# for emacs: -*- mode: sh; -*- # This file describes browser build for the ficAlb1 ######################################################################### # reuse photograph obtained for ficAlb2 experiment # (DONE - 2018-01-04 - Hiram) mkdir /hive/data/genomes/ficAlb1 cd /hive/data/genomes/ficAlb1 cp -p ../ficAlb2/photoReference.txt . cat -A photoReference.txt photoCreditURL^Ihttps://commons.wikimedia.org/wiki/File:Collared_flycatcher_(Ficedula_albicollis).jpg$ photoCreditName^IWikiMedia Commons: Andrej Chudý$ ######################################################################### # Initial steps (DONE - 2018-01-04 - Hiram) # To start this initialBuild.txt document, from a previous assembly document: mkdir ~/kent/src/hg/makeDb/doc/ficAlb1 cd ~/kent/src/hg/makeDb/doc/ficAlb1 # best to use a most recent document since it has the latest features and # procedures: sed -e 's/dipOrd2/ficAlb1/g; s/DipOrd2/FicAlb1/g; s/DONE/TBD/g;' \ ../dipOrd2/initialBuild.txt > initialBuild.txt mkdir /hive/data/genomes/ficAlb1/refseq cd /hive/data/genomes/ficAlb1/refseq time rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_other/Ficedula_albicollis/all_assembly_versions/GCF_000247815.1_FicAlb1.5/ ./ # sent 4417 bytes received 1507883526 bytes 14429549.69 bytes/sec # total size is 1507683242 speedup is 1.00 # real 1m44.830s # check assembly size for later reference: faSize G*5_genomic.fna.gz # 1118343587 bases (16019705 N's 1102323882 real 846711548 upper # 255612334 lower) in 21428 sequences in 1 files # Total size: mean 52190.8 sd 1925906.4 min 200 (NW_004797270.1) # max 157563209 (NC_021673.1) median 460 # %22.86 masked total, %23.19 masked real # this information is from the top of # ficAlb1/refseq/GCF_000247815.1_FicAlb1.5_assembly_report.txt # Assembly name: FicAlb1.5 # Organism name: Ficedula albicollis (collared flycatcher) # Isolate: OC2 # Sex: male # Taxid: 59894 # BioSample: SAMN02981387 # BioProject: PRJNA75089 # Submitter: Uppsala University # Date: 2013-6-27 # Assembly type: haploid # Release type: major # Assembly level: Chromosome # Genome representation: full # WGS project: AGTO02 # Assembly method: SOAPdenovo v. 1.05 # Expected final version: no # Reference guided assembly: de-novo # Genome coverage: 60.0x # Sequencing technology: Illumina GAII; Illumina HiSeq # RefSeq category: Representative Genome # GenBank assembly accession: GCA_000247815.2 # RefSeq assembly accession: GCF_000247815.1 # RefSeq assembly and GenBank assemblies identical: yes # ## Assembly-Units: ## GenBank Unit Accession RefSeq Unit Accession Assembly-Unit name ## GCA_000247825.2 GCF_000247825.1 Primary Assembly ## GCA_000416335.1 GCF_000416335.1 non-nuclear ############################################################################# # establish config.ra file (DONE - Hiram - 2018-01-04) # arguments here are: cd /hive/data/genomes/ficAlb1 $HOME/kent/src/hg/utils/automation/prepConfig.pl ficAlb1 vertebrate \ birds ./refseq/*_assembly_report.txt > ficAlb1.config.ra # verify it looks sane cat ficAlb1.config.ra # config parameters for makeGenomeDb.pl: db ficAlb1 clade vertebrate scientificName Ficedula albicollis commonName Collared flycatcher assemblyDate Jun. 2013 assemblyLabel Uppsala University assemblyShortLabel FicAlb1.5 orderKey 3614 # mitochondrial sequence included in refseq release # mitoAcc NC_021621.1 mitoAcc none fastaFiles /hive/data/genomes/ficAlb1/ucsc/*.fa.gz agpFiles /hive/data/genomes/ficAlb1/ucsc/*.agp # qualFiles none dbDbSpeciesDir birds photoCreditURL https://commons.wikimedia.org/wiki/File:Collared_flycatcher_(Ficedula_albicollis).jpg photoCreditName WikiMedia Commons: Andrej Chudý ncbiGenomeId 11872 ncbiAssemblyId 37801 ncbiAssemblyName FicAlb1.5 ncbiBioProject 75089 ncbiBioSample SAMN02981387 genBankAccessionID GCF_000247815.1 taxId 59894 ############################################################################# # setup UCSC named files (DONE - 2018-01-04 - Hiram) mkdir /hive/data/genomes/ficAlb1/ucsc cd /hive/data/genomes/ficAlb1/ucsc # check for duplicate sequences: time faToTwoBit -noMask ../refseq/G*5_genomic.fna.gz refseq.2bit # real 1m15.460s 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 refseq.2bit file time ~/kent/src/hg/utils/automation/ucscCompositeAgp.pl \ ../refseq/G*5_genomic.fna.gz \ ../refseq/G*5_assembly_structure/Primary_Assembly # real 5m45.505s NC_021671.1 chr1 NC_021672.1 chr1A NC_021673.1 chr2 NC_021674.1 chr3 NC_021675.1 chr4 NC_021676.1 chr4A NC_021677.1 chr5 NC_021678.1 chr6 NC_021679.1 chr7 NC_021680.1 chr8 NC_021681.1 chr9 NC_021682.1 chr10 NC_021683.1 chr11 NC_021684.1 chr12 NC_021685.1 chr13 NC_021686.1 chr14 NC_021687.1 chr15 NC_021688.1 chr17 NC_021689.1 chr18 NC_021690.1 chr19 NC_021691.1 chr20 NC_021692.1 chr21 NC_021693.1 chr22 NC_021694.1 chr23 NC_021695.1 chr24 NC_021696.1 chr25 NC_021697.1 chr26 NC_021698.1 chr27 NC_021699.1 chr28 NC_021700.1 chrZ NC_021701.1 chrLG34 NC_021702.1 chrLG35 NC_021703.1 chrLGE22 time ~/kent/src/hg/utils/automation/unplacedWithChroms.pl \ ../refseq/*_assembly_structure/Primary_Assembly # processed 21350 sequences into chrUn.fa.gz # real 4m8.699s # unlocalized sequences time ~/kent/src/hg/utils/automation/unlocalizedWithChroms.pl \ ../refseq/*_assembly_structure/Primary_Assembly # real 0m5.031s # 21 # LG34 # 26 # 2 # Z # 22 # 1 # 1A # 18 # 4A # 25 # 27 # 28 # LG35 # 4 # LGE22 # processed 44 sequences into chr*_random.gz 16 files # bash syntax here mitoAcc=`grep "^# mitoAcc" ../ficAlb1.config.ra | awk '{print $NF}'` printf "# mitoAcc %s\n" "$mitoAcc" # mitoAcc NC_021621.1 zcat \ ../refseq/*_assembly_structure/non-nuclear/assem*/AGP/chrMT.comp.agp.gz \ | grep -v "^#" | sed -e "s/^$mitoAcc/chrM/;" > chrM.agp printf ">chrM\n" > chrM.fa twoBitToFa -noMask refseq.2bit:$mitoAcc stdout | grep -v "^>" >> chrM.fa gzip chrM.fa # verify fasta and AGPs agree time faToTwoBit chr*.fa.gz test.2bit # real 0m32.917s time cat chr*.agp | checkAgpAndFa stdin test.2bit 2>&1 | tail -4 # All AGP and FASTA entries agree - both files are valid # real 0m15.568s # and no sequence lost from orginal: twoBitToFa test.2bit stdout | faSize stdin # 1118343587 bases (16019705 N's 1102323882 real 1102323882 upper 0 lower) # in 21428 sequences in 1 files # Total size: mean 52190.8 sd 1925906.4 min 200 (chrUn_NW_004797270v1) # max 157563209 (chr2) median 460 # same numbers as above # 1118343587 bases (16019705 N's 1102323882 real 846711548 upper # 255612334 lower) in 21428 sequences in 1 files # no longer need these temporary 2bit files rm refseq.2bit test.2bit ############################################################################# # Initial database build (DONE - 2018-01-04 - Hiram) cd /hive/data/genomes/ficAlb1 # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp ficAlb1.config.ra) > agp.log 2>&1 # *** All done! (through the 'agp' step) # real 1m19.664s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db ficAlb1.config.ra) > db.log 2>&1 # real 9m14.621s # check in the trackDb files created in TemporaryTrackDbCheckout/ # and add ficAlb1 to trackDb/makefile # temporary symlink until masked sequence is available cd /hive/data/genomes/ficAlb1 ln -s `pwd`/ficAlb1.unmasked.2bit /gbdb/ficAlb1/ficAlb1.2bit ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2018-01-04 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/cpgIslandsUnmasked cd /hive/data/genomes/ficAlb1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/ficAlb1/ficAlb1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku ficAlb1) > do.log 2>&1 # real 13m51.828s cat fb.ficAlb1.cpgIslandExtUnmasked.txt # 16741585 bases of 1102325870 (1.519%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2018-01-04 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/cytoBand cd /hive/data/genomes/ficAlb1/bed/cytoBand makeCytoBandIdeo.csh ficAlb1 ########################################################################## # run up idKeys files for chromAlias (DONE - 2018-01-04 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/idKeys cd /hive/data/genomes/ficAlb1/bed/idKeys time (doIdKeys.pl -twoBit=/hive/data/genomes/ficAlb1/ficAlb1.unmasked.2bit -buildDir=`pwd` ficAlb1) > do.log 2>&1 & # real 38m46.994s cat ficAlb1.keySignature.txt # ed755df88638d99bfa2de433f6f975d3 ########################################################################## # ucscToINSDC and ucscToRefSeq table/track (DONE - 2018-01-04 - 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 mkdir /hive/data/genomes/ficAlb1/bed/ucscToINSDC cd /hive/data/genomes/ficAlb1/bed/ucscToINSDC # if there is a chrM, use its INSDC name as a second argument: grep chrM ../../*.agp # chrM 1 16787 1 O NC_021621.1 1 16787 + # this is a RefSeq assembly, use the chrM refSeq name: ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../refseq/GCF_*structure/Primary_Assembly NC_021621.1 # this is actually ucscToRefSeq since this is a RefSeq assembly sort ucscToINSDC.txt > ucscToRefSeq.txt rm -f ucscToINSDC.txt # there is also a genbank release, need to make idKeys to match it mkdir /hive/data/genomes/ficAlb1/bed/ucscToINSDC/genbank cd /hive/data/genomes/ficAlb1/bed/ucscToINSDC/genbank ln -s /hive/data/outside/ncbi/genomes/genbank/vertebrate_other/Ficedula_albicollis/all_assembly_versions/GCA_000247815.2_FicAlb1.5/GCA_000247815.2_FicAlb1.5_genomic.fna.gz . faToTwoBit G*.fna.gz genbank.ficAlb1.2bit time (doIdKeys.pl -buildDir=`pwd` \ -twoBit=`pwd`/genbank.ficAlb1.2bit genbankFicAlb1) > do.log 2>&1 & # real 21m44.755s cd /hive/data/genomes/ficAlb1/bed/ucscToINSDC join -t$'\t' \ ../idKeys/ficAlb1.idKeys.txt genbank/genbankFicAlb1.idKeys.txt \ | cut -f2- | sort > ucscToINSDC.txt awk '{printf "%s\t%s\n", $2, $1}' ucscToRefSeq.txt \ | sort > refSeqToUcsc.txt awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > ucsc.coordinate.tab join -t$'\t' ucsc.coordinate.tab ucscToRefSeq.txt > ucscToRefSeq.bed join -t$'\t' ucsc.coordinate.tab ucscToINSDC.txt > ucscToINSDC.bed # should be same line counts throughout: wc -l * # 21428 refSeqToUcsc.txt # 21428 ucsc.coordinate.tab # 21428 ucscToINSDC.bed # 21428 ucscToINSDC.txt # 21428 ucscToRefSeq.bed # 21428 ucscToRefSeq.txt export chrSize=`cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1` echo $chrSize # 30 # use the 30 in this sed sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab ficAlb1 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 $chrSize # 30 sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | sed -e 's/INSDC/RefSeq/g;' > ucscToRefSeq.sql hgLoadSqlTab ficAlb1 ucscToRefSeq ./ucscToRefSeq.sql ucscToRefSeq.bed # checkTableCoords should be silent checkTableCoords ficAlb1 # each should cover %100 entirely: featureBits -countGaps ficAlb1 ucscToINSDC # 1118343587 bases of 1118343587 (100.000%) in intersection featureBits -countGaps ficAlb1 ucscToRefSeq # 1118343587 bases of 1118343587 (100.000%) in intersection ######################################################################### # add chromAlias table (DONE - 2018-01-05 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/chromAlias cd /hive/data/genomes/ficAlb1/bed/chromAlias # after ensembl idKeys have been made: join -t$'\t' ../idKeys/ficAlb1.idKeys.txt \ ../../ensembl/ensemblFicAlb1.idKeys.txt | cut -f2- > ucsc.ensembl.tab hgsql -N -e 'select chrom,name from ucscToRefSeq;' ficAlb1 \ > ucsc.refseq.tab hgsql -N -e 'select chrom,name from ucscToINSDC;' ficAlb1 \ > ucsc.genbank.tab ~/kent/src/hg/utils/automation/chromAlias.pl ucsc.*.tab \ > ficAlb1.chromAlias.tab for t in refseq genbank ensembl do c0=`cat ucsc.$t.tab | wc -l` c1=`grep $t ficAlb1.chromAlias.tab | wc -l` ok="OK" if [ "$c0" -ne "$c1" ]; then ok="ERROR" fi printf "# checking $t: $c0 =? $c1 $ok\n" done # checking refseq: 21428 =? 21428 OK # checking genbank: 21428 =? 21428 OK # checking ensembl: 21302 =? 21302 OK hgLoadSqlTab ficAlb1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ ficAlb1.chromAlias.tab ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2018-01-05 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/birds/ficAlb1 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" ficAlb1 \ | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c # 30842 AGTO.1 # 1 NC_.1 # implies a rule: '[AN][GC][TO0-9_]+(\.[0-9]+)?' # verify this rule will find them all and eliminate them all: hgsql -N -e "select frag from gold;" ficAlb1 | wc -l # 30843 hgsql -N -e "select frag from gold;" ficAlb1 \ | egrep -e '[AN][GC][TO0-9_]+(\.[0-9]+)?' | wc -l # 30843 hgsql -N -e "select frag from gold;" ficAlb1 \ | egrep -v -e '[AN][GC][TO0-9_]+(\.[0-9]+)?' | wc -l # 0 # hence, add to trackDb/chicken/ficAlb1/trackDb.ra searchTable gold shortCircuit 1 termRegex [AN][GC][TO0-9_]+(\.[0-9]+)? 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 - 2018-01-04 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/repeatMasker cd /hive/data/genomes/ficAlb1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku ficAlb1) > do.log 2>&1 & # real 358m9.085s egrep "bases|Total|masked" faSize.rmsk.txt \ | fold -w 75 -s | sed -e 's/^/# /;' # 1118343587 bases (16019705 N's 1102323882 real 989514048 upper 112809834 # lower) in 21428 sequences in 1 files # Total size: mean 52190.8 sd 1925906.4 min 200 (chrUn_NW_004797270v1) max # 157563209 (chr2) median 460 # %10.09 masked total, %10.23 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 ficAlb1 rmsk # 112813145 bases of 1118343587 (10.088%) in intersection # real 0m11.238s # 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;' ficAlb1 \ | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" # total 112813145.000000 # real 0m4.053s ########################################################################## # running simple repeat (DONE - 2018-01-04 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/simpleRepeat cd /hive/data/genomes/ficAlb1/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 ficAlb1) > do.log 2>&1 & # real 878m59.008s cat fb.simpleRepeat # 127427902 bases of 1102325870 (11.560%) in intersection # adding this trfMask to the other masking cd /hive/data/genomes/ficAlb1 # when using the Window Masker result: twoBitMask bed/windowMasker/ficAlb1.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed ficAlb1.2bit # you can safely ignore the warning about fields >= 13 # when using Rmsk results, add to rmsk after it is done: # twoBitMask ficAlb1.rmsk.2bit \ # -add bed/simpleRepeat/trfMask.bed ficAlb1.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa ficAlb1.2bit stdout | faSize stdin > faSize.ficAlb1.2bit.txt cat faSize.ficAlb1.2bit.txt egrep "bases|Total|masked" faSize.ficAlb1.2bit.txt \ | fold -w 78 -s | sed -e 's/^/# /;' # 1118343587 bases (16019705 N's 1102323882 real 837369763 upper 264954119 # lower) in 21428 sequences in 1 files # Total size: mean 52190.8 sd 1925906.4 min 200 (chrUn_NW_004797270v1) max # 157563209 (chr2) median 460 # %23.69 masked total, %24.04 masked real # reset the symlink rm /gbdb/ficAlb1/ficAlb1.2bit ln -s `pwd`/ficAlb1.2bit /gbdb/ficAlb1/ficAlb1.2bit ######################################################################### # CREATE MICROSAT TRACK (DONE - 2018-01-05 - Hiram) ssh hgwdev mkdir /cluster/data/ficAlb1/bed/microsat cd /cluster/data/ficAlb1/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 ficAlb1 microsat microsat.bed # Read 1858 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2018-01-05 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/windowMasker cd /hive/data/genomes/ficAlb1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev ficAlb1) > do.log 2>&1 # real 53m58.972s # Masking statistics cat faSize.ficAlb1.cleanWMSdust.txt egrep "bases|Total|masked" faSize.ficAlb1.cleanWMSdust.txt \ | fold -w 78 -s | sed -e 's/^/# /;' # 1118343587 bases (16019705 N's 1102323882 real 838249378 upper 264074504 # lower) in 21428 sequences in 1 files # Total size: mean 52190.8 sd 1925906.4 min 200 (chrUn_NW_004797270v1) max # 157563209 (chr2) median 460 # %23.61 masked total, %23.96 masked real cat fb.ficAlb1.rmsk.windowmaskerSdust.txt # 542455834 bases of 2236368823 (24.256%) in intersection ########################################################################## # cpgIslands - (DONE - 2018-01-05 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/cpgIslands cd /hive/data/genomes/ficAlb1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku ficAlb1) > do.log 2>&1 & # real 3m25.988s cat fb.ficAlb1.cpgIslandExt.txt # 10735328 bases of 1102325870 (0.974%) in intersection ############################################################################## # genscan - (DONE - 2018-01-05 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/genscan cd /hive/data/genomes/ficAlb1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku ficAlb1) > do.log 2>&1 & # real 22m10.156s # 2 jobs failed, running with 2000000 size window # real 3m24.459s # continuing time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -continue=makeBed -bigClusterHub=ku ficAlb1) > makeBed.log 2>&1 & # real 1m44.308s cat fb.ficAlb1.genscan.txt # 19263301 bases of 1102325870 (1.748%) in intersection cat fb.ficAlb1.genscanSubopt.txt # 22956236 bases of 1102325870 (2.083%) in intersection ############################################################################# # Ensembl genes (TBD - 2018-01-05 - Hiram) # the v91 release for Ensembl is for a previous assembly from this one # does not work on this assembly, the Ensembl genome is an older # obsolete assembly # after chromAlias work is done: cd /hive/data/genomes/ficAlb1/jkStuff join -t$'\t' <(sort -k1,1 ../chrom.sizes) \ <(sort ../bed/chromAlias/ucsc.ensembl.tab) \ | awk '{printf "0\t%s\t%d\t%s\t%d\n", $3,$2,$1,$2}' > ensToUcsc.lift cd /hive/data/genomes/ficAlb1 printf "# required db variable db ficAlb1 # specific lifting to translate names: liftUp /hive/data/genomes/ficAlb1/jkStuff/ensToUcsc.lift " > ficAlb1.ensGene.ra time (doEnsGeneUpdate.pl -ensVersion=91 ficAlb1.ensGene.ra) \ > ensGene.91.log 2>&1 # real 1m31.613s featureBits cavApe1 ensGene # 21842310 bases of 2716396567 (0.804%) in intersection ############################################################################# # augustus gene track (DONE - 2018-01-05 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/augustus cd /hive/data/genomes/ficAlb1/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=chicken -dbHost=hgwdev -workhorse=hgwdev ficAlb1) > do.log 2>&1 & # real 51m31.630s cat fb.ficAlb1.augustusGene.txt # 23463182 bases of 1102325870 (2.129%) in intersection ############################################################################## # lastz/chain/net swap human/hg38 (DONE - 2018-01-08 - Hiram) # original alignment cd /hive/data/genomes/hg38/bed/lastzFicAlb1.2018-01-08 cat fb.hg38.chainFicAlb1Link.txt # 147048785 bases of 3049335806 (4.822%) in intersection cat fb.hg38.chainSynFicAlb1Link.txt # 88694039 bases of 3049335806 (2.909%) in intersection cat fb.hg38.chainRBestFicAlb1Link.txt # 103439671 bases of 3049335806 (3.392%) in intersection # and for the swap: mkdir /hive/data/genomes/ficAlb1/bed/blastz.hg38.swap cd /hive/data/genomes/ficAlb1/bed/blastz.hg38.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/hg38/bed/lastzFicAlb1.2018-01-08/DEF \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -syntenicNet) > swap.log 2>&1 # real 14m32.727s cat fb.ficAlb1.chainHg38Link.txt # 118476285 bases of 1102325870 (10.748%) in intersection cat fb.ficAlb1.chainSynHg38Link.txt # 86852014 bases of 1102325870 (7.879%) in intersection time (doRecipBest.pl -load -workhorse=hgwdev -buildDir=`pwd` \ ficAlb1 hg38) > rbest.log 2>&1 # real 265m12.990s cat fb.ficAlb1.chainRBestHg38Link.txt # 103174487 bases of 1102325870 (9.360%) in intersection ############################################################################## # lastz/chain/net swap mouse/mm10 (DONE - 2018-01-09 - Hiram) # original alignment to mm10 cd /hive/data/genomes/mm10/bed/lastzFicAlb1.2018-01-09 cat fb.mm10.chainFicAlb1Link.txt # 98177848 bases of 2652783500 (3.701%) in intersection cat fb.mm10.chainRBestFicAlb1Link.txt # 76370866 bases of 2652783500 (2.879%) in intersection # and for the swap mkdir /hive/data/genomes/ficAlb1/bed/blastz.mm10.swap cd /hive/data/genomes/ficAlb1/bed/blastz.mm10.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/mm10/bed/lastzFicAlb1.2018-01-09/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 & # real 8m5.637s cat fb.ficAlb1.chainMm10Link.txt # 85384367 bases of 1102325870 (7.746%) in intersection time (doRecipBest.pl -load -workhorse=hgwdev ficAlb1 mm10 \ -buildDir=`pwd` -workhorse=hgwdev) > rbest.log 2>&1 & # real 209m22.159s cat fb.ficAlb1.chainRBestMm10Link.txt # 76183235 bases of 1102325870 (6.911%) in intersection ############################################################################## # Create kluster run files (DONE - 2018-01-08 - Hiram) # numerator is ficAlb1 gapless bases "real" as reported by: featureBits -noRandom -noHap ficAlb1 gap # 9190038 bases of 1034892040 (0.888%) 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 \( 1034892040 / 2861349177 \) \* 1024 # ( 1034892040 / 2861349177 ) * 1024 = 370.360059 # ==> use -repMatch=400 according to size scaled down from 1024 for human. # and rounded up to nearest 50 cd /hive/data/genomes/ficAlb1 blat ficAlb1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/ficAlb1.11.ooc \ -repMatch=400 # Wrote 14865 overused 11-mers to jkStuff/ficAlb1.11.ooc # check non-bridged gaps to see what the typical size is: hgsql -N \ -e 'select * from gap where bridge="no" order by size;' ficAlb1 \ | sort -k7,7nr | ave -col=7 stdin # all these gap sizes are 5000 gapToLift -verbose=2 -minGap=5000 ficAlb1 jkStuff/nonBridged.lft \ -bedFile=jkStuff/nonBridged.bed ######################################################################### # GENBANK AUTO UPDATE (DONE - 2018-01-08 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # #organism mrnaCnt estCnt refSeqCnt # Ficedula albicollis 0 0 8 # Ficedula hypoleuca 8 0 0 # edit etc/genbank.conf to add ficAlb1 just before geoFor1 # ficAlb1 (Ficedula albicollis - collard flycatcher) - taxId: 59894 ficAlb1.serverGenome = /hive/data/genomes/ficAlb1/ficAlb1.2bit ficAlb1.clusterGenome = /hive/data/genomes/ficAlb1/ficAlb1.2bit ficAlb1.ooc = /hive/data/genomes/ficAlb1/jkStuff/ficAlb1.11.ooc ficAlb1.lift = /hive/data/genomes/ficAlb1/jkStuff/nonBridged.lft ficAlb1.refseq.mrna.native.pslCDnaFilter = ${finished.refseq.mrna.native.pslCDnaFilter} ficAlb2.refseq.mrna.xeno.pslCDnaFilter = ${finished.refseq.mrna.xeno.pslCDnaFilter} ficAlb1.genbank.mrna.native.pslCDnaFilter = ${finished.genbank.mrna.native.pslCDnaFilter} ficAlb1.genbank.mrna.xeno.pslCDnaFilter = ${finished.genbank.mrna.xeno.pslCDnaFilter} ficAlb1.genbank.est.native.pslCDnaFilter = ${finished.genbank.est.native.pslCDnaFilter} # defaults yes: genbank.mrna.native.load and loadDesc # defaults yes: genbank.est.native.load # defaults yes: refseq.mrna.native.load, refseq.mrna.native.loadDesc # defaults yes: refseq.mrna.xeno.load, refseq.mrna.xeno.loadDesc # defaults no: genbank.mrna.xeno.load and loadDesc # defaults no: genbank.est.xeno.load, genbank.est.xeno.loadDesc # DO NOT NEED genbank.mrna.xeno except for human, mouse # there are no genbank.est natives: ficAlb1.genbank.est.native.load = no ficAlb1.downloadDir = ficAlb1 ficAlb1.perChromTables = no git commit -m \ 'adding ficAlb1 - Ficedula albicollis - collard flycatcher refs #9870' \ etc/genbank.conf src/lib/gbGenome.c git push git commit -m 'adding ficAlb1 Dipodomys ordii refs #20751' etc/genbank.conf git push # update /cluster/data/genbank/: make install-server make etc-update cd /cluster/data/genbank time ./bin/gbAlignStep -initial ficAlb1 # logFile: var/build/logs/2018.01.08-18:45:24.ficAlb1.initalign.log # real 184m19.477s tail -2 var/build/logs/2018.01.08-18:45:24.ficAlb1.initalign.log # hgwdev 2018.01.08-21:49:03 ficAlb1.initalign: Succeeded: ficAlb1 # hgwdev 2018.01.08-21:49:44 ficAlb1.initalign: finish # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.ficAlb1 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad ficAlb1 # logFile: var/dbload/hgwdev/logs/2018.01.08-22:05:29.ficAlb1.dbload.log # real 7m1.853s tail -1 var/dbload/hgwdev/logs/2018.01.08-22:05:29.ficAlb1.dbload.log # hgwdev 2018.01.08-22:12:31 ficAlb1.dbload: finish # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add ficAlb1 to: # etc/align.dbs etc/hgwdev.dbs git add etc/align.dbs etc/hgwdev.dbs git commit -m 'adding ficAlb1 to the update alignments refs #9870' etc/align.dbs etc/hgwdev.dbs git push make etc-update ############################################################################# # BLATSERVERS ENTRY (DONE - 2018-01-09 - 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 ("ficAlb1", "blat1b", "17892", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("ficAlb1", "blat1b", "17893", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################ ## reset default position to same protein location as hg38 ## found by blat (DONE - 2018-01-09 - Hiram) ssh hgwdev hgsql -e 'update dbDb set defaultPos="chr21:2266579-2296356" where name="ficAlb1";' hgcentraltest ######################################################################### # all.joiner update, downloads and in pushQ - (DONE - 2018-01-10 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output joinerCheck -database=ficAlb1 -tableCoverage all.joiner joinerCheck -database=ficAlb1 -times all.joiner joinerCheck -database=ficAlb1 -keys all.joiner cd /hive/data/genomes/ficAlb1 time (makeDownloads.pl -workhorse=hgwdev ficAlb1) > downloads.log 2>&1 # real 11m8.483s # now ready for pushQ entry mkdir /hive/data/genomes/ficAlb1/pushQ cd /hive/data/genomes/ficAlb1/pushQ time (makePushQSql.pl -redmineList ficAlb1) > ficAlb1.pushQ.sql 2> stderr.out # real 3m53.381s # check for errors in stderr.out, some are OK, e.g.: # WARNING: ficAlb1 does not have seq # WARNING: ficAlb1 does not have extFile # WARNING: ficAlb1 does not have estOrientInfo ## 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.ficAlb1.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.ficAlb1.table.list # 66 redmine.ficAlb1.table.list awk -F'.' '{ printf "hgsql -N -e \"show table status like '"'"'%s'"'"';\" %s\n", $2, $1 }' redmine.ficAlb1.table.list | while read L; do eval $L; done | wc -l # 66 # enter the path names to these files in the redmine issue to # make QA Ready: ls `pwd`/redmine* /hive/data/genomes/ficAlb1/pushQ/redmine.ficAlb1.file.list /hive/data/genomes/ficAlb1/pushQ/redmine.ficAlb1.releaseLog.txt /hive/data/genomes/ficAlb1/pushQ/redmine.ficAlb1.table.list ######################################################################### # tandemDups (DONE - 2018-02-27 - 03-14 - Hiram) mkdir /hive/data/genomes/ficAlb1/bed/tandemDups cd /hive/data/genomes/ficAlb1/bed/tandemDups time (~/kent/src/hg/utils/automation/doTandemDup.pl ficAlb1) > do.log 2>&1 # had immense trouble in the pairedEnds step, went on for weeks # made gigantic files: # -rw-rw-r-- 1 27432151329 Mar 9 17:00 chr1A.bed.gz # -rw-rw-r-- 1 30028043163 Mar 12 06:25 chr3.bed.gz # -rw-rw-r-- 1 41116614429 Mar 13 06:51 chr2.bed.gz # -rw-rw-r-- 1 36289190868 Mar 13 13:44 chr1.bed.gz # after recovery and getting that completed, continuing: time (~/kent/src/hg/utils/automation/doTandemDup.pl \ -continue=collapsePairedEnds ficAlb1) > collapsePairedEnds.log 2>&1 & #########################################################################