# for emacs: -*- mode: sh; -*- # This file describes browser build for the oreNil3 # Nile tilapia -- Oreochromis niloticus # chrMT listed in assembly == NC_013663.1 == GU238433.1 ############################################################################# # fetch sequence from new style download directory (DONE - 2017-07-27 - Hiram) mkdir -p /hive/data/genomes/oreNil3/refseq cd /hive/data/genomes/oreNil3/refseq time rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_other/Oreochromis_niloticus/all_assembly_versions/GCF_001858045.1_ASM185804v2/ ./ # real 1m49.631s # sent 2790 bytes received 1601058053 bytes 14621560.21 bytes/sec # total size is 1600852394 speedup is 1.00 # measure what we have here: faSize GCF_001858045.1_ASM185804v2_genomic.fna.gz # 1009856516 bases (4230001 N's 1005626515 real 714466436 upper # 291160079 lower) in 2567 sequences in 1 files # Total size: mean 393399.5 sd 3657283.6 min 5393 (NW_017615892.1) # max 62059223 (NC_031972.1) median 21833 # %28.83 masked total, %28.95 masked real faCount GCF_001858045.1_ASM185804v2_genomic.fna.gz | tail # seq len A C G T N cpg # total 1009856516 297968384 204671628 204958373 298028130 4230001 17363938 ############################################################################# # fixup to UCSC naming scheme (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/ucsc cd /hive/data/genomes/oreNil3/ucsc # verify no duplicate sequences: time faToTwoBit ../refseq/*2_genomic.fna.gz refseq.2bit # real 0m29.769s time twoBitDup refseq.2bit # should be silent output, otherwise the duplicates need to be removed # real 0m7.797s time ~/kent/src/hg/utils/automation/ucscCompositeAgp.pl \ ../refseq/*2_genomic.fna.gz ../refseq/*_assembly_structure/Primary_Assembly # NC_031965.1 chrLG1 # NC_031966.1 chrLG2 # NC_031967.1 chrLG3a # NC_031968.1 chrLG3b # NC_031969.1 chrLG4 # NC_031970.1 chrLG5 # NC_031971.1 chrLG6 # NC_031972.1 chrLG7 # NC_031973.1 chrLG8 # NC_031974.1 chrLG9 # NC_031975.1 chrLG10 # NC_031976.1 chrLG11 # NC_031977.1 chrLG12 # NC_031978.1 chrLG13 # NC_031979.1 chrLG14 # NC_031980.1 chrLG15 # NC_031981.1 chrLG17 # NC_031982.1 chrLG18 # NC_031983.1 chrLG19 # NC_031984.1 chrLG20 # NC_031985.1 chrLG22 # NC_031986.1 chrLG23 # NC_031987.1 chrLG16 # real 4m57.131s time ~/kent/src/hg/utils/automation/unplacedWithChroms.pl \ ../refseq/*_assembly_structure/Primary_Assembly # processed 2543 sequences into chrUn.fa.gz # real 0m47.519s # there are no unlocalized contigs # time ~/kent/src/hg/utils/automation/unlocalizedWithChroms.pl \ # ../refseq/*_assembly_structure/Primary_Assembly # Y # processed 9 sequences into chr*_random.gz 1 files # real 0m0.501s # bash syntax here G*2_assembly_report.txt mitoAcc=`grep MT ../refseq/G*2_assembly_report.txt | cut -f7` printf "# mitoAcc %s\n" "$mitoAcc" # mitoAcc NC_013663.1 zcat \ ../refseq/*_assembly_structure/non-nuclear/assem*/AGP/chrMT.comp.agp.gz \ | grep -v "^#" | sed -e "s/^$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 AGP match: time faToTwoBit chr*.fa.gz test.2bit # real 0m31.229s cat *.agp | checkAgpAndFa stdin test.2bit 2>&1 | tail # All AGP and FASTA entries agree - both files are valid # verify nothing lost compared to genbank: time twoBitToFa test.2bit stdout | faSize stdin # 1009856516 bases (4230001 N's 1005626515 real 767408040 upper # 238218475 lower) in 2567 sequences in 1 files # Total size: mean 393399.5 sd 3657283.6 min 5393 (chrUn_NW_017615892v1) # max 62059223 (chrLG7) median 21833 # %23.59 masked total, %23.69 masked real # real 0m20.750s # same totals as above: # 1009856516 bases (4230001 N's 1005626515 real 714466436 upper # 291160079 lower) in 2567 sequences in 1 files # Total size: mean 393399.5 sd 3657283.6 min 5393 (NW_017615892.1) # max 62059223 (NC_031972.1) median 21833 # %28.83 masked total, %28.95 masked real ############################################################################# # Initial database build (DONE - 2017-07-27 - Hiram) # same photoReference.txt as was used in oreNil2: cd /hive/data/genomes/oreNil3 cat ../oreNil2/photoReference.txt photoCreditURL http://fishbase.org/photos/PicturesSummary.php?resultPage=15&StartRow=9&ID=2&what=species&TotRec=11 photoCreditName Robbie Cada, fishbase.org cp -p ../oreNil2/photoReference.txt . # establish the config.ra file: ~/kent/src/hg/utils/automation/prepConfig.pl oreNil3 vertebrate oreNil \ refseq/G*2_assembly_report.txt > oreNil3.config.ra # verify this looks OK: cat oreNil3.config.ra # config parameters for makeGenomeDb.pl: db oreNil3 clade vertebrate scientificName Oreochromis niloticus commonName Nile tilapia assemblyDate Nov. 2016 assemblyLabel University of Maryland assemblyShortLabel ASM185804v2 orderKey 14418 # mitochondrial sequence included in refseq release # mitoAcc NC_013663.1 mitoAcc none fastaFiles /hive/data/genomes/oreNil3/ucsc/*.fa.gz agpFiles /hive/data/genomes/oreNil3/ucsc/*.agp # qualFiles none dbDbSpeciesDir oreNil photoCreditURL http://fishbase.org/photos/PicturesSummary.php?resultPage=15&StartRow=9&ID=2&what=species&TotRec=11 photoCreditName Robbie Cada, fishbase.org ncbiGenomeId 197 ncbiAssemblyId 893611 ncbiAssemblyName ASM185804v2 ncbiBioProject 354796 ncbiBioSample SAMN05859795 genBankAccessionID GCF_001858045.1 taxId 8128 # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp oreNil3.config.ra) > agp.log 2>&1 # real 1m1.424s # verify it ran OK: # *** All done! (through the 'agp' step) # then finish it off: time (~/kent/src/hg/utils/automation/makeGenomeDb.pl -workhorse=hgwdev \ -dbHost=hgwdev -fileServer=hgwdev -continue=db \ oreNil3.config.ra ) > db.log 2>&1 # real 8m19.946s # check in the trackDb files created and add to trackDb/makefile # temporary symlink until after masking ln -s `pwd`/oreNil3.unmasked.2bit /gbdb/oreNil3/oreNil3.2bit ############################################################################# # cytoBandIdeo - (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/cytoBand cd /hive/data/genomes/oreNil3/bed/cytoBand makeCytoBandIdeo.csh oreNil3 ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/cpgIslandsUnmasked cd /hive/data/genomes/oreNil3/bed/cpgIslandsUnmasked # run stepwise so the loading can be done in a different table time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/oreNil3/oreNil3.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku oreNil3) > do.log 2>&1 # real 4m24.538s cat fb.oreNil3.cpgIslandExtUnmasked.txt # 18179828 bases of 1009856516 (1.800%) in intersection ############################################################################# # running repeat masker (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/repeatMasker cd /hive/data/genomes/oreNil3/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku oreNil3) > do.log 2>&1 & # real 36m33.307s cat faSize.rmsk.txt # 1009856516 bases (4230001 N's 1005626515 real 951212348 upper # 54414167 lower) in 2567 sequences in 1 files # Total size: mean 393399.5 sd 3657283.6 min 5393 (chrUn_NW_017615892v1) # max 62059223 (chrLG7) median 21833 # %5.39 masked total, %5.41 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 oreNil3 rmsk # 54414168 bases of 1009856516 (5.388%) in intersection # real 0m5.322s # 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 # faster way to get the same result: time hgsql -N -e 'select genoName,genoStart,genoEnd from rmsk;' oreNil3 \ | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" # total 54414168.000000 # real 0m4.606s ########################################################################## # running simple repeat (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/simpleRepeat cd /hive/data/genomes/oreNil3/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ -trf409 5 oreNil3) > do.log 2>&1 & # real 9m52.732s cat fb.simpleRepeat # 31944076 bases of 1009856516 (3.163%) in intersection # using the Window Masker result: cd /hive/data/genomes/oreNil3] twoBitMask bed/windowMasker/oreNil3.cleanWMSdust.2bit \ -add bed/simpleRepeat/trfMask.bed oreNil3.2bit # you can safely ignore the warning about fields >= 13 # if using rmsk: # cd /hive/data/genomes/oreNil3 # twoBitMask oreNil3.rmsk.2bit \ # -add bed/simpleRepeat/trfMask.bed oreNil3.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa oreNil3.2bit stdout | faSize stdin > faSize.oreNil3.2bit.txt cat faSize.oreNil3.2bit.txt # 1009856516 bases (4230001 N's 1005626515 real 692122922 upper # 313503593 lower) in 2567 sequences in 1 files # Total size: mean 393399.5 sd 3657283.6 # min 5393 (chrUn_NW_017615892v1) max 62059223 (chrLG7) median 21833 # %31.04 masked total, %31.17 masked real rm /gbdb/oreNil3/oreNil3.2bit ln -s `pwd`/oreNil3.2bit /gbdb/oreNil3/oreNil3.2bit ############################################################################# # CREATE MICROSAT TRACK (DONE - 2017-07-27 - Hiram) ssh hgwdev mkdir /cluster/data/oreNil3/bed/microsat cd /cluster/data/oreNil3/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 oreNil3 microsat microsat.bed # Read 31297 elements of size 4 from microsat.bed ############################################################################# # ucscToINSDC table/track (DONE - 2017-07-29 - Hiram) # the sequence here is working for a 'refseq' assembly with a chrM # situation may be specific depending upon what is available in the assembly mkdir /hive/data/genomes/oreNil3/bed/ucscToINSDC cd /hive/data/genomes/oreNil3/bed/ucscToINSDC # find accession for chrM grep chrM ../../oreNil3.agp # chrM 1 16627 1 O NC_013663.1 1 16627 + # use that accession here: ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../refseq/GCF_*structure/Primary_Assembly NC_013663.1 # since this is a refseq assembly, this is actually ucscToRefSeq.txt sort -k2 ucscToINSDC.txt > ucscToRefSeq.txt rm -f ucscToINSDC.txt awk '{printf "%s\t%s\n", $2, $1}' ucscToRefSeq.txt | sort > refSeqToUcsc.txt # there is no INSDC name for chrM/NC_013663.1 sequence in the # assembly_report, use what was looked up earlier in this sed correction: grep -v "^#" ../../refseq/GCF*_assembly_report.txt | cut -f5,7 \ | sed -e 's/na\b/GU238433.1/;' | awk '{printf "%s\t%s\n", $2, $1}' \ | sort > refseq.insdc.txt # the sed \b means to match word awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > name.coordinate.tab # the tr commands avoid the problem of trying to use the -t argument # to the join command which doesn't accept -t'\t' but instead has # to use the unseen/can not copy command ctrl-v i 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 # should be same line counts throughout: wc -l * # 2567 name.coordinate.tab # 2567 refSeqToUcsc.txt # 2567 refseq.insdc.txt # 2567 ucscToINSDC.bed # 2567 ucscToRefSeq.bed # 2567 ucscToRefSeq.txt # verify chrM is correct: grep chrM *.bed # ucscToINSDC.bed:chrM 0 16627 GU238433.1 # ucscToRefSeq.bed:chrM 0 16627 NC_013663.1 cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1 # 20 # use the 20 in this sed sed -e "s/21/20/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab oreNil3 ucscToINSDC stdin ucscToINSDC.bed cut -f1 ucscToRefSeq.bed | awk '{print length($0)}' | sort -n | tail -1 sed -e "s/21/20/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab oreNil3 ucscToRefSeq stdin ucscToRefSeq.bed checkTableCoords oreNil3 # should cover %100 entirely: featureBits -countGaps oreNil3 ucscToINSDC # 1009856516 bases of 1009856516 (100.000%) in intersection featureBits -countGaps oreNil3 ucscToRefSeq # 1009856516 bases of 1009856516 (100.000%) in intersection ######################################################################### # add chromAlias table (DONE - 2017-07-29 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/chromAlias cd /hive/data/genomes/oreNil3/bed/chromAlias hgsql -N -e 'select chrom,name,"refseq" from ucscToRefSeq;' oreNil3 \ > ucsc.refseq.tab hgsql -N -e 'select chrom,name,"genbank" from ucscToINSDC;' oreNil3 \ > ucsc.genbank.tab # verify chrM is correct: grep chrM *.tab # ucsc.genbank.tab:chrM GU238433.1 genbank # ucsc.refseq.tab:chrM NC_013663.1 refseq awk '{printf "%s\t%s\t%s\n", $2,$1,$3}' ucsc.genbank.tab ucsc.refseq.tab \ | sort > oreNil3.chromAlias.tab hgLoadSqlTab oreNil3 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ oreNil3.chromAlias.tab ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2017-07-29 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/oreNil/oreNil3 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" oreNil3 \ | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c # 2566 MKQE.1 # 1 NC_.1 # implies a search rule of: '[MN][KC][EQ_]+[0-9]+(\.[0-9]+)?' # verify this rule will find them all or eliminate them all: hgsql -N -e "select frag from gold;" oreNil3 | wc -l # 2567 hgsql -N -e "select frag from gold;" oreNil3 \ | egrep -e '[MN][KC][EQ_]+[0-9]+(\.[0-9]+)?' | wc -l # 2567 hgsql -N -e "select frag from gold;" oreNil3 \ | egrep -v -e '[MN][KC][EQ_]+[0-9]+(\.[0-9]+)?' | wc -l # 0 # hence, add to trackDb/rhesus/oreNil3/trackDb.ra searchTable gold shortCircuit 1 termRegex [ACEFJN_]+[0-9]+(\.[0-9]+)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 git commit -m "Added gold table search rules refs #19879" trackDb.ra # verify searches work in the position box ########################################################################## ## WINDOWMASKER (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/windowMasker cd /hive/data/genomes/oreNil3/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev oreNil3) > do.log 2>&1 # real 43m54.433s # Masking statistics cat faSize.oreNil3.cleanWMSdust.txt # 1009856516 bases (4230001 N's 1005626515 real 692350940 upper # 313275575 lower) in 2567 sequences in 1 files # Total size: mean 393399.5 sd 3657283.6 min 5393 (chrUn_NW_017615892v1) # max 62059223 (chrLG7) median 21833 # %31.02 masked total, %31.15 masked real cat fb.oreNil3.rmsk.windowmaskerSdust.txt # 43101047 bases of 1009856516 (4.268%) in intersection ########################################################################## # cpgIslands - (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/cpgIslands cd /hive/data/genomes/oreNil3/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku oreNil3) > do.log 2>&1 # real 2m41.885s cat fb.oreNil3.cpgIslandExt.txt # 8831483 bases of 1009856516 (0.875%) in intersection ############################################################################## # ncbiRefSeq gene track (TBD - 2016-05-05 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/ncbiRefSeq cd /hive/data/genomes/oreNil3/bed/ncbiRefSeq # working on this script, running step by step: time (/cluster/home/hiram/kent/src/hg/utils/automation/doNcbiRefSeq.pl \ -stop=download -buildDir=`pwd` -bigClusterHub=ku \ -fileServer=hgwdev -workhorse=hgwdev -smallClusterHub=ku -dbHost=hgwdev \ refseq vertebrate_mammalian Pan_paniscus \ GCF_000258655.2_panpan1.1 oreNil3) > download.log 2>&1 # real 12m36.320s time (/cluster/home/hiram/kent/src/hg/utils/automation/doNcbiRefSeq.pl \ -continue=process -stop=process -buildDir=`pwd` -bigClusterHub=ku \ -fileServer=hgwdev -workhorse=hgwdev -smallClusterHub=ku -dbHost=hgwdev \ refseq vertebrate_mammalian Pan_paniscus \ GCF_000258655.2_panpan1.1 oreNil3) > process.log 2>&1 # real 4m22.621s time (/cluster/home/hiram/kent/src/hg/utils/automation/doNcbiRefSeq.pl \ -continue=load -stop=load -buildDir=`pwd` -bigClusterHub=ku \ -fileServer=hgwdev -workhorse=hgwdev -smallClusterHub=ku -dbHost=hgwdev \ refseq vertebrate_mammalian Pan_paniscus \ GCF_000258655.2_panpan1.1 oreNil3) > load.log 2>&1 # real 0m21.690s cat fb.ncbiRefSeq.oreNil3.txt # 74646536 bases of 2725937399 (2.738%) in intersection ############################################################################## # genscan - (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/genscan cd /hive/data/genomes/oreNil3/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku oreNil3) > do.log 2>&1 # real 31m40.539s # four broken, completed with window size 2000000, continuing: time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -continue=makeBed -bigClusterHub=ku oreNil3) > makeBed.log 2>&1 # real 1m12.510s cat fb.oreNil3.genscan.txt # 53054923 bases of 1009856516 (5.254%) in intersection cat fb.oreNil3.genscanSubopt.txt # 23284781 bases of 1009856516 (2.306%) in intersection ############################################################################# # augustus gene track (DONE - 2017-07-27 - Hiram) mkdir /hive/data/genomes/oreNil3/bed/augustus cd /hive/data/genomes/oreNil3/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=zebrafish -dbHost=hgwdev \ -workhorse=hgwdev oreNil3) > do.log 2>&1 # real 59m19.281s cat fb.oreNil3.augustusGene.txt # 59505339 bases of 1009856516 (5.892%) in intersection ######################################################################### # Create kluster run files (DONE - 2017-07-27 - Hiram) cd /hive/data/genomes/oreNil3 # numerator is oreNil3 gapless bases "real" as reported by: featureBits -noRandom -noHap oreNil3 gap # 0 bases of 868607890 (0.000%) 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 \( 868607890 / 2861349177 \) \* 1024 # ( 868607890 / 2861349177 ) * 1024 = 310.851429 # ==> use -repMatch=300 according to size scaled down from 1024 for human. # and rounded down to nearest 100 cd /hive/data/genomes/oreNil3 time blat oreNil3.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/oreNil3.11.ooc \ -repMatch=300 # Wrote 37324 overused 11-mers to jkStuff/oreNil3.11.ooc # real 0m17.355s # there are zero gaps in this assembly # check non-bridged gaps to see what the typical size is: # hgsql -N \ # -e 'select * from gap where bridge="no" order by size;' oreNil3 \ # | sort -k7,7nr # minimum size is 1000 # decide on a minimum gap for this break, use either 100 or 5000 will # generate 13387 liftOver rows, but if use 6000, only got 11703 rows. # so use 100 here to get more liftOver row. # gapToLift -verbose=2 -minGap=1000 oreNil3 jkStuff/oreNil3.nonBridged.lft \ # -bedFile=jkStuff/oreNil3.nonBridged.bed ######################################################################### # LIFTOVER TO oreNil2 (DONE - 2017-07-27 - Hiram ) mkdir /hive/data/genomes/oreNil3/bed/blat.oreNil2.2017-07-27 cd /hive/data/genomes/oreNil3/bed/blat.oreNil2.2017-07-27 # -debug run to create run dir, preview scripts... doSameSpeciesLiftOver.pl -debug \ -ooc /hive/data/genomes/oreNil3/jkStuff/oreNil3.11.ooc \ oreNil3 oreNil2 # Real run: time (doSameSpeciesLiftOver.pl -verbose=2 \ -ooc /hive/data/genomes/oreNil3/jkStuff/oreNil3.11.ooc \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ oreNil3 oreNil2) > do.log 2>&1 & # real 220m19.231s # verify browser can run a view in other genomes to oreNil3 from oreNil2 ######################################################################## # lastz/chain/net swap from hg38 (DONE - 2017-08-02 - Hiram) # alignment to hg38 cd /hive/data/genomes/hg38/bed/lastzOreNil3.2017-07-31 cat fb.hg38.chainOreNil3Link.txt # 59963651 bases of 3049335806 (1.966%) in intersection # and for the swap: mkdir /hive/data/genomes/oreNil3/bed/blastz.hg38.swap cd /hive/data/genomes/oreNil3/bed/blastz.hg38.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/hg38/bed/lastzOreNil3.2017-07-31/DEF \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -syntenicNet) > swap.log 2>&1 # real 8m59.810s cat fb.oreNil3.chainHg38Link.txt # 56298016 bases of 1009856516 (5.575%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` oreNil3 hg38) \ > rbest.log 2>&1 # real 263m17.904s ######################################################################### # lastz/chain/net swap from mm10 (DONE = 2017-08-02 - Hiram) # alignment to mm10 cd /hive/data/genomes/mm10/bed/lastzOreNil3.2017-07-31 cat fb.mm10.chainOreNil3Link.txt # 54152663 bases of 2652783500 (2.041%) in intersection # and for the swap mkdir /hive/data/genomes/oreNil3/bed/blastz.mm10.swap cd /hive/data/genomes/oreNil3/bed/blastz.mm10.swap time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/mm10/bed/lastzOreNil3.2017-07-31/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 & # real 8m5.590s cat fb.oreNil3.chainMm10Link.txt # 55291586 bases of 1009856516 (5.475%) in intersection time (doRecipBest.pl -workhorse=hgwdev oreNil3 mm10 \ -buildDir=`pwd` -workhorse=hgwdev) > rbest.log 2>&1 & # real 230m56.580s ######################################################################### # GENBANK AUTO UPDATE (DONE - 2017-07-29 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # organism mrnaCnt estCnt refSeqCnt # Oreochromis niloticus 861 120996 182 # Oreochromis niloticus niloticus 8 0 0 # edit etc/genbank.conf to add oreNil3 just before oreNil2 # oreNil3 (Nile tilapia) oreNil3.serverGenome = /hive/data/genomes/oreNil3/oreNil3.2bit oreNil3.clusterGenome = /hive/data/genomes/oreNil3/oreNil3.2bit oreNil3.ooc = /hive/data/genomes/oreNil3/jkStuff/oreNil3.11.ooc oreNil3.lift = no oreNil3.refseq.mrna.native.pslCDnaFilter = ${finished.refseq.mrna.native.pslCDnaFilter} oreNil3.refseq.mrna.xeno.pslCDnaFilter = ${finished.refseq.mrna.xeno.pslCDnaFilter} oreNil3.genbank.mrna.native.pslCDnaFilter = ${finished.genbank.mrna.native.pslCDnaFilter} oreNil3.genbank.mrna.xeno.pslCDnaFilter = ${finished.genbank.mrna.xeno.pslCDnaFilter} oreNil3.genbank.est.native.pslCDnaFilter = ${finished.genbank.est.native.pslCDnaFilter} oreNil3.refseq.mrna.native.load = yes oreNil3.refseq.mrna.xeno.load = yes oreNil3.genbank.mrna.xeno.load = no oreNil3.genbank.est.native.load = yes oreNil3.downloadDir = oreNil3 oreNil3.perChromTables = no git commit -m "Added oreNil3; refs #19879" etc/genbank.conf git push # update /cluster/data/genbank/: make etc-update # Edit src/lib/gbGenome.c to add new species. Skipped screen # control this business with a screen since it takes a while cd /cluster/data/genbank time ./bin/gbAlignStep -initial oreNil3 # logFile: var/build/logs/2017.07.27-20:58:31.oreNil3.initalign.log # real 266m50.129s tail -2 var/build/logs/2017.07.27-20:58:31.oreNil3.initalign.log # hgwdev 2017.07.28-01:25:08 oreNil3.initalign: Succeeded: oreNil3 # hgwdev 2017.07.28-01:25:21 oreNil3.initalign: finish # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.oreNil3 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad oreNil3 # logFile: var/dbload/hgwdev/logs/2017.07.28-14:27:37.oreNil3.dbload.log # real 19m51.130s # tail -1 var/dbload/hgwdev/logs/2017.07.28-14:27:37.oreNil3.dbload.log # hgwdev 2017.07.28-14:47:28 oreNil3.dbload: finish # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add oreNil3 to: # etc/align.dbs etc/hgwdev.dbs git add etc/align.dbs git add etc/hgwdev.dbs git commit -m "Added oreNil3 - Nile tilapia refs #19879" etc/align.dbs etc/hgwdev.dbs git push make etc-update ############################################################################## # BLATSERVERS ENTRY (DONE - 2017-07-31 - 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 ("oreNil3", "blat1b", "17882", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("oreNil3", "blat1b", "17883", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################## # set default position to GHR gene displays (DONE - 2017-08-03 - Hiram) hgsql -e \ 'update dbDb set defaultPos="chrLG7:32798427-32827906" where name="oreNil3";' \ hgcentraltest ############################################################################## # all.joiner update, downloads and in pushQ - (DONE - 2017-08-03 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output # tableCoverage and times should be silent, no output for clean check joinerCheck -database=oreNil3 -tableCoverage all.joiner joinerCheck -database=oreNil3 -times all.joiner joinerCheck -database=oreNil3 -keys all.joiner git commit -m 'adding oreNil3 rules refs #19879' all.joiner git push cd /hive/data/genomes/oreNil3 time (makeDownloads.pl oreNil3) > downloads.log 2>&1 # real 7m31.828s # now ready for pushQ entry mkdir /hive/data/genomes/oreNil3/pushQ cd /hive/data/genomes/oreNil3/pushQ time (makePushQSql.pl -redmineList oreNil3) \ > oreNil3.pushQ.sql 2> stderr.out # real 3m39.676s # check for errors in stderr.out, some are OK, e.g.: # writing redmine listings to # redmine.oreNil3.file.list # redmine.oreNil3.table.list # redmine.oreNil3.releaseLog.txt # WARNING: oreNil3 does not have seq # WARNING: oreNil3 does not have extFile #########################################################################