# for emacs: -*- mode: sh; -*- # This file describes browser build for the thaSir1 ######################################################################### # photograph obtained # (DONE - 2018-02-15 - Hiram) mkdir -p /hive/data/genomes/thaSir1/photo cd /hive/data/genomes/thaSir1/photo wget -O photoFile.jpg 'https://upload.wikimedia.org/wikipedia/commons/1/12/Thamnophis_sirtalis_%28Common_Garter_Snake%29.jpg' convert -quality 80 -sharpen 0 -normalize -geometry 400x400 photoFile.jpg Thamnophis_sirtalis.jpg cd /hive/data/genomes/thaSir1 printf 'photoCreditURL\thttps://www.flickr.com/people/19731486@N07?rb=1 photoCreditName Brian Gratwicke (Flickr) ' > photoReference.txt cat photoReference.txt photoCreditURL https://www.flickr.com/people/19731486@N07?rb=1 photoCreditName Brian Gratwicke (Flickr) ######################################################################### # Initial steps (DONE - 2018-03-12 - Hiram) # To start this initialBuild.txt document, from a previous assembly document: mkdir ~/kent/src/hg/makeDb/doc/thaSir1 cd ~/kent/src/hg/makeDb/doc/thaSir1 # best to use a most recent document with similar assembly structure # since it has the latest features and procedures: sed -e 's/petMar3/thaSir1/g; s/PetMar3/ThaSir1/g; s/DONE/TBD/g;' \ ../petMar3/initialBuild.txt > initialBuild.txt mkdir /hive/data/genomes/thaSir1/refseq cd /hive/data/genomes/thaSir1/refseq rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_other/Thamnophis_sirtalis/all_assembly_versions/GCF_001077635.1_Thamnophis_sirtalis-6.0/ ./ # sent 527 bytes received 1379603460 bytes 16927656.28 bytes/sec # total size is 1379432351 speedup is 1.00 # real 1m21.027s # check assembly size for later reference: faSize G*0_genomic.fna.gz # 1424897867 bases (302274245 N's 1122623622 real 743820074 upper # 378803548 lower) in 7930 sequences in 1 files # Total size: mean 179684.5 sd 338299.0 min 880 (NW_013665609.1) # max 3920154 (NW_013657680.1) median 26526 # %26.58 masked total, %33.74 masked real # this information is from the top of # thaSir1/refseq/GCF_001077635.1_Thamnophis_sirtalis-6.0_assembly_report.txt # Assembly name: Thamnophis_sirtalis-6.0 # Organism name: Thamnophis sirtalis (snakes) # Isolate: EDBJR-23777 # Sex: female # Taxid: 35019 # BioSample: SAMN03759628 # BioProject: PRJNA189551 # Submitter: The Genome Institute at Washington University School of Medicine (WUGSC) # Date: 2015-6-26 # Assembly type: haploid # Release type: major # Assembly level: Scaffold # Genome representation: full # WGS project: LFLD01 # Assembly method: ALLPATHS-LG v. May 2015 # Genome coverage: 72x # Sequencing technology: Illumina # RefSeq category: Representative Genome # GenBank assembly accession: GCA_001077635.2 # RefSeq assembly accession: GCF_001077635.1 # RefSeq assembly and GenBank assemblies identical: yes # ## Assembly-Units: ## GenBank Unit Accession RefSeq Unit Accession Assembly-Unit name ## GCA_001077645.2 GCF_001077645.1 Primary Assembly ############################################################################# # establish config.ra file (DONE - Hiram - 2018-03-12) # arguments here are: cd /hive/data/genomes/thaSir1 $HOME/kent/src/hg/utils/automation/prepConfig.pl thaSir1 vertebrate \ reptiles ./refseq/*_assembly_report.txt > thaSir1.config.ra # going to need a mitoAcc ? # fixup commonName to: commonName Garter snake # set mitoAcc to none, the refseq assembly did not have one # reset orderKey to be correct for "Garter snake": 7000 # verify it looks sane cat thaSir1.config.ra # config parameters for makeGenomeDb.pl: db thaSir1 clade vertebrate genomeCladePriority 70 scientificName Thamnophis sirtalis commonName Garter snake assemblyDate Jun. 2015 assemblyLabel The Genome Institute at Washington University School of Medicine (WUGSC) assemblyShortLabel Thamnophis_sirtalis-6.0 orderKey 7000 mitoAcc notFound fastaFiles /hive/data/genomes/thaSir1/ucsc/*.fa.gz agpFiles /hive/data/genomes/thaSir1/ucsc/*.agp # qualFiles none dbDbSpeciesDir reptiles photoCreditURL https://www.flickr.com/people/19731486@N07?rb=1 photoCreditName Brian Gratwicke (Flickr) ncbiGenomeId 16688 ncbiAssemblyId 472161 ncbiAssemblyName Thamnophis_sirtalis-6.0 ncbiBioProject 189551 ncbiBioSample SAMN03759628 genBankAccessionID GCF_001077635.1 taxId 35019 ############################################################################# # setup UCSC named files (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/ucsc cd /hive/data/genomes/thaSir1/ucsc # check for duplicate sequences: time faToTwoBit -noMask ../refseq/G*0_genomic.fna.gz refseq.2bit # real 0m26.340s twoBitDup refseq.2bit # no output is a good result, otherwise, would have to eliminate duplicates # the scripts creating the fasta here will be using this genbank.2bit file # there is no assembly structure directory in the genbank release # simply need to fix the contig names from .1 to v1 time zcat ../refseq/GCF_001077635.1_Thamnophis_sirtalis-6.0_assembly_structure/Primary_Assembly/unplaced_scaffolds/FASTA//unplaced.scaf.fna.gz \ | sed -e 's/.1 Thamnophis.*/v1/;' | gzip -c > chrUn.fa.gz # real 6m9.522s # survey the gap structure: twoBitInfo -nBed refseq.2bit stdout | gzip -c > refseq.N.bed.gz zcat ../refseq/GCF_001077635.1_Thamnophis_sirtalis-6.0_assembly_structure/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \ | sed -e 's/.1\t/v1\t/;' > chrUn.agp # the AGP does not mark all the N's: awk '$5 == "N"' chrUn.agp | ave -col=6 stdin | grep -w total # total 302196072.000000 zcat refseq.N.bed.gz | awk '{print $3-$2}' | ave stdin | grep -w total # total 302274245.000000 calc 302274245 - 302196072 # 302274245 - 302196072 = 78173.000000 # verify fasta and AGPs agree time faToTwoBit chrUn.fa.gz test.2bit # real 0m32.002s time checkAgpAndFa chrUn.agp test.2bit 2>&1 | tail -4 # All AGP and FASTA entries agree - both files are valid # real 0m11.384s # and no sequence lost from original: twoBitToFa test.2bit stdout | faSize stdin # 1424897867 bases (302274245 N's 1122623622 real 1122623622 upper 0 lower) # in 7930 sequences in 1 files # Total size: mean 179684.5 sd 338299.0 min 880 (NW_013665609v1) # max 3920154 (NW_013657680v1) median 26526 # %0.00 masked total, %0.00 masked real # original: # 1424897867 bases (302274245 N's 1122623622 real 743820074 upper # 378803548 lower) in 7930 sequences in 1 files # no longer need these temporary 2bit files rm refseq.2bit test.2bit refseq.N.bed.gz # run up idKeys for later use for ucscToRefSeq and chromAlias mkdir /hive/data/genomes/thaSir1/refseq/idKeys cd /hive/data/genomes/thaSir1/refseq/idKeys faToTwoBit \ ../GCF_001077635.1_Thamnophis_sirtalis-6.0_genomic.fna.gz \ refseq.thaSir1.2bit time (doIdKeys.pl -buildDir=`pwd` -twoBit=`pwd`/refseq.thaSir1.2bit \ refseqThaSir1) > do.log 2>&1 # real 7m41.207s cat refseqThaSir1.keySignature.txt # 7ee1e6311557786133154771c3da3ada ############################################################################# # Initial database build (DONE - 2018-03-12 - Hiram) cd /hive/data/genomes/thaSir1 # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp thaSir1.config.ra) > agp.log 2>&1 # *** All done! (through the 'agp' step) # real 1m22.411s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db thaSir1.config.ra) > db.log 2>&1 # real 9m25.908s # check in the trackDb files created in TemporaryTrackDbCheckout/ # and add thaSir1 to trackDb/makefile # temporary symlink until masked sequence is available cd /hive/data/genomes/thaSir1 ln -s `pwd`/thaSir1.unmasked.2bit /gbdb/thaSir1/thaSir1.2bit ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/cpgIslandsUnmasked cd /hive/data/genomes/thaSir1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/thaSir1/thaSir1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku thaSir1) > do.log 2>&1 # real 3m38.200s cat fb.thaSir1.cpgIslandExtUnmasked.txt # 18072438 bases of 1122701795 (1.610%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/cytoBand cd /hive/data/genomes/thaSir1/bed/cytoBand makeCytoBandIdeo.csh thaSir1 ########################################################################## # run up idKeys files for chromAlias (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/idKeys cd /hive/data/genomes/thaSir1/bed/idKeys time (doIdKeys.pl -twoBit=/hive/data/genomes/thaSir1/thaSir1.unmasked.2bit -buildDir=`pwd` thaSir1) > do.log 2>&1 & # real 7m42.466s cat thaSir1.keySignature.txt # 7ee1e6311557786133154771c3da3ada ########################################################################## # run up gapOverlap track (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/gapOverlap cd /hive/data/genomes/thaSir1/bed/gapOverlap time (doGapOverlap.pl thaSir1 \ -twoBit="/hive/data/genomes/thaSir1/thaSir1.unmasked.2bit" \ -buildDir=`pwd`) > do.log 2>&1 # real 14m16.065s cat fb.thaSir1.gapOverlap.txt # 813702 bases of 1424897867 (0.057%) in intersection ########################################################################## # run up tandemDups track (DONE - 2018-03-15 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/tandemDups cd /hive/data/genomes/thaSir1/bed/tandemDups time (~/kent/src/hg/utils/automation/doTandemDup.pl thaSir1) > do.log 2>&1 # real 30m34.636s cat fb.thaSir1.tandemDups.txt # 31689590 bases of 1424897867 (2.224%) in intersection bigBedInfo thaSir1.tandemDups.bb | sed -e 's/^/# /;' # version: 4 # fieldCount: 13 # hasHeaderExtension: yes # isCompressed: yes # isSwapped: 0 # extraIndexCount: 0 # itemCount: 224,609 # primaryDataSize: 6,499,638 # primaryIndexSize: 160,696 # zoomLevels: 6 # chromCount: 4709 # basesCovered: 421,569,139 # meanDepth (of bases covered): 2.406379 # minDepth: 1.000000 # maxDepth: 109.000000 # std of depth: 2.967056 ########################################################################## # ucscToINSDC and ucscToRefSeq table/track (DONE - 2018-03-12 - Hiram) # the sequence here is working for a 'refseq' assembly # beware of a chrM situation may be specific depending upon what is # available in the assembly # need the idKeys for the genbank sequence: mkdir /hive/data/genomes/thaSir1/genbank/idKeys faToTwoBit /hive/data/outside/ncbi/genomes/genbank/vertebrate_other/Thamnophis_sirtalis/all_assembly_versions/GCA_001077635.2_Thamnophis_sirtalis-6.0/GCA_001077635.2_Thamnophis_sirtalis-6.0_genomic.fna.gz \ genbank.thaSir1.2bit time (doIdKeys.pl -buildDir=`pwd` -twoBit=`pwd`/genbank.thaSir1.2bit \ genbankThaSir1) > do.log 2>&1 # real 21m35.295s cat genbankThaSir1.keySignature.txt # 7ee1e6311557786133154771c3da3ada mkdir /hive/data/genomes/thaSir1/bed/ucscToINSDC cd /hive/data/genomes/thaSir1/bed/ucscToINSDC join -t$'\t' ../idKeys/thaSir1.idKeys.txt \ ../../genbank/idKeys/genbankThaSir1.idKeys.txt | cut -f2- \ | sort -k1,1 | join -t$'\t' <(sort -k1,1 ../../chrom.sizes) - \ | awk '{printf "%s\t0\t%d\t%s\n", $1, $2, $3}' \ | sort -k1,1 -k2,2n > ucscToINSDC.bed join -t$'\t' ../idKeys/thaSir1.idKeys.txt \ ../../refseq/idKeys/refseqThaSir1.idKeys.txt | cut -f2- \ | sort -k1,1 | join -t$'\t' <(sort -k1,1 ../../chrom.sizes) - \ | awk '{printf "%s\t0\t%d\t%s\n", $1, $2, $3}' \ | sort -k1,1 -k2,2n > ucscToRefSeq.bed # should be same line counts throughout, only one for the RefSeq: wc -l * ../../chrom.sizes # 7930 ucscToINSDC.bed # 7930 ucscToRefSeq.bed # 7930 ../../chrom.sizes export chrSize=`cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1` echo $chrSize # 14 # use the 14 in this sed sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab thaSir1 ucscToINSDC stdin ucscToINSDC.bed # should be the same for ucscToRefSeq: export chrSize=`cut -f1 ucscToRefSeq.bed | awk '{print length($0)}' | sort -n | tail -1` echo $chrSize # 14 sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | sed -e 's/INSDC/RefSeq/g;' \ | hgLoadSqlTab thaSir1 ucscToRefSeq stdin ucscToRefSeq.bed # checkTableCoords should be silent checkTableCoords thaSir1 # each should cover %100 entirely: featureBits -countGaps thaSir1 ucscToINSDC # 1424897867 bases of 1424897867 (100.000%) in intersection featureBits -countGaps thaSir1 ucscToRefSeq # 1424897867 bases of 1424897867 (100.000%) in intersection ######################################################################### # add chromAlias table (DONE - 2018-03-14 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/chromAlias cd /hive/data/genomes/thaSir1/bed/chromAlias hgsql -N -e 'select chrom,name from ucscToRefSeq;' thaSir1 \ > ucsc.refseq.tab hgsql -N -e 'select chrom,name from ucscToINSDC;' thaSir1 \ > ucsc.genbank.tab ~/kent/src/hg/utils/automation/chromAlias.pl ucsc.*.tab \ > thaSir1.chromAlias.tab for t in refseq genbank do c0=`cat ucsc.$t.tab | wc -l` c1=`grep $t thaSir1.chromAlias.tab | wc -l` ok="OK" if [ "$c0" -ne "$c1" ]; then ok="ERROR" fi printf "# checking $t: $c0 =? $c1 $ok\n" done # checking refseq: 7930 =? 7930 OK # checking genbank: 7930 =? 7930 OK hgLoadSqlTab thaSir1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ thaSir1.chromAlias.tab ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2018-03-12 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/reptiles/thaSir1 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" thaSir1 \ | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c # 175977 LFLD.1 # implies a rule: 'LFLD01[0-9]{5}p\[0-9]*(.1)?' # verify this rule will find them all and eliminate them all: hgsql -N -e "select frag from gold;" thaSir1 | wc -l # 175977 hgsql -N -e "select frag from gold;" thaSir1 \ | egrep -e 'LFLD01[0-9]{5}[0-9]*(.1)?' | wc -l # 175977 hgsql -N -e "select frag from gold;" thaSir1 \ | egrep -v -e 'LFLD01[0-9]{5}[0-9]*(.1)?' | wc -l # 0 # hence, add to trackDb/reptile/thaSir1/trackDb.ra searchTable gold shortCircuit 1 termRegex LFLD01[0-9]{5}[0-9]*(.1)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 # verify searches work in the position box for these name patterns ########################################################################## # running repeat masker (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/repeatMasker cd /hive/data/genomes/thaSir1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku thaSir1) > do.log 2>&1 & # real 100m48.992s egrep "bases|Total|masked" faSize.rmsk.txt \ | fold -w 75 -s | sed -e 's/^/# /;' # 1424897867 bases (302274245 N's 1122623622 real 1060000477 upper 62623145 # lower) in 7930 sequences in 1 files # Total size: mean 179684.5 sd 338299.0 min 880 (NW_013665609v1) max 3920154 # (NW_013657680v1) median 26526 # %4.39 masked total, %5.58 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 thaSir1 rmsk # 62634553 bases of 1424897867 (4.396%) in intersection # real 0m8.743s # why is it different than the faSize above ? # because rmsk masks out some N's as well as bases, the faSize count above # separates out the N's from the bases, it doesn't show lower case N's # faster way to get the same result on high contig count assemblies: time hgsql -N -e 'select genoName,genoStart,genoEnd from rmsk;' thaSir1 \ | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" # total 62634553.000000 # real 0m5.735s ########################################################################## # running simple repeat (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/simpleRepeat cd /hive/data/genomes/thaSir1/bed/simpleRepeat # using trf409 3 here a bit smaller genome (human == 6) time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ -trf409 3 thaSir1) > do.log 2>&1 & # real 6m3.936s cat fb.simpleRepeat # 33580120 bases of 1122701795 (2.991%) in intersection # adding this trfMask to the other masking cd /hive/data/genomes/thaSir1 # when using the Window Masker result: twoBitMask bed/windowMasker/thaSir1.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed thaSir1.2bit # you can safely ignore the warning about fields >= 13 # when using Rmsk results, add to rmsk after it is done: # twoBitMask thaSir1.rmsk.2bit \ # -add bed/simpleRepeat/trfMask.bed thaSir1.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa thaSir1.2bit stdout | faSize stdin > faSize.thaSir1.2bit.txt egrep "bases|Total|masked" faSize.thaSir1.2bit.txt \ | fold -w 78 -s | sed -e 's/^/# /;' # 1424897867 bases (302274245 N's 1122623622 real 736577963 upper 386045659 # lower) in 7930 sequences in 1 files # Total size: mean 179684.5 sd 338299.0 min 880 (NW_013665609v1) max 3920154 # (NW_013657680v1) median 26526 # %27.09 masked total, %34.39 masked real # reset the symlink rm /gbdb/thaSir1/thaSir1.2bit ln -s `pwd`/thaSir1.2bit /gbdb/thaSir1/thaSir1.2bit ######################################################################### # CREATE MICROSAT TRACK (DONE - 2018-03-12 - Hiram) ssh hgwdev mkdir /cluster/data/thaSir1/bed/microsat cd /cluster/data/thaSir1/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 thaSir1 microsat microsat.bed # Read 30214 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2018-03-12 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/windowMasker cd /hive/data/genomes/thaSir1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev thaSir1) > do.log 2>&1 # real 156m2.988s # Masking statistics egrep "bases|Total|masked" faSize.thaSir1.cleanWMSdust.txt \ | fold -w 78 -s | sed -e 's/^/# /;' # 1424897867 bases (302274245 N's 1122623622 real 736777531 upper 385846091 # lower) in 7930 sequences in 1 files # Total size: mean 179684.5 sd 338299.0 min 880 (NW_013665609v1) max 3920154 # (NW_013657680v1) median 26526 # %27.08 masked total, %34.37 masked real cat fb.thaSir1.rmsk.windowmaskerSdust.txt # 55835996 bases of 1424897867 (3.919%) in intersection ########################################################################## # cpgIslands - (DONE - 2018-03-13 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/cpgIslands cd /hive/data/genomes/thaSir1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku thaSir1) > do.log 2>&1 & # real 3m5.988s cat fb.thaSir1.cpgIslandExt.txt # 10394140 bases of 1122701795 (0.926%) in intersection ############################################################################## # genscan - (DONE - 2018-03-13 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/genscan cd /hive/data/genomes/thaSir1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku thaSir1) > do.log 2>&1 & # real 27m48.117s cat fb.thaSir1.genscan.txt # 26205234 bases of 1122701795 (2.334%) in intersection cat fb.thaSir1.genscanSubopt.txt # 27408024 bases of 1122701795 (2.441%) in intersection ############################################################################# # augustus gene track (DONE - 2018-03-13 - Hiram) mkdir /hive/data/genomes/thaSir1/bed/augustus cd /hive/data/genomes/thaSir1/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=chicken -dbHost=hgwdev -workhorse=hgwdev thaSir1) \ > do.log 2>&1 & # real 70m5.863s cat fb.thaSir1.augustusGene.txt # 26140884 bases of 1122701795 (2.328%) in intersection ############################################################################## # lastz/chain/net swap human/hg38 (DONE - 2018-03-14 - Hiram) # alignment on hg38 cd /hive/data/genomes/hg38/bed/lastzThaSir1.2018-02-13 cat fb.hg38.chainThaSir1Link.txt # 99838535 bases of 3049335806 (3.274%) in intersection cat fb.hg38.chainSynThaSir1Link.txt # 29432396 bases of 3049335806 (0.965%) in intersection cat fb.hg38.chainRBestThaSir1Link.txt # 66452884 bases of 3049335806 (2.179%) in intersection # and for the swap: mkdir /hive/data/genomes/thaSir1/bed/blastz.hg38.swap cd /hive/data/genomes/thaSir1/bed/blastz.hg38.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/hg38/bed/lastzThaSir1.2018-02-13/DEF \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -syntenicNet) > swap.log 2>&1 & # real 16m52.956s cat fb.thaSir1.chainHg38Link.txt # 75915958 bases of 1122701795 (6.762%) in intersection cat fb.thaSir1.chainSynHg38Link.txt # 28865261 bases of 1122701795 (2.571%) in intersection time (doRecipBest.pl -workhorse=hgwdev -load -buildDir=`pwd` thaSir1 hg38)\ > rbest.log 2>&1 # real 269m56.160s cat fb.thaSir1.chainRBestHg38Link.txt # 66545073 bases of 1122701795 (5.927%) in intersection ############################################################################## # lastz/chain/net swap mouse/mm10 (DONE - 2018-03-13 - Hiram) # original alignment to mm10 cd /hive/data/genomes/mm10/bed/lastzThaSir1.2018-03-13 cat fb.mm10.chainThaSir1Link.txt # 78464036 bases of 2652783500 (2.958%) in intersection cat fb.mm10.chainRBestThaSir1Link.txt # 54099233 bases of 2652783500 (2.039%) in intersection # and for the swap mkdir /hive/data/genomes/thaSir1/bed/blastz.mm10.swap cd /hive/data/genomes/thaSir1/bed/blastz.mm10.swap time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/mm10/bed/lastzThaSir1.2018-03-13/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 & # real 11m28.892s cat fb.thaSir1.chainMm10Link.txt # 63814138 bases of 1122701795 (5.684%) in intersection cat fb.thaSir1.chainSynMm10Link.txt # 20728394 bases of 1122701795 (1.846%) in intersection time (doRecipBest.pl -workhorse=hgwdev -load -buildDir=`pwd` thaSir1 mm10) \ > rbest.log 2>&1 & # real 234m31.934s cat fb.thaSir1.chainRBestMm10Link.txt # 54778217 bases of 1122701795 (4.879%) in intersection ############################################################################## # Create kluster run files (DONE - 2018-03-13 - Hiram) # numerator is thaSir1 gapless bases "real" as reported by: featureBits -noRandom -noHap thaSir1 gap # 302196072 bases of 1122701795 (26.917%) in intersection # ^^^ # 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 \( 1122701795 / 2861349177 \) \* 1024 # ( 1122701795 / 2861349177 ) * 1024 = 401.784811 # ==> use -repMatch=400 according to size scaled down from 1024 for human. # and rounded to nearest 50 cd /hive/data/genomes/thaSir1 blat thaSir1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/thaSir1.11.ooc \ -repMatch=400 # Wrote 29799 overused 11-mers to jkStuff/thaSir1.11.ooc # check non-bridged gaps to see what the typical size is: # there are no non-bridged gaps in this contig assembly hgsql -N \ -e 'select * from gap where bridge="no" order by size;' thaSir1 \ | sort -k7,7nr | ave -col=7 stdin # Q1 127644.000000 # median 231535.000000 # Q3 500000.000000 # average 280957.457875 # min 50429.000000 # max 500000.000000 # count 273 # minimum non-bridged gap is 50000 # gapToLift -verbose=2 -minGap=50000 thaSir1 jkStuff/nonBridged.lft \ # -bedFile=jkStuff/nonBridged.bed ######################################################################### # GENBANK AUTO UPDATE (DONE - 2018-03-19 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # #organism mrnaCnt estCnt refSeqCnt # Thamnophis sirtalis 9 0 0 # edit src/lib/gbGenome.c to add thaSirNames definitions # static char *thaSirNames[] = {"Thamnophis sirtalis", NULL}; # . . . # {"thaSir", thaSirNames}, git commit -m "Added thaSir1/Thamnophis sirtalis/Garter snake; refs #16028" src/lib/gbGenome.c git push make install-server # edit etc/genbank.conf to add thaSir1 just after anoCar1 # thaSir1 (Thamnophis sirtalis/Garter snake taxId 35019 GCF_001077635.1) # 7930 contigs thaSir1.serverGenome = /hive/data/genomes/thaSir1/thaSir1.2bit thaSir1.clusterGenome = /hive/data/genomes/thaSir1/thaSir1.2bit thaSir1.ooc = /hive/data/genomes/jkStuff/thaSir1.11.ooc thaSir1.lift = no thaSir1.refseq.mrna.native.pslCDnaFilter = ${lowCover.refseq.mrna.native.pslCDnaFilter} thaSir1.refseq.mrna.xeno.pslCDnaFilter = ${lowCover.refseq.mrna.xeno.pslCDnaFilter} thaSir1.genbank.mrna.native.pslCDnaFilter = ${lowCover.genbank.mrna.native.pslCDnaFilter} thaSir1.genbank.mrna.xeno.pslCDnaFilter = ${lowCover.genbank.mrna.xeno.pslCDnaFilter} thaSir1.genbank.est.native.pslCDnaFilter = ${lowCover.genbank.est.native.pslCDnaFilter} thaSir1.genbank.est.xeno.pslCDnaFilter = ${lowCover.genbank.est.xeno.pslCDnaFilter} # defaults yes: genbank.mrna.native.load genbank.mrna.native.loadDesc # yes: genbank.est.native.load refseq.mrna.native.load # yes: refseq.mrna.native.loadDesc refseq.mrna.xeno.load # yes: refseq.mrna.xeno.loadDesc # defaults no: genbank.mrna.xeno.load genbank.mrna.xeno.loadDesc # no: genbank.est.native.loadDesc genbank.est.xeno.load # no: genbank.est.xeno.loadDesc # DO NOT NEED genbank.mrna.xeno except for human, mouse thaSir1.downloadDir = thaSir1 thaSir1.perChromTables = no # thaSir1.upstreamGeneTbl = ensGene git commit -m \ 'adding thaSir1 - Thamnophis sirtalis/Garter snake refs #16028' \ etc/genbank.conf # enable daily alignment and update of hgwdev # add thaSir1 to: # etc/align.dbs etc/hgwdev.dbs git commit -m 'adding thaSir1 to the update alignments refs #16028' etc/align.dbs etc/hgwdev.dbs git push # update /cluster/data/genbank/: make etc-update # wait a couple days, the automatic daily runs of genbank will pick # this up and get the initial run done ############################################################################# # ncbiRefSeq (DONE - 2018-03-15 - Hiram) # when assembly is from refseq, can run up this track, not so genbank mkdir /hive/data/genomes/thaSir1/bed/ncbiRefSeq cd /hive/data/genomes/thaSir1/bed/ncbiRefSeq time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev \ -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Thamnophis_sirtalis \ GCF_001077635.1_Thamnophis_sirtalis-6.0 thaSir1) > do.log 2>&1 # real 2m49.739s cat fb.ncbiRefSeq.thaSir1.txt # 43026874 bases of 1122701795 (3.832%) in intersection # interesting to intersect with refGene track featureBits thaSir1 -enrichment ncbiRefSeq refGene ############################################################################# # BLATSERVERS ENTRY (DONE - 2018-03-14 - 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 ("thaSir1", "blat1c", "17902", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("thaSir1", "blat1c", "17903", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################ ## reset default position to NOTCH3 gene location as found by blat of the cDNA ## (DONE - 2018-03-14 - Hiram) ## This gene works in concert with the FGF8/Wnt3a gradient to construct ## the extended vertebrate with the 'somitogenesis clock' ## https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3394116/ ssh hgwdev hgsql -e 'update dbDb set defaultPos="NW_013657986v1:596051-618240" where name="thaSir1";' hgcentraltest ######################################################################### # all.joiner update, downloads and in pushQ - (DONE - 2018-03-20 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # verify all required tables are present ~/kent/src/hg/utils/automation/verifyBrowser.pl thaSir1 # 58 tables in database thaSir1 - Garter snake, Thamnophis sirtalis # verified 58 tables, 0 extra tables, 9 optional tables # NCBI RefSeq genes 9 tables # 6 genbank tables found # verified 43 tables, 0 missing tables # fixup all.joiner until this is a clean output joinerCheck -database=thaSir1 -tableCoverage all.joiner joinerCheck -database=thaSir1 -times all.joiner joinerCheck -database=thaSir1 -keys all.joiner cd /hive/data/genomes/thaSir1 time (makeDownloads.pl -workhorse=hgwdev thaSir1) > downloads.log 2>&1 # real 11m32.203s # now ready for pushQ entry mkdir /hive/data/genomes/thaSir1/pushQ cd /hive/data/genomes/thaSir1/pushQ time (makePushQSql.pl -redmineList thaSir1) > thaSir1.pushQ.sql 2> stderr.out # real 3m59.567s # remove the tandemDups from the push sed -i -e "/tandemDups/d" redmine.thaSir1.table.list sed -i -e "/Tandem Dups/d" redmine.thaSir1.releaseLog.txt sed -i -e "/gapOverlap/d" redmine.thaSir1.table.list sed -i -e "/Gap Overlaps/d" redmine.thaSir1.releaseLog.txt # check for errors in stderr.out, some are OK, e.g.: # WARNING: thaSir1 does not have seq # WARNING: thaSir1 does not have extFile ## there are warnings about the RBest and Syn chainNet tables, which we ## are not interested in at this time. They can be left out. # verify the file listings are valid, should be no output to stderr: cat redmine.thaSir1.file.list \ | while read L; do ls -ogL $L; done > /dev/null # to verify the database.table list is correct, should be the same # line count for these two commands: wc -l redmine.thaSir1.table.list # 46 redmine.thaSir1.table.list awk -F'.' '{ printf "hgsql -N -e \"show table status like '"'"'%s'"'"';\" %s\n", $2, $1 }' redmine.thaSir1.table.list | while read L; do eval $L; done | wc -l # 46 # enter the path names to these files in the redmine issue to # make QA Ready: ls `pwd`/redmine* /hive/data/genomes/thaSir1/pushQ/redmine.thaSir1.file.list /hive/data/genomes/thaSir1/pushQ/redmine.thaSir1.releaseLog.txt /hive/data/genomes/thaSir1/pushQ/redmine.thaSir1.table.list #########################################################################