# for emacs: -*- mode: sh; -*- # This file describes browser build for the manLeu1 # Drill - Mandrillus leucophaeus # Assembly name: Mleu.le_1.0 # Organism name: Mandrillus leucophaeus (drill) # Isolate: KB7577 # Sex: female # Taxid: 9568 # BioSample: SAMN03121813 # BioProject: PRJNA279492 # Submitter: Baylor College of Medicine # Date: 2015-3-12 # Assembly type: haploid # Release type: major # Assembly level: Scaffold # Genome representation: full # WGS project: JYKQ01 # Assembly method: AllPathsLG v. 48744; Atlas Link v. 1.1; Atlas GapFill v. 2.2 # Expected final version: Yes # Genome coverage: 117.2x # Sequencing technology: Illumina # RefSeq category: Representative Genome # GenBank assembly accession: GCA_000951045.1 # RefSeq assembly accession: GCF_000951045.1 # RefSeq assembly and GenBank assemblies identical: yes # ## Assembly-Units: ## GenBank Unit Accession RefSeq Unit Accession Assembly-Unit name ## GCA_000951085.1 GCF_000951085.1 Primary Assembly ############################################################################# ## There is a RefSeq mitochondrion sequence not included in this release: ## NC_028442.1 from Dec 2015, will have to add that after these contigs. ############################################################################# ############################################################################# # fetch sequence from new style download directory (DONE - 2017-09-20 - Hiram) mkdir -p /hive/data/genomes/manLeu1/refseq cd /hive/data/genomes/manLeu1/refseq rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Mandrillus_leucophaeus/all_assembly_versions/GCF_000951045.1_Mleu.le_1.0/ ./ # measure what we have here: faSize GCF_000951045.1_Mleu.le_1.0_genomic.fna.gz # 3061992840 bases (340585302 N's 2721407538 real 1778481364 upper # 942926174 lower) in 12821 sequences in 1 files # Total size: mean 238826.4 sd 937352.3 min 304 (NW_012108651.1) # max 17385548 (NW_012098097.1) median 2228 # %30.79 masked total, %34.65 masked real ############################################################################# # fixup to UCSC naming scheme (DONE - 2017-09-20 - Hiram) mkdir /hive/data/genomes/manLeu1/ucsc cd /hive/data/genomes/manLeu1/ucsc # add the mitochondrion sequence export mitoAcc=NC_028442.1 wget -O ${mitoAcc}.fa \ "http://www.ncbi.nlm.nih.gov/sviewer/viewer.fcgi?db=nuccore&dopt=fasta&sendto=on&id=$mitoAcc" echo ">chrM" > chrM.fa grep -v "^>" ${mitoAcc}.fa >> chrM.fa export mSize=`faCount chrM.fa | grep total | awk '{print $2}'` /bin/echo -e "chrM\t1\t$mSize\t1\tF\t$mitoAcc\t1\t$mSize\t+" > chrM.agp # verify no duplicate sequences: faToTwoBit ../refseq/GCF_000951045.1_Mleu.le_1.0_genomic.fna.gz refseq.2bit twoBitDup refseq.2bit # should be silent # verify all names are .1: twoBitInfo refseq.2bit stdout | awk '{print $1}' \ | sed -e 's/_[0-9]\+//;' | sort | uniq -c # 12821 NW.1 # since they are all .1, change the names to be v1: zcat ../refseq/GCF_000951045.1_Mleu.le_1.0_assembly_structure/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \ | grep -v "^#" | sed -e 's/\.1/v1/;' > chrUn.manLeu1.agp zcat ../refseq/GCF_000951045.1_Mleu.le_1.0_assembly_structure/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fna.gz \ | sed -e 's/.1 Mandrillus .*/v1/;' > chrUn.manLeu1.fa gzip chrUn.manLeu1.fa chrM.fa cat chrUn.manLeu1.agp chrM.agp > manLeu1.ucsc.agp # verify these two files are compatible: faToTwoBit chrUn.manLeu1.fa.gz chrM.fa.gz test.2bit checkAgpAndFa manLeu1.ucsc.agp test.2bit 2>&1 | tail # All AGP and FASTA entries agree - both files are valid # no longer need these rm -f test.2bit refseq.2bit manLeu1.ucsc.agp ############################################################################# # photo (DONE - 2017-02-14 - Hiram) mkdir /hive/data/genomes/manLeu1/photo cd /hive/data/genomes/manLeu1/photo wget -O photoFile "https://upload.wikimedia.org/wikipedia/commons/e/e0/Drill_%28Mandrillus_leucophaeus%29.jpg" convert -sharpen 0 -normalize -geometry 400x400 -quality 80 photoFile Mandrillus_leucophaeus.jpg cd /hive/data/genomes/manLeu1 printf "photoCreditURL https://commons.wikimedia.org/wiki/User:Cl%C3%A9ment_Bardot photoCreditName WikiMedia Commons: Clément Bardot " > photoReference.txt # the e acute doesn't work in that cut/paste, should be: # WikiMedia Commons: Clément Bardot # for extended HTML markup # check that Mandrillus_leucophaeus.jpg file into source tree # src/hg/htdocs/images/ and copy to /usr/local/apache/htdocs/images/ ############################################################################# # Initial database build (DONE - 2017-09-20 - Hiram) cd /hive/data/genomes/manLeu1 ~/kent/src/hg/utils/automation/prepConfig.pl manLeu1 mammal primate \ refseq/*_assembly_report.txt > manLeu1.config.ra # going to need a mitoAcc ? # fixup the mitoAcc line: mitoAcc none # since it is already setup # verify it looks sane: cat *.ra # config parameters for makeGenomeDb.pl: db manLeu1 clade mammal genomeCladePriority 35 scientificName Mandrillus leucophaeus commonName Drill assemblyDate Mar. 2015 assemblyLabel Baylor College of Medicine assemblyShortLabel Mleu.le_1.0 orderKey 4874 mitoAcc none fastaFiles /hive/data/genomes/manLeu1/ucsc/*.fa.gz agpFiles /hive/data/genomes/manLeu1/ucsc/*.agp # qualFiles none dbDbSpeciesDir primate photoCreditURL https://commons.wikimedia.org/wiki/User:Cl%C3%A9ment_Bardot photoCreditName WikiMedia Commons: Clément Bardot ncbiGenomeId 36538 ncbiAssemblyId 311021 ncbiAssemblyName Mleu.le_1.0 ncbiBioProject 279492 ncbiBioSample SAMN03121813 genBankAccessionID GCF_000951045.1 taxId 9568 # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp manLeu1.config.ra) > agp.log 2>&1 # *** All done! (through the 'agp' step) # real 2m45.522s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db manLeu1.config.ra) > db.log 2>&1 # real 21m32.471s # new trackDb dependencies, fix the script, continue: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=trackDb manLeu1.config.ra) > trackDb.log 2>&1 # real 0m9.862s # fixup the description.html for the e acute: # WikiMedia Commons: Clément Bardot # check in the trackDb files created and add to trackDb/makefile # then, clean up: rm -fr TemporaryTrackDbCheckout/ ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2017-09-21 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/cpgIslandsUnmasked cd /hive/data/genomes/manLeu1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/manLeu1/manLeu1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku manLeu1) > do.log 2>&1 # real 6m48.049s cat fb.manLeu1.cpgIslandExtUnmasked.txt # 19104884 bases of 2721424086 (0.702%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2017-09-21 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/cytoBand cd /hive/data/genomes/manLeu1/bed/cytoBand makeCytoBandIdeo.csh manLeu1 ############################################################################# # running repeat masker (DONE - 2017-09-21 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/repeatMasker cd /hive/data/genomes/manLeu1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku manLeu1) > do.log 2>&1 # real 513m36.866s cat faSize.rmsk.txt # 3062009387 bases (340585302 N's 2721424085 real 1376487602 upper # 1344936483 lower) in 12822 sequences in 1 files # Total size: mean 238809.0 sd 937317.8 min 304 (NW_012108651v1) # max 17385548 (NW_012098097v1) median 2228 # %43.92 masked total, %49.42 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 manLeu1 rmsk # 1346389644 bases of 3062009387 (43.971%) in intersection # real 0m51.890s # 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 - 2017-09-21 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/simpleRepeat cd /hive/data/genomes/manLeu1/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ -trf409 5 manLeu1) > do.log 2>&1 # real 12m30.019s cat fb.simpleRepeat # 89300635 bases of 2721424086 (3.281%) in intersection # using the rmsk result cd /hive/data/genomes/manLeu1 twoBitMask bed/repeatMasker/manLeu1.rmsk.2bit \ -add bed/simpleRepeat/trfMask.bed manLeu1.2bit # you can safely ignore the warning about fields >= 13 # if using windowMasker result: # twoBitMask bed/windowMasker/manLeu1.cleanWMSdust.2bit \ # -add bed/simpleRepeat/trfMask.bed manLeu1.2bit twoBitToFa manLeu1.2bit stdout | faSize stdin > faSize.manLeu1.2bit.txt cat faSize.manLeu1.2bit.txt # 3062009387 bases (340585302 N's 2721424085 real 1375191201 upper # 1346232884 lower) in 12822 sequences in 1 files # Total size: mean 238809.0 sd 937317.8 min 304 (NW_012108651v1) # max 17385548 (NW_012098097v1) median 2228 # %43.97 masked total, %49.47 masked real rm /gbdb/manLeu1/manLeu1.2bit ln -s `pwd`/manLeu1.2bit /gbdb/manLeu1/manLeu1.2bit ########################################################################## # CREATE MICROSAT TRACK (DONE - 2017-09-22 - Hiram) ssh hgwdev mkdir /cluster/data/manLeu1/bed/microsat cd /cluster/data/manLeu1/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, $16);}' \ ../simpleRepeat/simpleRepeat.bed > microsat.bed hgLoadBed manLeu1 microsat microsat.bed # Read 26827 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2017-09-21 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/windowMasker cd /hive/data/genomes/manLeu1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev manLeu1) > do.log 2>&1 # real 265m59.904s # Masking statistics cat faSize.manLeu1.cleanWMSdust.txt # 3062009387 bases (340585302 N's 2721424085 real 1763962353 upper # 957461732 lower) in 12822 sequences in 1 files # Total size: mean 238809.0 sd 937317.8 min 304 (NW_012108651v1) # max 17385548 (NW_012098097v1) median 2228 # %31.27 masked total, %35.18 masked real cat fb.manLeu1.rmsk.windowmaskerSdust.txt # 727173946 bases of 3062009387 (23.748%) in intersection ########################################################################## # cpgIslands - (DONE - 2017-09-22 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/cpgIslands cd /hive/data/genomes/manLeu1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku manLeu1) > do.log 2>&1 & # real 6m52.738s cat fb.manLeu1.cpgIslandExt.txt # 13635922 bases of 2721424086 (0.501%) in intersection ############################################################################## # augustus - (DONE - 2017-09-22 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/augustus cd /hive/data/genomes/manLeu1/bed/augustus export db="manLeu1" export species="human" time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=$species -dbHost=hgwdev -utr -workhorse=hgwdev $db) > do.log 2>&1 # real 116m47.705s cat fb.manLeu1.augustusGene.txt # 45617591 bases of 2721424086 (1.676%) in intersection ######################################################################### # genscan - (DONE - 2017-09-22 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/genscan cd /hive/data/genomes/manLeu1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku manLeu1) > do.log 2>&1 & # real 162m41.174s cat fb.manLeu1.genscan.txt # 50153337 bases of 2721424086 (1.843%) in intersection cat fb.manLeu1.genscanSubopt.txt # 53430104 bases of 2721424086 (1.963%) in intersection ######################################################################### # ucscToINSDC table/track (DONE - 2017-09-22 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/ucscToINSDC cd /hive/data/genomes/manLeu1/bed/ucscToINSDC # check for chrM accession: grep chrM ../../manLeu1.agp # chrM 1 16547 1 F NC_028442.1 1 16547 + # use that accession as an argument to this command ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../refseq/*0_assembly_structure/Primary_Assembly NC_028442.1 # this is actually ucscToRefSeq since this is a RefSeq assembly: sort -k2 ucscToINSDC.txt > ucscToRefSeq.txt rm -f ucscToINSDC.txt awk '{printf "%s\t%s\n", $2, $1}' ucscToRefSeq.txt | sort > refseqToUcsc.txt # need to add the chrM RefSeq and INSDC name with the extra printf: ( printf "NC_028442.1\tKT696596.1\n"; grep -v "^#" ../../refseq/GCF*_assembly_report.txt | cut -f5,7 \ | awk '{printf "%s\t%s\n", $2, $1}') | sort > refseq.insdc.txt awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > name.coordinate.tab join -2 2 refseq.insdc.txt ucscToRefSeq.txt | tr '[ ]' '[\t]' | sort -k3 \ | join -2 3 name.coordinate.tab - | tr '[ ]' '[\t]' | cut -f1-3,5 \ > ucscToINSDC.bed join -2 2 refseq.insdc.txt ucscToRefSeq.txt | tr '[ ]' '[\t]' | sort -k3 \ | join -2 3 name.coordinate.tab - | tr '[ ]' '[\t]' | cut -f1-4 \ > ucscToRefSeq.bed # verify all names are coming through, should be same line count: wc -l * # 12822 name.coordinate.tab # 12822 refseq.insdc.txt # 12822 refseqToUcsc.txt # 12822 ucscToINSDC.bed # 12822 ucscToINSDC.txt # 12822 ucscToRefSeq.bed # 12822 ucscToRefSeq.txt # verify chrM is correct: grep chrM *.bed # ucscToINSDC.bed:chrM 0 16547 KT696596.1 # ucscToRefSeq.bed:chrM 0 16547 NC_028442.1 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 manLeu1 ucscToINSDC stdin ucscToINSDC.bed cut -f1 ucscToRefSeq.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 manLeu1 ucscToRefSeq stdin ucscToRefSeq.bed # checkTableCoords should be silent for no errors: checkTableCoords manLeu1 # should cover %100 entirely: featureBits -countGaps manLeu1 ucscToINSDC # 3062009387 bases of 3062009387 (100.000%) in intersection featureBits -countGaps manLeu1 ucscToRefSeq # 3062009387 bases of 3062009387 (100.000%) in intersection ######################################################################### # add chromAlias table (DONE - 2017-09-22 - Hiram) mkdir /hive/data/genomes/manLeu1/bed/chromAlias cd /hive/data/genomes/manLeu1/bed/chromAlias hgsql -N -e 'select chrom,name,"refseq" from ucscToRefSeq;' manLeu1 \ > ucsc.refseq.tab hgsql -N -e 'select chrom,name,"genbank" from ucscToINSDC;' manLeu1 \ > ucsc.genbank.tab # verify chrM is correct: grep chrM * ucsc.genbank.tab:chrM KT696596.1 genbank ucsc.refseq.tab:chrM NC_028442.1 refseq awk '{printf "%s\t%s\t%s\n", $2,$1,$3}' ucsc.genbank.tab ucsc.refseq.tab \ | sort > manLeu1.chromAlias.tab hgLoadSqlTab manLeu1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ manLeu1.chromAlias.tab cd /hive/data/genomes/manLeu1/bed/chromAlias # add ensembl names 2017-12-14 mkdir previous mv *.tab previous join -t$'\t' ../idKeys/manLeu1.idKeys.txt \ ../../ensembl/ensemblManLeu1.idKeys.txt \ | cut -f2,3 | sort > ucsc.ensembl.tab cut -f1,2 previous/ucsc.refseq.tab > ucsc.refseq.tab cut -f1,2 previous/ucsc.genbank.tab > ucsc.genbank.tab ~/kent/src/hg/utils/automation/chromAlias.pl sort -o manLeu1.chromAlias.tab manLeu1.chromAlias.tab for t in refseq genbank ensembl do c0=`cat ucsc.$t.tab | wc -l` c1=`grep $t manLeu1.chromAlias.tab | wc -l` ok="OK" if [ "$c0" -ne "$c1" ]; then ok="ERROR" fi printf "# checking $t: $c0 =? $c1 $ok\n" done # checking refseq: 12822 =? 12822 OK # checking genbank: 12822 =? 12822 OK # checking ensembl: 12822 =? 12822 OK hgLoadSqlTab manLeu1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ manLeu1.chromAlias.tab ######################################################################### # Create kluster run files (DONE - 2017-09-22 - Hiram) cd /hive/data/genomes/manLeu1 # numerator is manLeu1 gapless bases "real" as reported by: head -1 faSize.manLeu1.2bit.txt # 3062009387 bases (340585302 N's 2721424085 real 1375191201 upper # 1346232884 lower) in 12822 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 \( 2721424085 / 2861349177 \) \* 1024 # ( 2721424085 / 2861349177 ) * 1024 = 973.924569 # ==> use -repMatch=950 according to size scaled down from 1024 for human. # and rounded down to nearest 50 cd /hive/data/genomes/manLeu1 time blat manLeu1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/manLeu1.11.ooc \ -repMatch=950 # Wrote 26979 overused 11-mers to jkStuff/manLeu1.11.ooc # real 0m46.946s # 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;' manLeu1 # | 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 manLeu1 jkStuff/manLeu1.nonBridged.lft \ # -bedFile=jkStuff/manLeu1.nonBridged.bed # hgsql -N \ # -e 'select * from gap where bridge="no" order by size;' manLeu1 \ # | 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 # contig count: 12822, total size: 3062009387, one half size: 1531004693 # cumulative N50 count contig contig size # 1530667727 284 NW_012102255v1 3188400 # 1531004693 one half size # 1533854475 285 NW_012104798v1 3186748 ############################################################################# # GENBANK AUTO UPDATE (DONE - 2017-09-22 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows one mRNA, three refSeq: # organism mrnaCnt estCnt refSeqCnt # Mandrillus leucophaeus 1 0 3 # Mandrillus sphinx 12 0 0 # edit etc/genbank.conf to add manLeu1 just after criGri1 # manLeu1 (Drill - Mandrillus leucophaeus) taxId 9568 manLeu1.serverGenome = /hive/data/genomes/manLeu1/manLeu1.2bit manLeu1.clusterGenome = /hive/data/genomes/manLeu1/manLeu1.2bit manLeu1.ooc = /hive/data/genomes/manLeu1/jkStuff/manLeu1.11.ooc manLeu1.lift = no manLeu1.refseq.mrna.xeno.pslCDnaFilter = ${lowCover.refseq.mrna.xeno.pslCDnaFilter} manLeu1.genbank.mrna.native.pslCDnaFilter = ${lowCover.genbank.mrna.native.pslCDnaFilter} manLeu1.genbank.mrna.xeno.pslCDnaFilter = ${lowCover.genbank.mrna.xeno.pslCDnaFilter} manLeu1.genbank.est.native.pslCDnaFilter = ${lowCover.genbank.est.native.pslCDnaFilter} manLeu1.genbank.est.xeno.pslCDnaFilter = ${lowCover.genbank.est.xeno.pslCDnaFilter} manLeu1.downloadDir = manLeu1 manLeu1.refseq.mrna.native.load = yes manLeu1.refseq.mrna.xeno.load = yes # DO NOT NEED genbank.mrna.xeno except for human, mouse manLeu1.genbank.mrna.xeno.load = no manLeu1.genbank.mrna.native.load = no manLeu1.genbank.est.native.load = no manLeu1.perChromTables = no # And edit src/lib/gbGenome.c to add new species. Adding lines: # static char *manLeuNames[] = {"Mandrillus leucophaeus", NULL}; # {"manLeu", manLeuNames}, git commit -m "Added manLeu1/Drill; refs #20190" \ etc/genbank.conf src/lib/gbGenome.c git push # update /cluster/data/genbank/: make etc-update make install-server screen # control this business with a screen since it takes a while cd /cluster/data/genbank time ./bin/gbAlignStep -initial manLeu1 # logFile: var/build/logs/2017.09.23-20:40:37.manLeu1.initalign.log # real 32m58.945s # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.manLeu1 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad manLeu1 # logFile: var/dbload/hgwdev/logs/2017.09.24-14:05:05.manLeu1.dbload.log # real 5m30.217s # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add manLeu1 to: # vi etc/align.dbs etc/hgwdev.dbs git commit -m "Added manLeu1 - Drill/Mandrillus leucophaeus refs #20190" \ etc/align.dbs etc/hgwdev.dbs git push make etc-update ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2017-09-22 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/primate/manLeu1 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" manLeu1 \ | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c # 246054 JYKQ.1 # 1 NC_.1 # implies a rule: '[JN][CY][K_][Q0-9]+(\.[0-9]+)?' # verify this rule will find them all and eliminate them all: hgsql -N -e "select frag from gold;" manLeu1 | wc -l # 246055 hgsql -N -e "select frag from gold;" manLeu1 \ | egrep -e '[JN][CY][K_][Q0-9]+(\.[0-9]+)?' | wc -l # 246055 hgsql -N -e "select frag from gold;" manLeu1 \ | egrep -v -e '[JN][CY][K_][Q0-9]+(\.[0-9]+)?' | wc -l # 0 # hence, add to trackDb/chicken/manLeu1/trackDb.ra searchTable gold shortCircuit 1 termRegex [JN][CY][K_][Q0-9]+(\.[0-9]+)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 # verify searches work in the position box, check full accession name, # accession name without .1 # truncated accession name produces multiple results # and the two chrM accessions, with and without the .1 and partial name # use: accessionName:n-m to display locations n to m on that accession git commit -m 'add gold/assembly track search rule refs #20190' *.ra git push ######################################################################### # lastz/chain/net swap from hg38 (DONE - 2017-09-25 - Hiram) # alignment to hg38 cd /hive/data/genomes/hg38/bed/lastzManLeu1.2017-09-25 cat fb.hg38.chainManLeu1Link.txt # 2515634714 bases of 3049335806 (82.498%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` hg38 manLeu1) \ > rbest.log 2>&1 & # real 287m28.553s # and for the swap: mkdir /hive/data/genomes/manLeu1/bed/blastz.hg38.swap cd /hive/data/genomes/manLeu1/bed/blastz.hg38.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/hg38/bed/lastzManLeu1.2017-09-25/DEF \ -swap -chainMinScore=5000 -chainLinearGap=medium \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -syntenicNet) > swap.log 2>&1 # real 79m34.951s cat fb.manLeu1.chainHg38Link.txt # 2409820927 bases of 2721424086 (88.550%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` manLeu1 hg38) \ > rbest.log 2>&1 # real 388m23.086s ######################################################################### # lastz/chain/net swap from mm10 (DONE - 2017-09-25 - Hiram) # alignment on mm10 cd /hive/data/genomes/mm10/bed/lastzManLeu1.2017-09-25 cat fb.mm10.chainManLeu1Link.txt # 905203366 bases of 2652783500 (34.123%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` mm10 manLeu1) \ > rbest.log 2>&1 & # real 362m58.840s mkdir /hive/data/genomes/manLeu1/bed/blastz.mm10.swap cd /hive/data/genomes/manLeu1/bed/blastz.mm10.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/mm10/bed/lastzManLeu1.2017-09-25/DEF \ -swap -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 64m55.226s cat fb.manLeu1.chainMm10Link.txt # 895668222 bases of 2721424086 (32.912%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` manLeu1 mm10) \ > rbest.log 2>&1 # real 338m57.422s ############################################################################## # BLATSERVERS ENTRY (DONE - 2017-09-25 - 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 ("manLeu1", "blat1a", "17878", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("manLeu1", "blat1a", "17879", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################ ## reset default position to similar location as hg38 default DONE ## (DONE - 2017-09-26 - Hiram) ssh hgwdev hgsql -e 'update dbDb set defaultPos="NW_012105586v1:3197838-3365219" where name="manLeu1";' hgcentraltest ############################################################################ # all.joiner update, downloads and in pushQ - (DONE - 2017-09-26 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output joinerCheck -database=manLeu1 -tableCoverage all.joiner joinerCheck -database=manLeu1 -times all.joiner joinerCheck -database=manLeu1 -keys all.joiner git commit -m "Added manLeu1 - Drill/Mandrillus leucophaeus refs #20190" \ all.joiner # to get this installed, run a 'make alpha' in the hgTables directory # in a clean source tree that has been fully constructed cd /hive/data/genomes/manLeu1 time (makeDownloads.pl manLeu1) > downloads.log 2>&1 # real 18m27.907s # now ready for pushQ entry mkdir /hive/data/genomes/manLeu1/pushQ cd /hive/data/genomes/manLeu1/pushQ time (makePushQSql.pl -redmineList manLeu1) \ > manLeu1.pushQ.sql 2> stderr.out # real 5m58.381s # check for errors in stderr.out, some are OK, e.g.: # writing redmine listings to # redmine.manLeu1.file.list # redmine.manLeu1.table.list # redmine.manLeu1.releaseLog.txt # WARNING: manLeu1 does not have seq # WARNING: manLeu1 does not have extFile # WARNING: manLeu1 does not have estOrientInfo # Enter the full path names of these listing files: # /hive/data/genomes/manLeu1/pushQ/redmine.manLeu1.file.list # /hive/data/genomes/manLeu1/pushQ/redmine.manLeu1.releaseLog.txt # /hive/data/genomes/manLeu1/pushQ/redmine.manLeu1.table.list # into the Redmine #20190 and set to QA Ready. #########################################################################