americanus_v1/ ./ mkdir -p /hive/data/genomes/necAme1/ws245 cd /hive/data/genomes/necAme1/ws245 wget --no-parent --timestamping -m -nH --cut-dirs=6 \ ftp://ftp.sanger.ac.uk/pub/wormbase/releases/WS245/species/PRJNA231479 # The genbank and WS245 sequences are identical mkdir /hive/data/genomes/necAme1/ucsc cd /hive/data/genomes/necAme1/ucsc zcat ../genbank/GCA_000507365.1_N_americanus_v1_assembly_structure/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fna.gz \ | sed -e 's/.1 Necator americanus .*/v1/;' | gzip -c > necAme1.fa.gz zcat ../genbank/GCA_000507365.1_N_americanus_v1_assembly_structure/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \ | sed -e 's/\.1/v1/;' > necAme1.agp | sed -e 's/.1 Necator americanus .*/v1/;' | gzip -c > necAme1.fa.gz zcat ../genbank/GCA_000688135.1_Acey_2013.11.30.genDNA_genomic.fna.gz \ | sed -e 's/.1 Ancylostoma ceylanicum .*/v1/;' | gzip -c > necAme1.fa.gz # obtain photo, CDC public domain photo, reuse same photo from ancCey1 wget --timestamping \ http://www.cdc.gov/dpdx/images/hookworm/Hookworm_filariform_A.jpg # this image is slightly larger than 400x300, reduce to 400x300: convert -geometry "400x300" Hookworm_filariform_A.jpg \ Ancylostoma_ceylanicum.jpg # check into source tree # src/hg/htdocs/images/Ancylostoma_ceylanicum.jpg # and copy to /usr/local/apache/htdocs/images/Ancylostoma_ceylanicum.jpg ############################################################################# # Initial database build (DONE - 2015-07-15 - Hiram) cd /hive/data/genomes/necAme1 cat << '_EOF_' > necAme1.config.ra # Config parameters for makeGenomeDb.pl: db necAme1 clade worm genomeCladePriority 86 scientificName Necator americanus commonName Hookworm assemblyDate Dec. 2013 assemblyLabel The Genome Institute at Washington University assemblyShortLabel WS242/N_americanus_v1 orderKey 8679 mitoAcc AJ417719.2 fastaFiles /hive/data/genomes/necAme1/ucsc/necAme1.fa.gz agpFiles /hive/data/genomes/necAme1/ucsc/necAme1.agp # qualFiles none dbDbSpeciesDir worm photoCreditURL http://www.cdc.gov/dpdx/hookworm/ hotoCreditName Centers for Disease Control and Prevention ncbiGenomeId 770 ncbiAssemblyId 88321 ncbiAssemblyName WS242/N_americanus_v1 ncbiBioProject 72135 genBankAccessionID GCA_000507365.1 taxId 51031 '_EOF_' # << happy emacs # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp necAme1.config.ra) > agp.log 2>&1 # *** All done! (through the 'agp' step) # real 0m32.382s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db necAme1.config.ra) > db.log 2>&1 # real 2m16.737s # check in the trackDb files created and add to trackDb/makefile ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/cpgIslandsUnmasked cd /hive/data/genomes/necAme1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/necAme1/necAme1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku necAme1) > do.log 2>&1 # real 68m58.962s cat fb.necAme1.cpgIslandExtUnmasked.txt # 11575810 bases of 208187215 (5.560%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/cytoBand cd /hive/data/genomes/necAme1/bed/cytoBand makeCytoBandIdeo.csh necAme1 ######################################################################### # ucscToINSDC table/track (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/ucscToINSDC cd /hive/data/genomes/necAme1/bed/ucscToINSDC # find the chrM accession: grep chrM ../../*.agp # chrM 1 13605 2 F AJ417719.2 1 13605 + # use that as the second argument here: ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../genbank/GCA_000507365.1_N_americanus_v1_assembly_structure/Primary_Assembly AJ417719.2 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 * # 11865 name.coordinate.tab # 11865 ucscToINSDC.bed # 11865 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 necAme1 ucscToINSDC stdin ucscToINSDC.bed checkTableCoords necAme1 # should cover %100 entirely: featureBits -countGaps necAme1 ucscToINSDC # 244088665 bases of 244088665 (100.000%) in intersection ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2015-07-15 - Hiram) hgsql -N -e "select frag from gold;" necAme1 | sort | head -3 AJ417719.2 ANCG01000001.1 ANCG01000002.1 hgsql -N -e "select frag from gold;" necAme1 | sort | tail -2 ANCG01065212.1 ANCG01065213.1 # verify this rule will find them all or eliminate them all: hgsql -N -e "select frag from gold;" necAme1 | wc -l # 65214 hgsql -N -e "select frag from gold;" necAme1 \ | egrep -e '(ANCG010|AJ417719)[0-9]*(\.[12])?' | wc -l # 65214 hgsql -N -e "select frag from gold;" necAme1 \ | egrep -v -e '(ANCG010|AJ417719)[0-9]*(\.[12])?' | wc -l # 0 # hence, add to trackDb/worm/necAme1/trackDb.ra searchTable gold shortCircuit 1 termRegex (ANCG010|AJ417719)[0-9]*(\.[12])? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 ########################################################################## # running repeat masker (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/repeatMasker cd /hive/data/genomes/necAme1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku necAme1) > do.log 2>&1 # real 44m44.824s cat faSize.rmsk.txt # 244088665 bases (35901450 N's 208187215 real 205566359 upper # 2620856 lower) in 11865 sequences in 1 files # Total size: mean 20572.2 sd 82630.6 min 201 (KI658258v1) # max 1890151 (KI657455v1) median 1315 # %1.07 masked total, %1.26 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 necAme1 rmsk # 2620956 bases of 244088665 (1.074%) in intersection # real 0m4.884s ########################################################################## # running simple repeat (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/simpleRepeat cd /hive/data/genomes/necAme1/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ necAme1) > do.log 2>&1 # real 9m24.231s cat fb.simpleRepeat # 1474019 bases of 208187215 (0.708%) in intersection # using the Window Masker result as indicated below ########################################################################## # CREATE MICROSAT TRACK (DONE - 2015-07-15 - Hiram) ssh hgwdev mkdir /cluster/data/necAme1/bed/microsat cd /cluster/data/necAme1/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 necAme1 microsat microsat.bed # Read 409 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/windowMasker cd /hive/data/genomes/necAme1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev necAme1) > do.log 2>&1 # real 19m24.750s # Masking statistics cat faSize.necAme1.cleanWMSdust.txt # 244088665 bases (35901450 N's 208187215 real 154506078 upper # 53681137 lower) in 11865 sequences in 1 files # Total size: mean 20572.2 sd 82630.6 min 201 (KI658258v1) # max 1890151 (KI657455v1) median 1315 # %21.99 masked total, %25.79 masked real cat fb.necAme1.rmsk.windowmaskerSdust.txt # 2166063 bases of 244088665 (0.887%) in intersection # using this Window Masker result for final masking:: cd /hive/data/genomes/necAme1 # you can safely ignore the warning about fields >= 13 twoBitMask bed/windowMasker/necAme1.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed necAme1.2bit # measure the final masking: twoBitToFa necAme1.2bit stdout | faSize stdin > faSize.necAme1.2bit.txt cat faSize.necAme1.2bit.txt # 244088665 bases (35901450 N's 208187215 real 154468942 upper # 53718273 lower) in 11865 sequences in 1 files # Total size: mean 20572.2 sd 82630.6 min 201 (KI658258v1) # max 1890151 (KI657455v1) median 1315 # %22.01 masked total, %25.80 masked real # and reset the symlink rm /gbdb/necAme1/necAme1.2bit ln -s /hive/data/genomes/necAme1/necAme1.2bit /gbdb/necAme1/necAme1.2bit ########################################################################## # cpgIslands - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/cpgIslands cd /hive/data/genomes/necAme1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku necAme1) > do.log 2>&1 & # real 48m2.637s cat fb.necAme1.cpgIslandExt.txt # 2764104 bases of 208187215 (1.328%) in intersection ######################################################################### # augustus - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/augustus cd /hive/data/genomes/necAme1/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 necAme1) > do.log 2>&1 & # real 64m43.147s cat fb.necAme1.augustusGene.txt # 16924922 bases of 208187215 (8.130%) in intersection ######################################################################### # genscan - (DONE - 2015-07-15 - Hiram) mkdir /hive/data/genomes/necAme1/bed/genscan cd /hive/data/genomes/necAme1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku necAme1) > do.log 2>&1 # real 33m54.384s cat fb.necAme1.genscan.txt # 11294976 bases of 208187215 (5.425%) in intersection cat fb.necAme1.genscanSubopt.txt # 11510060 bases of 208187215 (5.529%) in intersection ######################################################################## # Create kluster run files (TBD - 2015-07-08 - Hiram) cd /hive/data/genomes/necAme1 # numerator is necAme1 gapless bases "real" as reported by: head -1 faSize.necAme1.2bit.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/necAme1 time blat necAme1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/necAme1.11.ooc \ -repMatch=100 # Wrote 8527 overused 11-mers to jkStuff/necAme1.11.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;' necAme1 \ | 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 necAme1 jkStuff/necAme1.nonBridged.lft \ -bedFile=jkStuff/necAme1.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 necAme1 just before priPac1 # necAme1 (P. pacificus) necAme1.serverGenome = /hive/data/genomes/necAme1/necAme1.2bit necAme1.clusterGenome = /hive/data/genomes/necAme1/necAme1.2bit necAme1.ooc = /hive/data/genomes/necAme1/jkStuff/necAme1.11.ooc necAme1.lift = /hive/data/genomes/necAme1/jkStuff/necAme1.nonBridged.lft necAme1.refseq.mrna.native.pslCDnaFilter = ${ordered.refseq.mrna.native.pslCDnaFilter} necAme1.refseq.mrna.xeno.pslCDnaFilter = ${ordered.refseq.mrna.xeno.pslCDnaFilter} necAme1.genbank.mrna.native.pslCDnaFilter = ${ordered.genbank.mrna.native.pslCDnaFilter} necAme1.genbank.mrna.xeno.pslCDnaFilter = ${ordered.genbank.mrna.xeno.pslCDnaFilter} necAme1.genbank.est.native.pslCDnaFilter = ${ordered.genbank.est.native.pslCDnaFilter} necAme1.refseq.mrna.native.load = yes necAme1.refseq.mrna.xeno.load = yes necAme1.refseq.mrna.xeno.loadDesc = yes # DO NOT NEED genbank.mrna.xeno except for human, mouse necAme1.genbank.mrna.xeno.load = no necAme1.genbank.est.native.load = yes necAme1.genbank.est.native.loadDesc = no necAme1.downloadDir = necAme1 necAme1.perChromTables = no git commit -m "Added necAme1 - 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 necAme1 # logFile: var/build/logs/2015.07.02-11:39:01.necAme1.initalign.log # real 91m20.362s # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.necAme1 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad necAme1 # logFile: var/dbload/hgwdev/logs/2015.07.06-09:51:31.necAme1.dbload.log # real 18m34.687s # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add necAme1 to: # vi etc/align.dbs etc/hgwdev.dbs git commit -m "Added necAme1 - 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=necAme1 -tableCoverage all.joiner joinerCheck -database=necAme1 -times all.joiner joinerCheck -database=necAme1 -keys all.joiner cd /hive/data/genomes/necAme1 time makeDownloads.pl necAme1 > downloads.log 2>&1 # real 13m42.027s # now ready for pushQ entry mkdir /hive/data/genomes/necAme1/pushQ cd /hive/data/genomes/necAme1/pushQ makePushQSql.pl necAme1 > necAme1.pushQ.sql 2> stderr.out # check for errors in stderr.out, some are OK, e.g.: # WARNING: hgwdev does not have /gbdb/necAme1/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/necAme1/wib/quality.wib # WARNING: hgwdev does not have /gbdb/necAme1/bbi/qualityBw/quality.bw # WARNING: necAme1 does not have seq # WARNING: necAme1 does not have extFile # WARNING: necAme1 does not have estOrientInfo # WARNING: necAme1 does not have mrnaOrientInfo # copy it to hgwbeta scp -p necAme1.pushQ.sql qateam@hgwbeta:/tmp ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/necAme1.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly #############################################################################