2e3edf11e977a5c3a12aa2e1fed9cbfa25ffe928 hiram Mon Apr 26 15:02:21 2021 -0700 cleaning up crisprAll working data about 326 Tb no redmine diff --git src/hg/makeDb/doc/regenRn1/initialBuild.txt src/hg/makeDb/doc/regenRn1/initialBuild.txt index 1f7c4d1..a23496f 100644 --- src/hg/makeDb/doc/regenRn1/initialBuild.txt +++ src/hg/makeDb/doc/regenRn1/initialBuild.txt @@ -1,992 +1,999 @@ # for emacs: -*- mode: sh; -*- # This file describes browser build for the regenRn1 # Can use existing photograph (otherwise find one before starting here) ######################################################################### # Initial steps, use existing photograph (DONE - 2019-09-08 - Hiram) # To start this initialBuild.txt document, from a previous assembly document: mkdir ~/kent/src/hg/makeDb/doc/regenRn1 cd ~/kent/src/hg/makeDb/doc/regenRn1 sed -e 's/Rn0/Rn1/g; s/DONE/TBD/g;' \ ../regenRn0/initialBuild.txt > initialBuild.txt mkdir -p /hive/data/genomes/regenRn1/regeneron # sequences were obtained via email instructions from Regeneron # downloaded from a URl for their secure situation # -rw-r--r-- 1 643996250 Sep 8 08:26 DarkAgouti_chromosomes.tgz # their file is very odd. it is not a tgz, it requires two: # unzip # unzip # then it is a tar file, which makes a directory: -rw-r--r-- 1 2395238400 Sep 8 08:26 file.tar tar xvf file.tat drwxrwxr-x 2 4096 Sep 7 17:39 DA_annotated_chromosomes # examples: -rw-rw-r-- 1 115245581 Sep 3 12:06 chr1.fasta -rw-rw-r-- 1 41589339 Sep 3 13:00 chr12.fasta # Can use existing photograph cp -p ../regenRn0/photoReference.txt . cat photoReference.txt photoCreditURL http://www.genome.gov/dmd/img.cfm?node=Photos/Animals/Rat%20(brown%20norway)&id=79111 photoCreditName NHGRI press photos # check assembly size for later reference: faSize DA_annotated_chromosomes/*.fasta 2395213054 bases (112730866 N's 2282482188 real 2276778425 upper 5703763 lower) in 23 sequences in 23 files Total size: mean 104139698.0 sd 52890062.5 min 2246 (chrM) max 236134952 (chr2) median 102991095 # this assembly just has the 23 chromosome sequences ############################################################################# # establish config.ra file (DONE - Hiram - 2019-09-08) cd /hive/data/genomes/regenRn1 # copy the rn6 file and edit: cat regenRn9.config.ra # Config parameters for makeGenomeDb.pl: db regenRn1 clade mammal scientificName Rattus norvegicus commonName Rat assemblyDate May. 2019 assemblyLabel Regeneron rat assemblyShortLabel Regen Rn1 orderKey 18021 mitoAcc none fastaFiles /hive/data/genomes/regenRn1/ucsc/*.fa.gz agpFiles /hive/data/genomes/regenRn1/ucsc/fake.agp # qualFiles none dbDbSpeciesDir rat photoCreditURL http://www.genome.gov/10005141 photoCreditName NHGRI Press Photos ncbiGenomeId 73 ncbiAssemblyId 382928 ncbiAssemblyName RegenRn1 ncbiBioProject 10629 ncbiBioSample SAMN02808228 genBankAccessionID GCA_000001895.4 taxId 10116 # compare with previous version to see if it is sane: diff regenRn1.config.ra ../regenRn0/regenRn0.config.ra ############################################################################# # setup UCSC named files (DONE - 2019-09-08 - Hiram) # copy the *.fasta files from ../regeneron/DA_annotated_chromosomes/ # to here, renaming to .fa, then gzip mkdir /hive/data/genomes/regenRn1/ucsc cd /hive/data/genomes/regenRn1/regeneron/DA_annotated_chromosomes ls *.fasta | sed -e 's/.fasta//;' | while read C do cp -p $C.fasta ../../ucsc/$C.fa gzip ../../ucsc/$C.fa & done wait cd /hive/data/genomes/regenRn1/ucsc # construct a fakeAgp: faToTwoBit *.fa.gz test.2bit twoBitDup test.2bit twoBitToFa test.2bit stdout | hgFakeAgp -minContigGap=1 \ -minScaffoldGap=100 stdin fake.agp checkAgpAndFa fake.agp test.2bit 2>&1 | tail -4 # All AGP and FASTA entries agree - both files are valid twoBitToFa test.2bit stdout | faSize stdin # 2395213054 bases (112730866 N's 2282482188 real 2276778425 upper # 5703763 lower) in 23 sequences in 1 files # Total size: mean 104139698.0 sd 52890062.5 min 2246 (chrM) # max 236134952 (chr2) median 102991095 # %0.24 masked total, %0.25 masked real # no longer need the temporary 2bit file rm test.2bit ############################################################################# # Initial database build (DONE - 2019-09-08 - Hiram) # create the database time (makeGenomeDb.pl -stop=db regenRn1.config.ra -workhorse=hgwdev) \ > db.log 2>&1 # real 14m41.734s # finish it off time (makeGenomeDb.pl -continue=dbDb regenRn1.config.ra -workhorse=hgwdev) \ > dbDb.log 2>&1 # real 0m11.683s # check in the trackDb files created in TemporaryTrackDbCheckout/ # and add regenRn1 to trackDb/makefile # temporary symlink until masked sequence is available cd /hive/data/genomes/regenRn1 ln -s `pwd`/regenRn1.unmasked.2bit /gbdb/regenRn1/regenRn1.2bit ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2019-09-08 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/cpgIslandsUnmasked cd /hive/data/genomes/regenRn1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/regenRn1/regenRn1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku regenRn1) > do.log 2>&1 # real 3m16.335s cat fb.regenRn1.cpgIslandExtUnmasked.txt # 10642803 bases of 2282482188 (0.466%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2019-09-08 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/cytoBand cd /hive/data/genomes/regenRn1/bed/cytoBand makeCytoBandIdeo.csh regenRn1 ############################################################################# # run up idKeys files for chromAlias/ncbiRefSeq (TBD - 2019-09-08 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/idKeys cd /hive/data/genomes/regenRn1/bed/idKeys time (doIdKeys.pl \ -twoBit=/hive/data/genomes/regenRn1/regenRn1.unmasked.2bit \ -buildDir=`pwd` regenRn1) > do.log 2>&1 & # real 0m47.105s cat regenRn1.keySignature.txt # 7850e2d5dabb6134fdc9d7083f1a3a54 ############################################################################# # gapOverlap (DONE - 2019-09-08 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/gapOverlap cd /hive/data/genomes/regenRn1/bed/gapOverlap time (doGapOverlap.pl \ -twoBit=/hive/data/genomes/regenRn1/regenRn1.unmasked.2bit regenRn1 ) \ > do.log 2>&1 & # real 1m49.029s # there are 7 items found: wc -l bed.tab # 7 bed.tab cat fb.regenRn1.gapOverlap.txt # 5912 bases of 2762428432 (0.000%) in intersection ############################################################################# # tandemDups (DONE - 2019-09-08 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/tandemDups cd /hive/data/genomes/regenRn1/bed/tandemDups time (~/kent/src/hg/utils/automation/doTandemDup.pl \ -twoBit=/hive/data/genomes/regenRn1/regenRn1.unmasked.2bit regenRn1) \ > do.log 2>&1 & # real 371m3.411s cat fb.regenRn1.tandemDups.txt # 75673026 bases of 2395213054 (3.159%) in intersection bigBedInfo regenRn1.tandemDups.bb | sed -e 's/^/# /;' # version: 4 # fieldCount: 13 # hasHeaderExtension: yes # isCompressed: yes # isSwapped: 0 # extraIndexCount: 0 # itemCount: 899,735 # primaryDataSize: 23,487,472 # primaryIndexSize: 63,376 # zoomLevels: 9 # chromCount: 22 # basesCovered: 1,254,230,011 # meanDepth (of bases covered): 6.233185 # minDepth: 1.000000 # maxDepth: 253.000000 # std of depth: 11.790108 ######################################################################### # ucscToINSDC and ucscToRefSeq table/track (TBD - 2019-09-08 - Hiram) # construct idKeys for the refseq sequence mkdir /hive/data/genomes/regenRn1/refseq/idKeys cd /hive/data/genomes/regenRn1/refseq/idKeys faToTwoBit ../GCF_000002315.5_GRCg6a_genomic.fna.gz regenRn1.refSeq.2bit time (doIdKeys.pl -buildDir=`pwd` \ -twoBit=`pwd`/regenRn1.refSeq.2bit refseqRegenRn1) > do.log 2>&1 & # real 0m48.786s cat refseqRegenRn1.keySignature.txt # 7850e2d5dabb6134fdc9d7083f1a3a54 # and the genbank sequence needs keys too: mkdir /hive/data/genomes/regenRn1/refseq/idKeysGenbank cd /hive/data/genomes/regenRn1/refseq/idKeysGenbank faToTwoBit /hive/data/outside/ncbi/genomes/genbank/vertebrate_other/Gallus_gallus/all_assembly_versions/GCA_000002315.5_GRCg6a/GCA_000002315.5_GRCg6a_genomic.fna.gz regenRn1.genbank.2bit time (doIdKeys.pl -buildDir=`pwd` \ -twoBit=`pwd`/regenRn1.genbank.2bit genbankRegenRn1) > do.log 2>&1 & cat genbankRegenRn1.keySignature.txt # a20fdad3318d371fcb34fcc66bab3752 mkdir /hive/data/genomes/regenRn1/bed/chromAlias join -t$'\t' ../idKeys/regenRn1.idKeys.txt \ ../../refseq/idKeysGenbank/genbankRegenRn1.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/regenRn1.idKeys.txt \ ../../refseq/idKeys/refseqRegenRn1.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: wc -l * ../../chrom.sizes # 463 ucscToINSDC.bed # 464 ucscToRefSeq.bed # 464 ../../chrom.sizes # need to find the accession for the INSDC equivalent to chrM: egrep chrM * # ucscToRefSeq.bed:chrM 0 16775 NC_001323.1 # lookup that accession at NCBI Entrez: X52392.1 # and add to ucscToINSDC.bed: printf "chrM\t0\t16775\tX52392.1\n" >> ucscToINSDC.bed # verify: grep chrM * # ucsc.genbank.tab:chrM X52392.1 # ucsc.refseq.tab:chrM NC_001323.1 # ucscToINSDC.bed:chrM 0 16775 X52392.1 # ucscToRefSeq.bed:chrM 0 16775 NC_001323.1 export chrSize=`cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1` echo $chrSize # 27 # use the $chrSize in this sed sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab regenRn1 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 $chrSize # 27 sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | sed -e 's/INSDC/RefSeq/g;' \ | hgLoadSqlTab regenRn1 ucscToRefSeq stdin ucscToRefSeq.bed # should be quiet for all OK checkTableCoords regenRn1 # should cover %100 entirely: featureBits -countGaps regenRn1 ucscToINSDC # 1065365425 bases of 1065365425 (100.000%) in intersection featureBits -countGaps regenRn1 ucscToRefSeq # 1065365425 bases of 1065365425 (100.000%) in intersection ######################################################################### # add chromAlias table (TBD - 2019-09-08 - Hiram) # did not run this up, there aren't any names to alias mkdir /hive/data/genomes/regenRn1/bed/chromAlias cd /hive/data/genomes/regenRn1/bed/chromAlias ln -s ../../ucsc/ucsc.regeneron.name.translate regen.tab ~/kent/src/hg/utils/automation/chromAlias.pl regen.tab \ > regenRn1.chromAlias.tab join regen.tab <(sort -k1,1 ../../chrom.sizes) \ | awk '{printf "%s\t0\t%d\t%s\n", $1, $3, $2}' > regen.alias.bed wc -l * # 6846 regen.alias.bed # 6846 regen.tab # 6846 regenRn1.chromAlias.tab hgLoadSqlTab regenRn1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ regenRn1.chromAlias.tab # display the Regeneron AGP file as a track: ln -s ../../ucsc/ucsc.agp ./translatedRegen.agp grep -v "^#" translatedRegen.agp | grep -w scaffold \ | awk '{printf "%s\t%d\t%d\tgap\t0\t.\n", $1, $2-1, $3}' > regen.gaps.bed grep -v "^#" translatedRegen.agp | grep -v -w scaffold \ | awk '{printf "%s\t%d\t%d\t%s\t0\t%s\n", $1, $2-1, $3, $6, $9}' \ > regen.contigs.bed sort -k1,1 -k2,2n regen.gaps.bed regen.contigs.bed > regen.agp.bed hgLoadBed regenRn1 regenAGP regen.agp.bed # Read 17962 elements of size 6 from regen.agp.bed cut -f1-4 regen.gaps.bed | hgLoadBed regenRn1 regenAGPGap stdin # Read 5558 elements of size 4 from stdin ######################################################################### # fixup search rule for assembly track/gold table (TBD - 2019-09-08 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/rat/regenRn1 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" regenRn1 \ | sed -e 's/[0-9_FR]\+//;' | sort | uniq -c 6725 pi 67 pisub 47725 ss # implies a rule: '[ps][is](sub)?[0-9]+(_[0-9FR]+)?' # verify this rule will find them all and eliminate them all: hgsql -N -e "select frag from gold;" regenRn1 | wc -l # 54517 hgsql -N -e "select frag from gold;" regenRn1 \ | egrep -e '[ps][is](sub)?[0-9]+(_[0-9FR]+)?' | wc -l # 54517 hgsql -N -e "select frag from gold;" regenRn1 \ | egrep -v -e '[ps][is](sub)?[0-9]+(_[0-9FR]+)?' | wc -l # 0 # hence, add to trackDb/chicken/regenRn1/trackDb.ra searchTable gold shortCircuit 1 termRegex [ps][is](sub)?[0-9]+(_[0-9FR]+)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 # verify searches work in the position box ########################################################################## # running repeat masker (DONE - 2019-09-08 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/repeatMasker cd /hive/data/genomes/regenRn1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku regenRn1) > do.log 2>&1 & # real 343m22.076s cat faSize.rmsk.txt # 2395213054 bases (112730866 N's 2282482188 real 1288908568 upper # 993573620 lower) in 23 sequences in 1 files # Total size: mean 104139698.0 sd 52890062.5 min 2246 (chrM) # max 236134952 (chr2) median 102991095 # %41.48 masked total, %43.53 masked real egrep -i "versi|relea" do.log # RepeatMasker version development-$Id: RepeatMasker,v 1.332 2017/04/17 19:01:11 rhubley Exp $ # February 01 2017 (open-4-0-8) 1.332 version of RepeatMasker # grep RELEASE /hive/data/staging/data/RepeatMasker/Libraries/RepeatMaskerLib.embl # CC Dfam_Consensus RELEASE 20181026; * time featureBits -countGaps regenRn1 rmsk # 993577605 bases of 2395213054 (41.482%) in intersection # real 0m31.877s # 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;' regenRn1 \ | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" # total 993577605.000000 # real 0m19.017s ########################################################################## # running simple repeat (DONE - 2019-09-08 - Hiram) # The '-trf409 4' is a bit smaller than human which is 6 mkdir /hive/data/genomes/regenRn1/bed/simpleRepeat cd /hive/data/genomes/regenRn1/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ -trf409=4 regenRn1) > do.log 2>&1 & # real 14m43.287s cat fb.simpleRepeat # 77526109 bases of 2282482188 (3.397%) in intersection cd /hive/data/genomes/regenRn1 # when using the Window Masker result: cd /hive/data/genomes/regenRn1 # twoBitMask bed/windowMasker/regenRn1.cleanWMSdust.2bit \ # -add bed/simpleRepeat/trfMask.bed regenRn1.2bit # you can safely ignore the warning about fields >= 13 # or using the rmsk result after it is done: twoBitMask regenRn1.rmsk.2bit \ -add bed/simpleRepeat/trfMask.bed regenRn1.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa regenRn1.2bit stdout | faSize stdin > faSize.regenRn1.2bit.txt cat faSize.regenRn1.2bit.txt # 2395213054 bases (112730866 N's 2282482188 real 1285490790 upper # 996991398 lower) in 23 sequences in 1 files # Total size: mean 104139698.0 sd 52890062.5 min 2246 (chrM) # max 236134952 (chr2) median 102991095 # %41.62 masked total, %43.68 masked real rm /gbdb/regenRn1/regenRn1.2bit ln -s `pwd`/regenRn1.2bit /gbdb/regenRn1/regenRn1.2bit ######################################################################### # CREATE MICROSAT TRACK (DONE - 2019-09-09 - Hiram) ssh hgwdev mkdir /cluster/data/regenRn1/bed/microsat cd /cluster/data/regenRn1/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 regenRn1 microsat microsat.bed # Read 208104 elements of size 4 from microsat.bed featureBits regenRn1 microsat # 9627095 bases of 2282482188 (0.422%) in intersection ########################################################################## ## WINDOWMASKER (DONE - 2019-09-09 - Hiram) # wait for RepeatMasker to finish before this, since this is going # to compare itself with the rmsk result mkdir /hive/data/genomes/regenRn1/bed/windowMasker cd /hive/data/genomes/regenRn1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev regenRn1) > do.log 2>&1 # real 26m58.753s # Masking statistics cat faSize.regenRn1.cleanWMSdust.txt # 2762428432 bases (227617579 N's 2534810853 real 1629368138 upper # 905442715 lower) in 6846 sequences in 1 files # Total size: mean 403509.8 sd 6476189.8 min 101 (pi012238F) # max 248247500 (ss100001) median 10668 # %32.78 masked total, %35.72 masked real cat fb.regenRn1.rmsk.windowmaskerSdust.txt # 696560115 bases of 2762428432 (25.215%) in intersection ########################################################################## # cpgIslands - (DONE - 2019-09-09 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/cpgIslands cd /hive/data/genomes/regenRn1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku regenRn1) > do.log 2>&1 # real 3m12.993s cat fb.regenRn1.cpgIslandExt.txt # 9240105 bases of 2282482188 (0.405%) in intersection ############################################################################## # genscan - (DONE - 2019-09-09 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/genscan cd /hive/data/genomes/regenRn1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku regenRn1) > do.log 2>&1 # real 113m30.148s cat fb.regenRn1.genscan.txt # 48448954 bases of 2282482188 (2.123%) in intersection cat fb.regenRn1.genscanSubopt.txt # 51171443 bases of 2282482188 (2.242%) in intersection ######################################################################### # Create kluster run files (DONE - 2019-09-11 - Hiram) # numerator is regenRn1 gapless bases "real" as reported by: featureBits -noRandom -noHap regenRn1 gap # 112730866 bases of 2282482188 (4.939%) 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 \( 2282482188 / 2861349177 \) \* 1024 # ( 2282482188 / 2861349177 ) * 1024 = 816.839056 # ==> use -repMatch=800 according to size scaled down from 1024 for human. # and rounded down to nearest 50 cd /hive/data/genomes/regenRn1 blat regenRn1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/regenRn1.11.ooc \ -repMatch=800 # Wrote 30001 overused 11-mers to jkStuff/regenRn1.11.ooc # regenRn0 was repMatch 900 and: # Wrote 29409 overused 11-mers to jkStuff/regenRn1.11.ooc # rn6 was repMatch 1000 and: # Wrote 27021 overused 11-mers to jkStuff/rn6.11.ooc # rn5 at repMatch 950 and: # Wrote 34513 overused 11-mers to jkStuff/rn5.11.ooc # check non-bridged gaps to see what the typical size is: hgsql -N \ -e 'select * from gap where bridge="no" order by size;' regenRn1 \ | sort -k7,7nr | ave -col=7 stdin # min 100.000000 # max 136961.000000 gapToLift -verbose=2 -minGap=100 regenRn1 jkStuff/nonBridged.lift \ -bedFile=jkStuff/nonBridged.bed ### kent source version 387 ### # chrom count: 23 # found 23482 gaps of size >= 100 # bed output requested to jkStuff/nonBridged.bed # no gaps on chrom: chr11, size: 83994014 # no gaps on chrom: chr2, size: 236134952 # no gaps on chrom: chr8, size: 119798696 # no gaps on chrom: chrM, size: 2246 wc -l jkStuff/nonBri* # 23505 jkStuff/nonBridged.bed # 23505 jkStuff/nonBridged.lift ######################################################################## # lastz/chain/net swap human/hg38 (DONE - 2019-09-12 - Hiram) # original alignment cd /hive/data/genomes/hg38/bed/lastzRegenRn1.2019-09-12 cat fb.hg38.chainRegenRn1Link.txt # 863041482 bases of 3095998939 (27.876%) in intersection cat fb.hg38.chainSynRegenRn1Link.txt # 809332351 bases of 3095998939 (26.141%) in intersection cat fb.hg38.chainRBest.RegenRn1.txt # 792532873 bases of 3095998939 (25.599%) in intersection # and for the swap: mkdir /hive/data/genomes/regenRn1/bed/blastz.hg38.swap cd /hive/data/genomes/regenRn1/bed/blastz.hg38.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/hg38/bed/lastzRegenRn1.2019-09-12/DEF \ -swap -chainMinScore=3000 -chainLinearGap=medium -noDbNameCheck \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -syntenicNet) > swap.log 2>&1 # real 63m47.773s cat fb.regenRn1.chainHg38Link.txt # 827167139 bases of 2282482188 (36.240%) in intersection cat fb.regenRn1.chainSynHg38Link.txt # 788025130 bases of 2282482188 (34.525%) in intersection time (doRecipBest.pl -workhorse=hgwdev -load -buildDir=`pwd` regenRn1 hg38) > rbest.log 2>&1 # real 300m50.570s cat fb.regenRn1.chainRBest.Hg38.txt # 793987134 bases of 2282482188 (34.786%) in intersection ######################################################################### # lastz/chain/net swap mouse/mm10 (DONE - 2019-11-26 - Hiram) # original alignment cd /hive/data/genomes/mm10/bed/lastzRegenRn1.2019-09-12 cat fb.mm10.chainRegenRn1Link.txt # 1699005386 bases of 2652783500 (64.046%) in intersection cat fb.mm10.chainSynRegenRn1Link.txt # 1591231371 bases of 2652783500 (59.983%) in intersection cat fb.mm10.chainRBest.RegenRn1.txt # 1565180527 bases of 2652783500 (59.001%) in intersection mkdir /hive/data/genomes/regenRn1/bed/blastz.mm10.swap cd /hive/data/genomes/regenRn1/bed/blastz.mm10.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/mm10/bed/lastzRegenRn1.2019-09-12/DEF \ -swap -syntenicNet -noDbNameCheck \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=medium) > swap.log 2>&1 # real 94m19.060s cat fb.regenRn1.chainMm10Link.txt # 1634389849 bases of 2282482188 (71.606%) in intersection cat fb.regenRn1.chainSynMm10Link.txt # 1564752158 bases of 2282482188 (68.555%) in intersection time (doRecipBest.pl -workhorse=hgwdev -load -buildDir=`pwd` regenRn1 mm10) > rbest.log 2>&1 # real 488m25.471s cat fb.regenRn1.chainRBest.Mm10.txt # 1565313924 bases of 2282482188 (68.579%) in intersection ############################################################################## # GENBANK AUTO UPDATE (DONE - 2019-09-11 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # #organism mrnaCnt estCnt refSeqCnt # Rattus norvegicus 130204 1103594 17712 # edit etc/genbank.conf to add regenRn1 just before galGal5 # regenRn1 (Regeneron rat assembly) regenRn1.serverGenome = /hive/data/genomes/regenRn1/regenRn1.2bit regenRn1.ooc = /hive/data/genomes/regenRn1/jkStuff/regenRn1.11.ooc regenRn1.lift = /hive/data/genomes/regenRn1/jkStuff/nonBridged.lift regenRn1.refseq.mrna.native.pslCDnaFilter = ${finished.refseq.mrna.native.pslCDnaFilter} regenRn1.refseq.mrna.xeno.pslCDnaFilter = ${finished.refseq.mrna.xeno.pslCDnaFilter} regenRn1.genbank.mrna.native.pslCDnaFilter = ${finished.genbank.mrna.native.pslCDnaFilter} regenRn1.genbank.mrna.xeno.pslCDnaFilter = ${finished.genbank.mrna.xeno.pslCDnaFilter} regenRn1.genbank.est.native.pslCDnaFilter = ${finished.genbank.est.native.pslCDnaFilter} regenRn1.downloadDir = regenRn1 regenRn1.refseq.mrna.xeno.load = yes regenRn1.refseq.mrna.xeno.loadDesc = yes regenRn1.genbank.mrna.xeno.load = yes regenRn1.perChromTables = no regenRn1.mgc = yes # regenRn1.upstreamGeneTbl = ensGene # regenRn1.upstreamMaf = multiz13way /hive/data/genomes/regenRn1/bed/multiz13way/species.list.txt # verify the files specified exist before checking in the file: grep ^regenRn1 etc/genbank.conf | grep hive | awk '{print $NF}' | xargs ls -og # -rw-rw-r-- 1 840371 Sep 9 07:27 /hive/data/genomes/regenRn1/jkStuff/nonBridged.lift # -rw-rw-r-- 1 120012 Sep 11 08:54 /hive/data/genomes/regenRn1/jkStuff/regenRn1.11.ooc # -rw-rw-r-- 1 625487112 Sep 9 07:14 /hive/data/genomes/regenRn1/regenRn1.2bit git commit -m "Added regenRn1 - Regeneron rat refs #21579" etc/genbank.conf git push # update /cluster/data/genbank/: make etc-update # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add regenRn1 to: # etc/align.dbs etc/hgwdev.dbs git add etc/align.dbs etc/hgwdev.dbs git commit -m "Added regenRn1 - Regeneron rat refs #21579" etc/hgwdev.dbs etc/align.dbs git push make etc-update # wait a few days for genbank magic to take place, the tracks will # appear ############################################################################# # augustus gene track (DONE - 2019-09-11 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/augustus cd /hive/data/genomes/regenRn1/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=human -dbHost=hgwdev \ -workhorse=hgwdev regenRn1) > do.log 2>&1 # real 144m4.269s cat fb.regenRn1.augustusGene.txt # 43428122 bases of 2282482188 (1.903%) in intersection ######################################################################### # ncbiRefSeq (TBD - 2019-09-08 - Hiram) # There are no source files for this mkdir /hive/data/genomes/regenRn1/bed/ncbiRefSeq cd /hive/data/genomes/regenRn1/bed/ncbiRefSeq # running step wise just to be careful time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev \ -stop=download -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.5_GRCg6a regenRn1) > download.log 2>&1 # real 1m19.029s time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -continue=process -bigClusterHub=ku -dbHost=hgwdev \ -stop=process -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.5_GRCg6a regenRn1) > process.log 2>&1 # real 2m6.030s time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -continue=load -bigClusterHub=ku -dbHost=hgwdev \ -stop=load -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.5_GRCg6a regenRn1) > load.log 2>&1 # real 0m22.312s cat fb.ncbiRefSeq.regenRn1.txt # 88641893 bases of 1055588482 (8.397%) in intersection # need to add: include ../../refSeqComposite.ra alpha # to the chicken/regenRn1/trackDb.ra to turn on the track in the browser # there was one gene that claimed to have a protein, but the # protein sequence was not included in the protein.faa file # discovered from joinerCheck # manual fix to blank out this one protein, to see the entry hgsql -e 'select * from ncbiRefSeqLink where protAcc="NP_989875.1";' regenRn1 hgsql -e 'update ncbiRefSeqLink set protAcc="" where protAcc="NP_989875.1";' regenRn1 # this makes the 'protein' link disappear from the gene details page # curious that this gene is marked as a non-coding gene anyway ? # gene: FET1 at chr4:63,102,774-63,105,516- featureBits -enrichment regenRn1 refGene ncbiRefSeq # refGene 1.374%, ncbiRefSeq 8.397%, both 1.370%, cover 99.73%, enrich 11.88x featureBits -enrichment regenRn1 ncbiRefSeq refGene # ncbiRefSeq 8.397%, refGene 1.374%, both 1.370%, cover 16.32%, enrich 11.88x featureBits -enrichment regenRn1 ncbiRefSeqCurated refGene # ncbiRefSeqCurated 1.368%, refGene 1.374%, both 1.364%, cover 99.71%, enrich 72.59x featureBits -enrichment regenRn1 refGene ncbiRefSeqCurated # refGene 1.374%, ncbiRefSeqCurated 1.368%, both 1.364%, cover 99.32%, enrich 72.59x ######################################################################### # LIFTOVER TO rn6 (DONE - 2019-09-09 - Hiram) ssh hgwdev mkdir /hive/data/genomes/regenRn1/bed/blat.rn6.2019-09-09 cd /hive/data/genomes/regenRn1/bed/blat.rn6.2019-09-09 doSameSpeciesLiftOver.pl -verbose=2 \ -debug -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/regenRn1/jkStuff/regenRn1.11.ooc \ regenRn1 rn6 time (doSameSpeciesLiftOver.pl -verbose=2 \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/regenRn1/jkStuff/regenRn1.11.ooc \ regenRn1 rn6) > doLiftOverToRn6.log 2>&1 # real 171m37.650s # see if the liftOver menus function in the browser from regenRn1 to rn6 # would like to see this as a track: hgLoadChain regenRn1 chainRn6 regenRn1ToRn6.over.chain.gz # Loading 127292 chains into regenRn1.chainRn6 # real 0m9.461s chainSort regenRn1ToRn6.over.chain.gz stdout \ | chainPreNet stdin \ /hive/data/genomes/regenRn1/chrom.sizes \ /hive/data/genomes/rn6/chrom.sizes stdout \ | chainNet stdin -minSpace=1 /hive/data/genomes/regenRn1/chrom.sizes \ /hive/data/genomes/rn6/chrom.sizes stdout /dev/null \ | netSyntenic stdin noClass.net netClass -verbose=0 -noAr noClass.net regenRn1 rn6 regenRn1.rn6.net netFilter -minGap=10 regenRn1.rn6.net \ | hgLoadNet -verbose=0 regenRn1 netRn6 stdin featureBits regenRn1 chainRn6Link > fb.regenRn1.chainRn6Link.txt 2>&1 cat fb.regenRn1.chainRn6Link.txt # 2211701054 bases of 2282482188 (96.899%) in intersection ######################################################################### # LIFTOVER TO regenRn0 (DONE - 2019-09-09 - Hiram) ssh hgwdev mkdir /hive/data/genomes/regenRn1/bed/blat.regenRn0.2019-09-09 cd /hive/data/genomes/regenRn1/bed/blat.regenRn0.2019-09-09 doSameSpeciesLiftOver.pl -verbose=2 \ -debug -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/regenRn1/jkStuff/regenRn1.11.ooc \ regenRn1 regenRn0 time (doSameSpeciesLiftOver.pl -verbose=2 \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -ooc=/hive/data/genomes/regenRn1/jkStuff/regenRn1.11.ooc \ regenRn1 regenRn0) > doLiftOverToRegenRn0.log 2>&1 # real 176m49.811s # see if the liftOver menus function in the browser from regenRn1 to galGal5 # would like to see this as a track: hgLoadChain regenRn1 chainRegenRn0 regenRn1ToRegenRn0.over.chain.gz # Loading 56 chains into regenRn1.chainRegenRn0 chainSort regenRn1ToRegenRn0.over.chain.gz stdout \ | chainPreNet stdin \ /hive/data/genomes/regenRn1/chrom.sizes \ /hive/data/genomes/regenRn0/chrom.sizes stdout \ | chainNet stdin -minSpace=1 /hive/data/genomes/regenRn1/chrom.sizes \ /hive/data/genomes/regenRn0/chrom.sizes stdout /dev/null \ | netSyntenic stdin noClass.net netClass -verbose=0 -noAr noClass.net regenRn1 regenRn0 regenRn1.regenRn0.net netFilter -minGap=10 regenRn1.regenRn0.net \ | hgLoadNet -verbose=0 regenRn1 netRegenRn0 stdin featureBits regenRn1 chainRegenRn0Link > fb.regenRn1.chainRegenRn0Link.txt 2>&1 cat fb.regenRn1.chainRegenRn0Link.txt # 2395098699 bases of 2282482188 (104.934%) in intersection ######################################################################### # BLATSERVERS ENTRY (DONE - 2019-09-13 - 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 ("regenRn1", "blat1c", "17912", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("regenRn1", "blat1c", "17913", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################ ## reset default position to MEPE gene (egg shell protein) ## (TBD - 2019-09-08 - Hiram) # as found from the galGal5 to regenRn1 liftOver ssh hgwdev hgsql -e 'update dbDb set defaultPos="chr4:45667017-45672928" where name="regenRn1";' hgcentraltest ######################################################################### # all.joiner update, downloads and in pushQ - (TBD - 2018-10-17 - Hiram) xyz cd $HOME/kent/src/hg/makeDb/schema # verify all the business is done for release ~/kent/src/hg/utils/automation/verifyBrowser.pl regenRn1 # fixup all.joiner until this is a clean output joinerCheck -database=regenRn1 -tableCoverage all.joiner joinerCheck -database=regenRn1 -times all.joiner joinerCheck -database=regenRn1 -keys all.joiner # when clean, check in: git commit -m 'adding rules for regenRn1 refs #21579' all.joiner git push # run up a 'make alpha' in hg/hgTables to get this all.joiner file # into the hgwdev/genome-test system cd /hive/data/genomes/regenRn1 time (makeDownloads.pl regenRn1) > downloads.log 2>&1 XXX - running - Tue Oct 22 08:15:47 PDT 2019 # real 10m7.605s # now ready for pushQ entry mkdir /hive/data/genomes/regenRn1/pushQ cd /hive/data/genomes/regenRn1/pushQ time (makePushQSql.pl -redmineList regenRn1) > regenRn1.pushQ.sql 2> stderr.out # real 9m58.779s # remove the extra chainNet files from the listings: sed -i -e "/etNig1/d" redmine.regenRn1.file.list sed -i -e "/asAcu1/d" redmine.regenRn1.file.list sed -i -e "/etNig1/d" redmine.regenRn1.table.list sed -i -e "/onAlb1/d" redmine.regenRn1.table.list sed -i -e "/asAcu1/d" redmine.regenRn1.table.list sed -i -e "/Stickleback/d" redmine.regenRn1.releaseLog.txt sed -i -e "/Tetraodon/d" redmine.regenRn1.releaseLog.txt sed -i -e "/sparrow/d" redmine.regenRn1.releaseLog.txt # remove the tandemDups and gapOverlap from the file list: sed -i -e "/tandemDups/d" redmine.regenRn1.table.list sed -i -e "/Tandem Dups/d" redmine.regenRn1.releaseLog.txt sed -i -e "/gapOverlap/d" redmine.regenRn1.table.list sed -i -e "/Gap Overlaps/d" redmine.regenRn1.releaseLog.txt # real 7m21.629s # check for errors in stderr.out, some are OK, e.g.: # WARNING: hgwdev does not have /gbdb/regenRn1/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/regenRn1/wib/quality.wib # WARNING: hgwdev does not have /gbdb/regenRn1/bbi/quality.bw # WARNING: regenRn1 does not have seq # WARNING: regenRn1 does not have extFile # add the path names to the listing files in the redmine issue # in the three appropriate entry boxes: # /hive/data/genomes/regenRn1/pushQ/redmine.regenRn1.file.list # /hive/data/genomes/regenRn1/pushQ/redmine.regenRn1.releaseLog.txt # /hive/data/genomes/regenRn1/pushQ/redmine.regenRn1.table.list ############################################################################## # crispr whole genome (DONE - 2019-10-15,22 - Hiram) mkdir /hive/data/genomes/regenRn1/bed/crisprAll cd /hive/data/genomes/regenRn1/bed/crisprAll # the large shoulder argument will cause the entire genome to be scanned # and since this is a new genome, samtools fa indexing will take place time (~/kent/src/hg/utils/automation/doCrispr.pl -verbose=2 -stop=ranges \ regenRn1 refGene -shoulder=250000000 -tableName=crisprAll \ -fileServer=hgwdev -buildDir=`pwd` -smallClusterHub=hgwdev \ -bigClusterHub=ku -workhorse=hgwdev) > ranges.log 2>&1 # real 65m1.474s time (~/kent/src/hg/utils/automation/doCrispr.pl -verbose=2 \ -continue=guides -stop=guides \ regenRn1 knownGene -shoulder=250000000 -tableName=crisprAll -fileServer=hgwdev \ -buildDir=`pwd` -smallClusterHub=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev) > guides.log 2>&1 # real 5m21.112s # one of the jobs failed on ku, but worked OK on hgwdev stand alone /cluster/software/bin/python \ /hive/data/outside/crisprTrack/scripts/findGuides.py \ /hive/data/genomes/regenRn1/bed/crisprAll/ranges/tmp/040.fa \ /hive/data/genomes/regenRn1/chrom.sizes tmp/040.bed tmp/040.fa Completed: 99 of 100 jobs Crashed: 1 jobs CPU time in finished jobs: 9475s 157.92m 2.63h 0.11d 0.000 y IO & Wait Time: 371s 6.18m 0.10h 0.00d 0.000 y Average job time: 99s 1.66m 0.03h 0.00d Longest finished job: 524s 8.73m 0.15h 0.01d Submission to last job: 539s 8.98m 0.15h 0.01d time (~/kent/src/hg/utils/automation/doCrispr.pl -verbose=2 \ -continue=specScoreJobList -stop=specScores -fileServer=hgwdev \ regenRn1 knownGene -shoulder=250000000 -tableName=crisprAll \ -buildDir=`pwd` -smallClusterHub=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev) > specScores.log 2>&1 # real 4764m4.884s # user 0m1.693s # sys 0m1.684s # Number of specScores: 204371688 # Completed: 2733924 of 2733924 jobs # CPU time in finished jobs: 220930932s 3682182.20m 61369.70h 2557.07d 7.006 y # IO & Wait Time: 5388062s 89801.04m 1496.68h 62.36d 0.171 y # Average job time: 83s 1.38m 0.02h 0.00d # Longest finished job: 189s 3.15m 0.05h 0.00d # Submission to last job: 228205s 3803.42m 63.39h 2.64d ### remember to get back to hgwdev to run this time (~/kent/src/hg/utils/automation/doCrispr.pl -verbose=2 \ -continue=effScores -stop=load regenRn1 knownGene \ -shoulder=250000000 -tableName=crisprAll -fileServer=hgwdev \ -buildDir=`pwd` -smallClusterHub=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev) > load.log 2>&1 # real 834m5.635s # user 0m0.962s # sys 0m1.894s # effScores: Completed: 24067 of 24067 jobs CPU time in finished jobs: 12223610s 203726.84m 3395.45h 141.48d 0.388 y IO & Wait Time: 143817s 2396.95m 39.95h 1.66d 0.005 y Average job time: 514s 8.56m 0.14h 0.01d Longest finished job: 1269s 21.15m 0.35h 0.01d Submission to last job: 12710s 211.83m 3.53h 0.15d # offTargets: Completed: 136697 of 136697 jobs CPU time in finished jobs: 2016502s 33608.36m 560.14h 23.34d 0.064 y IO & Wait Time: 2570740s 42845.67m 714.09h 29.75d 0.082 y Average job time: 34s 0.56m 0.01h 0.00d Longest finished job: 77s 1.28m 0.02h 0.00d Submission to last job: 4643s 77.38m 1.29h 0.05d + # cleaning 2021-04-24 - Hiram + time (~/kent/src/hg/utils/automation/doCrispr.pl -verbose=2 \ + -continue=cleanup regenRn1 -tableName=crisprAll -fileServer=hgwdev \ + -buildDir=`pwd` -smallClusterHub=hgwdev -bigClusterHub=ku \ + -workhorse=hgwdev) > cleanup.log 2>&1 + # real 286m17.939s + #########################################################################