5002bc264773860c46470525d2f8802f2c45679a hiram Wed Sep 11 09:37:10 2019 -0700 liftOver to regenRn0 and regenRn1 refs #21579 diff --git src/hg/makeDb/doc/rn6.txt src/hg/makeDb/doc/rn6.txt index c5a770f..c2165bc 100644 --- src/hg/makeDb/doc/rn6.txt +++ src/hg/makeDb/doc/rn6.txt @@ -1,4383 +1,4444 @@ # for emacs: -*- mode: sh; -*- # # the above keeps emacs happy while working with this text document # This file describes how we made the browser database on the # Rattus norvegicus genome, March 2012 update (Rnor5.0) from Baylor. # http://www.ncbi.nlm.nih.gov/bioproject/16219 # http://www.ncbi.nlm.nih.gov/genome/73 # http://www.ncbi.nlm.nih.gov/Traces/wgs/?val=AABR07 # Genome Coverage : 3x BAC; 6x WGS ABI Sanger reads 10.0x PcaBio # chrMt: NC_001665.2 # DATE: 01-Jul-2014 # ORGANISM: Rattus norvegicus # TAXID: 10116 # ASSEMBLY LONG NAME: Rnor_6.0 # ASSEMBLY SHORT NAME: Rnor_6.0 # ASSEMBLY SUBMITTER: Rat Genome Sequencing Consortium # ASSEMBLY TYPE: Haploid # NUMBER OF ASSEMBLY-UNITS: 2 # ASSEMBLY ACCESSION: GCA_000001895.4 # ##Below is a 2 column list with assembly-unit id and name. # ##The Primary Assembly unit is listed first. # GCA_000000225.4 Primary Assembly # GCA_000002045.1 non-nuclear # FTP-RELEASE DATE: 03-Jul-2014 ######################################################################### ## Download sequence (DONE - 2014-07-07 - Hiram) mkdir -p /hive/data/genomes/rn6/genbank cd /hive/data/genomes/rn6/genbank time rsync -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Rattus_norvegicus/Rnor_6.0/ ./ # sent 4565 bytes received 2064907254 bytes 10350435.18 bytes/sec # total size is 2064638723 speedup is 1.00 # real 3m19.289s # measure total sequence: faSize Primary_Assembly/assembled_chromosomes/FASTA/chr*.fa.gz \ Primary_Assembly/unlocalized_scaffolds/FASTA/chr*.fa.gz \ Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fa.gz \ non-nuclear/assembled_chromosomes/FASTA/chrMT.fa.gz # 2870184193 bases (140322104 N's 2729862089 real 2729862089 upper 0 lower) # in 955 sequences in 46 files # Total size: mean 3005428.5 sd 21316687.8 # min 230 (gi|661406602|gb|AABR07024268.1|) # max 282763074 (gi|661902784|gb|CM000072.5|) median 7105 ######################################################################### # fixup names for UCSC standards (DONE - 2014-07-07 - Hiram) mkdir /hive/data/genomes/rn6/ucsc cd /hive/data/genomes/rn6/ucsc # can re-use the scripts from hg38 build cp -p ../../hg38/ucsc/ucscCompositeAgp.pl . time ./ucscCompositeAgp.pl # real 0m53.149s cp -p ../../hg38/ucsc/unlocalized.pl . # fixed this script to be more generic so it should work on any # assembly in the future as long as the fasta headers maintain the # observed structre time ./unlocalized.pl # real 0m2.097s cp -p ../../hg38/ucsc/unplaced.pl . # fixed this script to be more generic so it should work on any # assembly in the future as long as the fasta headers maintain the # observed structre time ./unplaced.pl # real 0m0.494s zcat ../genbank/non-nuclear/assembled_chromosomes/FASTA/chrMT.fa.gz \ | sed -e 's/^>.*/>chrM/' > chrM.fa zcat ../genbank/non-nuclear/assembled_chromosomes/AGP/chrMT.comp.agp.gz \ | grep AY172581 | sed -e 's/^AY172581.1/chrM/;' > chrM.agp # more efficient to have everything zipped: time gzip *.fa # real 16m8.206s time gzip *.agp # real 0m1.725s # verify all the sequence is still here after all this rigamarole: time faSize *.fa.gz # 2870184193 bases (140322104 N's 2729862089 real 2729862089 upper 0 lower) # in 955 sequences in 46 files # Total size: mean 3005428.5 sd 21316687.8 min 230 (chrUn_AABR07024268v1) # max 282763074 (chr1) median 7105 # real 1m3.343s # verify same numbers as was in the original files measured above: # 2870184193 bases (140322104 N's 2729862089 real 2729862089 upper 0 lower) # in 955 sequences in 46 files # Total size: mean 3005428.5 sd 21316687.8 # min 230 (gi|661406602|gb|AABR07024268.1|) # max 282763074 (gi|661902784|gb|CM000072.5|) median 7105 ######################################################################### # Create .ra file and run makeGenomeDb.pl (DONE - Hiram - 2014-07-07) cd /hive/data/genomes/rn6 cat << '_EOF_' >rn6.config.ra # Config parameters for makeGenomeDb.pl: db rn6 clade mammal scientificName Rattus norvegicus commonName Rat assemblyDate Jul. 2014 assemblyLabel RGSC Rnor_6.0 assemblyShortLabel RGSC 6.0 orderKey 1558 # chrMt NC_001665.2 included in the genbank release mitoAcc none fastaFiles /hive/data/genomes/rn6/ucsc/chr*.fa.gz agpFiles /hive/data/genomes/rn6/ucsc/chr*.agp.gz # qualFiles none dbDbSpeciesDir rat photoCreditURL http://www.genome.gov/10005141 photoCreditName NHGRI Press Photos ncbiGenomeId 73 ncbiAssemblyId 382928 ncbiAssemblyName Rnor_6.0 ncbiBioProject 10629 genBankAccessionID GCA_000001895.4 taxId 10116 # http://www.ncbi.nlm.nih.gov/Traces/wgs/?val=AABR07 '_EOF_' # << happy emacs # run agp step first to verify fasta and agp files agree time makeGenomeDb.pl -stop=agp rn6.config.ra > agp.log 2>&1 # ERROR: duplicate sequence found in rn6.unmasked.2bit # echo ERROR: duplicate sequence found in rn6.unmasked.2bit twoBitDup rn6.unmasked.2bit # chrUn_AABR07022993v1 and chrUn_AABR07022518v1 are identical # chrUn_AABR07023006v1 and chrUn_AABR07022518v1 are identical # remove chrUn_AABR07022518v1 chrUn_AABR07023006v1 cd ucsc echo "chrUn_AABR07022518v1 chrUn_AABR07023006v1" > dup.list faSomeRecords -exclude chrUn.fa.gz dup.list chrUn.clean.fa mv chrUn.fa.gz chrUn.fa.gz.dups mv chrUn.clean.fa chrUn.fa gzip chrUn.fa zcat chrUn.agp.gz | egrep -v "chrUn_AABR07022518v1|chrUn_AABR07023006v1" \ > chrUn.agp mv chrUn.agp.gz chrUn.agp.gz.dups gzip chrUn.agp cd .. rm -fr bed jkStuff rn6.unmasked.2bit mv agp.log agp.log.0 time makeGenomeDb.pl -stop=agp rn6.config.ra > agp.log 2>&1 # verify end of agp.log indictates: # All AGP and FASTA entries agree - both files are valid # the non-dup sequence is now: twoBitToFa rn6.unmasked.2bit stdout | faSize stdin # 2870182909 bases (140322104 N's 2729860805 real 2729860805 upper # 0 lower) in 953 sequences in 1 files # continue with the build, don't let this run without a workhorse # specified, it can end up on a machine that can't run the wigToBigWig # for the gc5Base track: time makeGenomeDb.pl -workhorse=hgwdev -fileServer=hgwdev \ -continue=db rn6.config.ra > db.log 2>&1 # real 23m10.146s ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2014-07-09 - Hiram) export maxLen=`hgsql -N -e 'select frag from gold;' rn6 | awk '{print length($0)}' | sort -run | head -1` echo $maxLen # 14 export C=1 while [ $C -le $maxLen ]; do echo -n " $C: " hgsql -N -e 'select frag from gold;' rn6 | sort -u \ | awk '{ print substr($0,'$C',1) }' | sort -u | xargs echo | sed -e 's/ //g' C=`echo $C | awk '{print $1+1}'` done # 1: A # 2: ACY # 3: 012B # 4: 012345789R # 5: 0123456789 # 6: 0123456789 # 7: 0123456789 # 8: 0123456789 # 9: .0123456789 # 10: 0123456789 # 11: 0123456789 # 12: 0123456789 # 13: . # 14: 1 # since we have the composite fragment IDs, there are a variety of # dot versions: hgsql -N -e 'select frag from gold;' rn6 | sed -e 's/.*\.//' \ | sort | uniq -c | sort -rn # 73744 1 # 574 5 # 540 4 # 390 6 # ... etc ... # hence, add to trackDb/rat/rn6/trackDb.ra searchTable gold searchMethod prefix searchType bed shortCircuit 1 termRegex A[ACY][B012][R0-9][0-9]+(\.[0-9]+)* query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 # test pattern: hgsql -N -e 'select frag from gold;' rn6 | wc -l # 76195 hgsql -N -e 'select frag from gold;' rn6 \ | egrep -e 'A[ACY][B012][R0-9][0-9]+(\.[0-9]+)*' | wc -l # 76195 hgsql -N -e 'select frag from gold;' rn6 \ | egrep -v -e 'A[ACY][B012][R0-9][0-9]+(\.[0-9]+)*' | wc -l # 0 ############################################################################ # create ucscToINSDC name mapping (DONE - 2014-04-11 - Hiram) mkdir /hive/data/genomes/rn6/bed/ucscToINSDC cd /hive/data/genomes/rn6/bed/ucscToINSDC # this script has been maturing over time, it is close to complete. # to find a latest copy of it: # ls -ogrt /hive/data/genomes/*/bed/ucscToINSDC/translateNames.sh cp -p /hive/data/genomes/poeRet1/bed/ucscToINSDC/translateNames.sh . # to see chrM accession: hgsql -e 'select * from gold;' rn6 | grep chrM # use that accession here: ./translateNames.sh AY172581.1 # eliminate the duplicates, and needs to be sorted to work with join egrep -v "chrUn_AABR07022518v1|chrUn_AABR07023006v1" ucscToINSDC.txt \ | sort > ucscToINSDC.tab awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes | sort \ > name.coordinate.tab join name.coordinate.tab ucscToINSDC.tab | tr '[ ]' '[\t]' > ucscToINSDC.bed cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1 # 27 # use the 27 in this sed: sed -e "s/21/27/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab rn6 ucscToINSDC stdin ucscToINSDC.bed checkTableCoords rn6 ucscToINSDC # should cover all bases featureBits -countGaps rn6 ucscToINSDC # 2870182909 bases of 2870182909 (100.000%) in intersection ############################################################################## # running repeat masker (DONE - 2014-07-07,09 - Hiram) mkdir /hive/data/genomes/rn6/bed/repeatMasker cd /hive/data/genomes/rn6/bed/repeatMasker time doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku rn6 > do.log 2>&1 & # real 1921m43.379s cat faSize.rmsk.txt # 2870182909 bases (140322104 N's 2729860805 real 1551246793 upper # 1178614012 lower) in 953 sequences in 1 files # Total size: mean 3011734.4 sd 21338622.2 min 230 (chrUn_AABR07024268v1) # max 282763074 (chr1) median 7294 # %41.06 masked total, %43.17 masked real grep -i versi do.log # RepeatMasker version open-4.0.5 # January 31 2015 (open-4-0-5) version of RepeatMasker featureBits -countGaps rn6 rmsk # 1178980732 bases of 2870182909 (41.077%) in intersection # 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 count lower case N's ########################################################################## # running simple repeat (DONE - 2014-07-07 - Hiram) mkdir /hive/data/genomes/rn6/bed/simpleRepeat cd /hive/data/genomes/rn6/bed/simpleRepeat time doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku rn6 > do.log 2>&1 & # real 8m49.125s cat fb.simpleRepeat # 106804757 bases of 2731211732 (3.911%) in intersection # add the TRF mask to the rmsk sequence: # it masks more sequence cd /hive/data/genomes/rn6 twoBitMask rn6.rmsk.2bit \ -add bed/simpleRepeat/trfMask.bed rn6.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa rn6.2bit stdout | faSize stdin > faSize.rn6.2bit.txt cat faSize.rn6.2bit.txt # 2870182909 bases (140322104 N's 2729860805 real 1547693783 upper # 1182167022 lower) in 953 sequences in 1 files # Total size: mean 3011734.4 sd 21338622.2 min 230 (chrUn_AABR07024268v1) # max 282763074 (chr1) median 7294 # %41.19 masked total, %43.31 masked real # replace the previous symLink which goes to the unmasked 2bit rm /gbdb/rn6/rn6.2bit ln -s `pwd`/rn6.2bit /gbdb/rn6/rn6.2bit ######################################################################### # Verify all gaps are marked (DONE - 2014-07-08 - Hiram) mkdir /hive/data/genomes/rn6/bed/gap cd /hive/data/genomes/rn6/bed/gap time nice -n +19 findMotif -motif=gattaca -verbose=4 \ -strand=+ ../../rn6.unmasked.2bit > findMotif.txt 2>&1 # real 0m40.856s grep "^#GAP " findMotif.txt | sed -e "s/^#GAP //" > allGaps.bed time featureBits -countGaps rn6 -not gap -bed=notGap.bed # 2731211732 bases of 2870182909 (95.158%) in intersection # real 0m15.822s time featureBits -countGaps rn6 allGaps.bed notGap.bed -bed=new.gaps.bed # 1350927 bases of 2870182909 (0.047%) in intersection # real 0m56.160s awk '{print $3-$2}' new.gaps.bed | ave stdin | grep total # total 1350927.000000 awk '{print $3-$2,$0}' new.gaps.bed | sort -nr | head # 10377 chr2 172413175 172423552 chr2.4447 # 7000 chr8 119186607 119193607 chr8.2918 # 6344 chrX 123693517 123699861 chrX.3045 # what is the highest index in the existing gap table: hgsql -N -e "select ix from gap;" rn6 | sort -n | tail -1 # 14766 cat << '_EOF_' > mkGap.pl #!/bin/env perl use strict; use warnings; my $ix=`hgsql -N -e "select ix from gap;" rn6 | sort -n | tail -1`; chomp $ix; open (FH,"<new.gaps.bed") or die "can not read new.gaps.bed"; while (my $line = <FH>) { my ($chrom, $chromStart, $chromEnd, $rest) = split('\s+', $line); ++$ix; printf "%s\t%d\t%d\t%d\tN\t%d\tother\tyes\n", $chrom, $chromStart, $chromEnd, $ix, $chromEnd-$chromStart; } close (FH); '_EOF_' # << happy emacs chmod +x ./mkGap.pl ./mkGap.pl > other.bed featureBits -countGaps rn6 other.bed # 1350927 bases of 2870182909 (0.047%) in intersection wc -l other.bed # 75068 # verify no mistake here: featureBits -countGaps rn6 gap other.bed # 0 bases of 2870182909 (0.000%) in intersection hgLoadBed -sqlTable=$HOME/kent/src/hg/lib/gap.sql \ -noLoad rn6 otherGap other.bed # starting with this many hgsql -e "select count(*) from gap;" rn6 # 70664 hgsql rn6 -e 'load data local infile "bed.tab" into table gap;' # result count: hgsql -e "select count(*) from gap;" rn6 # 145732 calc 70664 \+ 75068 # 70664 + 75068 = 145732.000000 # verify we aren't adding gaps where gaps already exist # this would output errors if that were true: gapToLift -minGap=1 rn6 nonBridged.lift -bedFile=nonBridged.bed # see example in danRer7.txt when problems arise # this can happen and is not an error: # WARNING: overlapping gap at chrY:692274-692374(scaffold) and chrY:692374-692375(other) # the scaffold following the (scaffold) gap begins with a single N # how that can happen I have no idea # there are bridged and non-bridged gaps here: hgsql -N -e "select bridge from gap;" rn6 | sort | uniq -c # 440 no # 145292 yes ######################################################################### # cytoBandIdeo - (DONE - 2014-07-14 - Hiram) mkdir /hive/data/genomes/rn6/bed/cytoBand cd /hive/data/genomes/rn6/bed/cytoBand makeCytoBandIdeo.csh rn6 ######################################################################### ## CYTOBAND - ideogram track (DONE - 2017-03-15 - Hiram) mkdir /hive/data/genomes/rn6/bed/cytoBand cd /hive/data/genomes/rn6/bed/cytoBand # fetch the ideogram file: rsync -a -P rsync://ftp.ncbi.nlm.nih.gov/pub/gdp/ideogram_10116_GCF_000000225.4_NA_V1 ./ # Create bed file $HOME/kent/src/utils/ncbi/createNcbiCytoBand.pl \ ideogram_10116_GCF_000000225.4_NA_V1 # add in the default bands for all other chroms: egrep "_|chrM" firstTrack/rn6.cytoBand \ | awk '{printf "%s\t%s\t%d\t%d\t%s\n", $1,$1,$2,$3,$4}' \ | sed -e 's/^chr[nUXY0-9]\+_//; s/v1_random/.1/;' \ | awk '{printf "%s\t%d\t%d\t\t%s\n", $2,$3,$4,$5}' >> cytoBand.bed ## can now verify before load: $HOME/kent/src/utils/ncbi/cytoBandVerify.pl # everything checks out OK on 953 chroms # these chrom names are too long for the SQL keys: cut -f1 cytoBand.bed | sort -u | awk '{print length($0)}' | sort -rn | head -1 # 27 sed -e 's/12/27/;' $HOME/kent/src/hg/lib/cytoBand.sql \ > cytoBand.sql # Load the bed file hgLoadBed -tab -noBin -sqlTable=cytoBand.sql rn6 cytoBand cytoBand.bed # Read 1179 elements of size 5 from cytoBand.bed # Make cytoBandIdeo track for ideogram gif on hgTracks page. # The cytoBandIdeo is just a replicate of the cytoBand track. hgsql -e "CREATE TABLE cytoBandIdeo SELECT * FROM cytoBand;" rn6 checkTableCoords rn6 cytoBand checkTableCoords rn6 cytoBandIdeo # should cover %100 featureBits -countGaps rn6 cytoBand # 2870182909 bases of 2870182909 (100.000%) in intersection featureBits -countGaps rn6 cytoBandIdeo # 2870182909 bases of 2870182909 (100.000%) in intersection ########################################################################## ## WINDOWMASKER (DONE - 2014-07-09 - Hiram) mkdir /hive/data/genomes/rn6/bed/windowMasker cd /hive/data/genomes/rn6/bed/windowMasker time nice -n +19 doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev rn6 > do.log 2>&1 & # real 205m59.097s # Masking statistics cat fb.rn6.windowmaskerSdust.clean.txt # 949668059 bases of 2870182909 (33.087%) in intersection cat fb.rn6.rmsk.windowmaskerSdust.txt # 712388228 bases of 2870182909 (24.820%) in intersection ######################################################################### # MASK SEQUENCE WITH WM+TRF # not running this since RM + TRF is plenty of masking # cd /hive/data/genomes/rn6 # twoBitMask -add bed/windowMasker/rn6.cleanWMSdust.2bit \ # bed/simpleRepeat/trfMask.bed rn6.2bit # safe to ignore the warnings about BED file with >=13 fields # twoBitToFa rn6.2bit stdout | faSize stdin > faSize.rn6.txt # cat faSize.rn6.txt # create symlink to gbdb # ssh hgwdev # rm /gbdb/rn6/rn6.2bit # ln -s `pwd`/rn6.2bit /gbdb/rn6/rn6.2bit ######################################################################### # PREPARE LINEAGE SPECIFIC REPEAT FILES FOR BLASTZ (TBD - 2012-03-23 - Hiram) ssh ku mkdir /hive/data/genomes/rn6/bed/linSpecRep cd /hive/data/genomes/rn6/bed/linSpecRep # split the RM output by chromosome name into separate files mkdir rmsk dateRepeats head -3 ../repeatMasker/rn6.sorted.fa.out > rmsk.header.txt headRest 3 ../repeatMasker/rn6.sorted.fa.out \ | splitFileByColumn -ending=.out -col=5 -head=rmsk.header.txt stdin rmsk ls -1S rmsk/* > rmOut.list wc -l rmOut.list # 904 rmOut.list wc -l ../../chrom.sizes # 953 ../../chrom.sizes #evidently there are 49 segments without RM annotations, some are very small cat << '_EOF_' > mkLSR #!/bin/csh -fe rm -f dateRepeats/$1_homo-sapiens_mus-musculus /scratch/data/RepeatMasker140131/DateRepeats \ $1 -query rat -comp human -comp mouse mv $1_homo-sapiens_mus-musculus dateRepeats '_EOF_' # << happy emacs chmod +x mkLSR cat << '_EOF_' > template #LOOP ./mkLSR $(path1) {check out line+ dateRepeats/$(file1)_homo-sapiens_mus-musculus} #ENDLOOP '_EOF_' # << happy emacs gensub2 rmOut.list single template jobList para create jobList para try ... check ... push ... etc... para time # Completed: 904 of 904 jobs # CPU time in finished jobs: 21844s 364.06m 6.07h 0.25d 0.001 y # IO & Wait Time: 20183s 336.39m 5.61h 0.23d 0.001 y # Average job time: 46s 0.77m 0.01h 0.00d # Longest finished job: 96s 1.60m 0.03h 0.00d # Submission to last job: 682s 11.37m 0.19h 0.01d mkdir notInHuman notInMouse for F in dateRepeats/chr*.out_homo-sapiens* do B=`basename ${F}` B=${B/.out*/} echo $B /cluster/bin/scripts/extractRepeats 1 ${F} > \ notInHuman/${B}.out.spec /cluster/bin/scripts/extractRepeats 2 ${F} > \ notInMouse/${B}.out.spec done # Verify that these two things are actually different # To check identical find ./notInHuman ./notInMouse -name "*.out.spec" | \ while read FN; do echo `cat ${FN} | sum -r` ${FN}; done \ | sort -k1,1n | sort -t"/" -k3,3 > check.same # some of them are the same, but not all: sed -e 's#./notInHuman/##; s#./notInMouse/##' check.same \ | sort | uniq -c | sort -rn | less # you will see a count of two at the beginning, but it becomes one soon # Copy to data/staging for cluster replication mkdir /hive/data/staging/data/rn6 rsync -a -P ./notInMouse/ /hive/data/staging/data/rn6/notInMouse/ rsync -a -P ./notInHuman/ /hive/data/staging/data/rn6/notInHuman/ # We also need the nibs for the lastz runs with lineage specific repeats mkdir /hive/data/staging/data/rn6/nib mkdir /hive/data/genomes/rn6/nib cd /hive/data/genomes/rn6 cut -f1 chrom.sizes | while read C do twoBitToFa -seq=${C} rn6.2bit stdout | faToNib -softMask stdin nib/${C}.nib ls -og nib/$C.nib done # verify sequence remains the same cut -f1 chrom.sizes | while read C do nibFrag -masked nib/${C}.nib 0 `grep -w ${C} chrom.sizes | cut -f2` + \ stdout done | faSize stdin # 2870182909 bases (140322104 N's 2729860805 real 1547693783 upper # 1182167022 lower) in 953 sequences in 1 files # Total size: mean 3011734.4 sd 21338622.2 # min 230 (nib/chrUn_AABR07024268v1.nib:0-230) # max 282763074 (nib/chr1.nib:0-282763074) median 7294 # %41.19 masked total, %43.31 masked real # Compare to original: cat faSize.rn6.2bit.txt # 2870182909 bases (140322104 N's 2729860805 real 1547693783 upper # 1182167022 lower) in 953 sequences in 1 files # Total size: mean 3011734.4 sd 21338622.2 # min 230 (chrUn_AABR07024268v1) max 282763074 (chr1) median 7294 # %41.19 masked total, %43.31 masked real # Copy to data/genomes staging for cluster replication rsync -a -P ./nib/ /hive/data/staging/data/rn6/nib/ ######################################################################### # cpgIslands - (DONE - 2014-07-09 - Hiram) mkdir /hive/data/genomes/rn6/bed/cpgIslands cd /hive/data/genomes/rn6/bed/cpgIslands time (doCpgIslands.pl -workhorse=hgwdev -buildDir=`pwd` \ -dbHost=hgwdev -smallClusterHub=ku -bigClusterHub=ku rn6) > do.log 2>&1 # real 2m32.227s cat fb.rn6.cpgIslandExt.txt # 11404910 bases of 2729860805 (0.418%) in intersection ############################################################################# # CPG Islands Unmasked track (DONE - 2014-07-08 - Hiram) mkdir /hive/data/genomes/rn6/bed/cpgIslandsUnmasked cd /hive/data/genomes/rn6/bed/cpgIslandsUnmasked time (doCpgIslands.pl -buildDir=`pwd` -bigClusterHub=ku \ -tableName=cpgIslandExtUnmasked -dbHost=hgwdev -smallClusterHub=ku \ -workhorse=hgwdev \ -maskedSeq=/hive/data/genomes/rn6/rn6.unmasked.2bit rn6) > do.log 2>&1 # real 14m0.085s cat fb.rn6.cpgIslandExtUnmasked.txt # 13483527 bases of 2731211732 (0.494%) in intersection ######################################################################### # genscan - (DONE - 2014-07-09 - Hiram) mkdir /hive/data/genomes/rn6/bed/genscan cd /hive/data/genomes/rn6/bed/genscan time (doGenscan.pl -bigClusterHub=ku -workhorse=hgwdev \ -dbHost=hgwdev -buildDir=`pwd` rn6) > do.log 2>&1 & # real 137m3.209s cat fb.rn6.genscan.txt # 57944086 bases of 2729860805 (2.123%) in intersection cat fb.rn6.genscanSubopt.txt # 62473326 bases of 2729860805 (2.289%) in intersection ######################################################################### # MAKE 11.OOC FILE FOR BLAT/GENBANK (DONE - 2014-07-09 - Hiram) # Use -repMatch=1000, based on size -- for human we use 1024 # use the "real" number from the faSize measurement, # hg19 is 2897316137, calculate the ratio factor for 1024: calc \( 2729860805 / 2897316137 \) \* 1024 # ( 2729860805 / 2897316137 ) * 1024 = 964.816172 # round up to 1000 (rn5 was 950) cd /hive/data/genomes/rn6 time blat rn6.2bit /dev/null /dev/null -tileSize=11 \ -makeOoc=jkStuff/rn6.11.ooc -repMatch=1000 # Wrote 27021 overused 11-mers to jkStuff/rn6.11.ooc # rn5 had: Wrote 34513 overused 11-mers to jkStuff/rn6.11.ooc # real 1m10.077s # there are non-bridged gaps, create lift file needed for genbank hgsql -N -e "select bridge from gap;" rn6 | sort | uniq -c # 440 no # 145292 yes cd /hive/data/genomes/rn6/jkStuff gapToLift rn6 rn6.nonBridged.lift -bedFile=rn6.nonBridged.bed # largest non-bridged contig: awk '{print $3-$2,$0}' rn6.nonBridged.bed | sort -nr | head # 52752077 chr18 35449852 88201929 chr18.06 ######################################################################### # AUTO UPDATE GENBANK (DONE - 2012-05-04 - Hiram) # examine the file: /cluster/data/genbank/data/organism.lst # for your species to see what counts it has for: # organism mrnaCnt estCnt refSeqCnt # Rattus norvegicus 125869 1103595 17503 # to decide which "native" mrna or ests you want to specify in genbank.conf ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # edit etc/genbank.conf to add: # rn6 (rat) rn6.serverGenome = /hive/data/genomes/rn6/rn6.2bit rn6.clusterGenome = /hive/data/genomes/rn6/rn6.2bit rn6.ooc = /hive/data/genomes/rn6/jkStuff/rn6.11.ooc rn6.lift = /hive/data/genomes/rn6/jkStuff/rn6.nonBridged.lift rn6.refseq.mrna.native.pslCDnaFilter = ${finished.refseq.mrna.native.pslCDnaFilter} rn6.refseq.mrna.xeno.pslCDnaFilter = ${finished.refseq.mrna.xeno.pslCDnaFilter} rn6.genbank.mrna.native.pslCDnaFilter = ${finished.genbank.mrna.native.pslCDnaFilter} rn6.genbank.mrna.xeno.pslCDnaFilter = ${finished.genbank.mrna.xeno.pslCDnaFilter} rn6.genbank.est.native.pslCDnaFilter = ${finished.genbank.est.native.pslCDnaFilter} rn6.downloadDir = rn6 rn6.refseq.mrna.xeno.load = yes rn6.refseq.mrna.xeno.loadDesc = yes rn6.genbank.mrna.xeno.load = yes rn6.perChromTables = no rn6.mgc = yes # rn6.upstreamGeneTbl = ensGene # rn6.upstreamMaf = multiz13way # /hive/data/genomes/rn6/bed/multiz13way/species.list.txt # end of section added to etc/genbank.conf git commit -m "adding rn6 rat refs #13578" etc/genbank.conf git push make etc-update ssh hgwdev # used to do this on "genbank" machine screen -S rn6 # long running job managed in screen cd /cluster/data/genbank time nice -n +19 ./bin/gbAlignStep -initial rn6 & # var/build/logs/2014.07.09-12:07:40.rn6.initalign.log # real 902m50.425s # load database when finished ssh hgwdev cd /cluster/data/genbank time nice -n +19 ./bin/gbDbLoadStep -drop -initialLoad rn6 & # logFile: var/dbload/hgwdev/logs/2014.07.10-18:27:18.rn6.dbload.log # real 68m41.070s # enable daily alignment and update of hgwdev (DONE - 2012-02-09 - Hiram) cd ~/kent/src/hg/makeDb/genbank git pull # add rn6 to: etc/align.dbs etc/hgwdev.dbs git commit -m "Added rn6 refs #13578" etc/align.dbs etc/hgwdev.dbs git push make etc-update ######################################################################### # setup search rule for assembly track (DONE - 2014-07-08 - Hiram) export maxLen=`hgsql -N -e 'select frag from gold;' rn6 | awk '{print length($0)}' | sort -run | head -1` echo $maxLen # 14 export C=1 while [ $C -le $maxLen ]; do echo -n " $C: " hgsql -N -e 'select frag from gold;' rn6 | sort -u \ | awk '{ print substr($0,'$C',1) }' | sort -u | xargs echo | sed -e 's/ //g' C=`echo $C | awk '{print $1+1}'` done # 1: A # 2: ACY # 3: 012B # 4: 012345789R # 5: 0123456789 # 6: 0123456789 # 7: 0123456789 # 8: 0123456789 # 9: .0123456789 # 10: 0123456789 # 11: 0123456789 # 12: 0123456789 # 13: . # 14: 1 searchTable gold searchMethod prefix searchType bed shortCircuit 1 termRegex A[ACY][012B][0-9R][0-9]+(\.1)* query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 # test pattern: hgsql -N -e 'select frag from gold;' rn6 | wc -l # 76195 hgsql -N -e 'select frag from gold;' rn6 \ | egrep -e 'A[ACY][B012][R0-9][0-9]+(\.[0-9]+)*' | wc -l # 76195 hgsql -N -e 'select frag from gold;' rn6 | sed -e 's/\.[0-9]+$//' \ | egrep -e 'A[ACY][B012][R0-9][0-9]+(\.[0-9]+)*' | wc -l # 76195 hgsql -N -e 'select frag from gold;' rn6 \ | egrep -v -e 'A[ACY][B012][R0-9][0-9]+(\.[0-9]+)*' | wc -l # 0 ############################################################################## # construct liftOver from rn5 to rn6 (DONE - 2014-07-14 - Hiram) # documentation for this step is in rn5.txt ######################################################################### ## Default position to same as rn5 found via lift over data n rn5 ## (DONE - 2014-07-14 - Hiram) ssh hgwdev hgsql -e 'update dbDb set defaultPos="chr1:80608553-80639261" where name="rn6";' hgcentraltest ############################################################################## # construct downloads files (DONE - 2014-07-14 - Hiram) # before starting downloads, the joinerCheck should be clean # after rn6 is added to all.joiner: joinerCheck -keys -database=rn6 all.joiner cd /hive/data/genomes/rn6 makeDownloads.pl -dbHost=hgwdev -workhorse=hgwdev rn6 \ > downloads.log 2>&1 # real 28m50.409s # examine the goldenPath/*/README.txt files to verify the text ############################################################################## # ready for first pushQ entry (DONE - 2014-07-14 - Hiram) mkdir /hive/data/genomes/rn6/pushQ cd /hive/data/genomes/rn6/pushQ makePushQSql.pl rn6 > rn6.sql 2> stderr.out # real 1m52.294s # some errors are legitimate and OK: head stderr.out # WARNING: hgwdev does not have /gbdb/rn6/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/rn6/wib/quality.wib # WARNING: hgwdev does not have /gbdb/rn6/bbi/qualityBw/quality.bw # WARNING: rn6 does not have seq # WARNING: rn6 does not have extFile scp -p rn6.sql qateam@hgwbeta:/tmp ssh qateam@hgwbeta './bin/x86_64/hgsql qapushq < /tmp/rn6.sql' ############################################################################## ## blat server turned on (DONE - 2014-01-13 - 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 ("rn6", "blat4b", "17784", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("rn6", "blat4b", "17785", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################ # LIFTOVER TO Rn5 (DONE - 2014-07-14 - Hiram ) mkdir /hive/data/genomes/rn6/bed/blat.rn5.2014-07-25 cd /hive/data/genomes/rn6/bed/blat.rn5.2014-07-25 # -debug run to create run dir, preview scripts... doSameSpeciesLiftOver.pl -ooc=/hive/data/genomes/rn6/jkStuff/rn6.11.ooc \ -debug rn6 rn5 # Real run: time doSameSpeciesLiftOver.pl \ -bigClusterHub=ku -workhorse=hgwdev -dbHost=hgwdev -buildDir=`pwd` \ -ooc=/hive/data/genomes/rn6/jkStuff/rn6.11.ooc rn6 rn5 > do.log 2>&1 # real 73m17.426s # test with sequence ############################################################################# # UCSC to RefSeq name correspondence (DONE - 2015-04-15 - Hiram) mkdir /hive/data/genomes/rn6/bed/ucscToRefSeq cd /hive/data/genomes/rn6/bed/ucscToRefSeq rsync -avPL \ rsync://ftp.ncbi.nlm.nih.gov/genomes/genbank/vertebrate_mammalian/Rattus_norvegicus/all_assembly_versions/GCA_000001895.4_Rnor_6.0/GCA_000001895.4_Rnor_6.0_assembly_report.txt ./ # this assembly_report has "UCSC-style-name" in column 10 # but it does not name everything # duplicate contigs removed: 1 AABR07023006.1 1 AABR07022518.1 # columns 5 and 7 are the INSDC and RefSeq names # eliminate the two duplicate contigs grep -v "^#" GCA_000001895.4_Rnor_6.0_assembly_report.txt \ | egrep -v "AABR07023006.1|AABR07022518.1" \ | awk -F'\t' '{printf "%s\t%s\n", $5,$7}' | sort > insdc.refSeq.tab hgsql -N -e 'select name,chrom,chromStart,chromEnd from ucscToINSDC;' \ rn6 | sort > insdc.ucsc.tab join insdc.ucsc.tab insdc.refSeq.tab | tr '[ ]' '[\t]' \ | cut -f2- > ucsc.refSeq.tab # when working perfectly, all these tab files have the same line count: wc -l *.tab # 953 insdc.refSeq.tab # 953 insdc.ucsc.tab # 953 ucsc.refSeq.tab export chrSize=`cut -f1 ucsc.refSeq.tab | awk '{print length($0)}' | sort -n | tail -1` sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | sed -e 's/INSDC/RefSeq/g;' > ucscToRefSeq.sql hgLoadSqlTab rn6 ucscToRefSeq ./ucscToRefSeq.sql ucsc.refSeq.tab checkTableCoords rn6 -table=ucscToRefSeq # should cover %100 all bases: featureBits -countGaps rn6 ucscToRefSeq # 2870182909 bases of 2870182909 (100.000%) in intersection ######################################################################### # download and load ncbiGene track db=rn6 mkdir /cluster/data/genomes/$db/bed/ncbiGene cd /cluster/data/genomes/$db/bed/ncbiGene ftpFile=ftp://ftp.ncbi.nlm.nih.gov/genomes/R_norvegicus/GFF/ref_Rnor_6.0_top_level.gff3.gz gff3File=`basename $ftpFile` echo "select * from ucscToRefSeq" | hgsql $db | tail -n +2 | awk '{print 0, $4, $3, $1, $3}' > refSeqToUcsc.lft rm -f $gff3File wget $ftpFile /cluster/home/braney/bin/x86_64/gff3ToGenePred -useName -warnAndContinue -attrsOut=attrs -bad=bad.gp $gff3File stdout 2> convertErr.txt | liftUp -type=.gp -extGenePred lift.gp refSeqToUcsc.lft warn stdin 2> liftErr.txt wc -l lift.gp # 64457 lift.gp wc -l bad.gp # 0 tawk '{print $1}' attrs | sort | uniq > meta wc -l meta # 63278 meta for i in product Dbxref gene gbkey do echo $i tawk -v attr=$i '$2==attr {print $1,$3}' attrs | sort | uniq | join -t $'\t' /dev/stdin meta > out mv out meta done wc -l meta # 62775 meta cat curated.gp predicted.gp | awk '{print $1}' | sort -u > tmp1 cat meta | awk '{print $1}' | sort -u > tmp2 join -v 1 tmp1 tmp2 | wc -l # 0 egrep "^N(M|R)|^YP" lift.gp > curated.gp egrep "^X(M|R)" lift.gp > predicted.gp wc -l curated.gp predicted.gp # 17961 curated.gp # 43265 predicted.gp # 61226 total grep dropping convertErr.txt | wc -l # 0 awk '/isn/ {print $1}' liftErr.txt | sort -u # nothing hgLoadGenePred -genePredExt $db ncbiRefCurated curated.gp hgLoadGenePred -genePredExt $db ncbiRefPredicted predicted.gp hgLoadSqlTab $db ncbiRefLink $kent/src/hg/lib/ncbiRefLink.sql meta hgsql -e 'INSERT INTO trackVersion \ (db, name, who, version, updateTime, comment, source, dateReference) VALUES("rn6", "ncbiRefSeq", "braney", "105", now(), "http://www.ncbi.nlm.nih.gov/genome/annotation_euk/Rattus_norvegicus/105/", "ftp://ftp.ncbi.nlm.nih.gov/genomes/R_norvegicus", "7 August 2014" );' hgFixed ############################################################################ # SWAP LASTZ human/hg19 vs. rat/rn6 - (DONE - 2015-06-08 - Hiram) # the original alignment cd /hive/data/genomes/hg19/bed/lastzRn6.2015-06-08 cat fb.hg19.chainRn6Link.txt # 924289029 bases of 2897316137 (31.902%) in intersection # and for the swap: mkdir /hive/data/genomes/rn6/bed/blastz.hg19.swap cd /hive/data/genomes/rn6/bed/blastz.hg19.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/hg19/bed/lastzRn6.2015-06-08/DEF \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -syntenicNet) > swap.log 2>&1 # real 77m36.480s cat fb.rn6.chainHg19Link.txt # 949880616 bases of 2729860805 (34.796%) in intersection time (doRecipBest.pl -buildDir=`pwd` rn6 hg19) > rbest.log 2>&1 # real 32m19.569s ######################################################################### # GENEID GENE PREDICTIONS (DONE - 2015-06-26 - Hiram) ssh hgwdev mkdir /hive/data/genomes/rn6/bed/geneid cd /hive/data/genomes/rn6/bed/geneid wget --timestamping \ http://genome.crg.es/genepredictions/R.norvegicus/rn6/geneid_v1.4/rn6.geneid.prot wget --timestamping \ http://genome.crg.es/genepredictions/R.norvegicus/rn6/geneid_v1.4/rn6.geneid.gtf ldHgGene -gtf -genePredExt rn6 geneid rn6.geneid.gtf # Read 41652 transcripts in 307250 lines in 1 files # 41652 groups 883 seqs 1 sources 3 feature types # 41652 gene predictions featureBits -countGaps rn6 geneid # 42028722 bases of 2730871774 (1.539%) in intersection featureBits -countGaps rn4 geneid # 40255245 bases of 2834127293 (1.420%) in intersection ########################################################################## # SGP GENES (DONE - 2015-07-30 - Hiram) mkdir /hive/data/genomes/rn6/bed/sgpGene cd /hive/data/genomes/rn6/bed/sgpGene wget --timestamping \ http://genome.crg.es/genepredictions/R.norvegicus/rn6/SGP2/hg38/00README wget --timestamping \ http://genome.crg.es/genepredictions/R.norvegicus/rn6/SGP2/hg38/rn6.sgp2.gtf wget --timestamping \ http://genome.crg.es/genepredictions/R.norvegicus/rn6/SGP2/hg38/rn6.sgp2.gff3 ldHgGene -gtf -genePredExt rn6 sgpGene rn6.sgp2.gtf # Read 37981 transcripts in 306550 lines in 1 files # 37981 groups 468 seqs 1 sources 3 feature types # 37981 gene predictions featureBits -enrichment rn6 refGene:CDS sgpGene # refGene:CDS 0.963%, sgpGene 1.399%, both 0.821%, cover 85.27%, enrich 60.93x ########################################################################### # lastz zebrafish danRer10 (DONE - 2015-09-18,14 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S rn6DanRer10 mkdir /hive/data/genomes/rn6/bed/lastzDanRer10.2015-09-18 cd /hive/data/genomes/rn6/bed/lastzDanRer10.2015-09-18 printf "%s\n" \ '# Rat vs. zebrafish BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz BLASTZ_Y=3400 BLASTZ_L=6000 BLASTZ_K=2200 BLASTZ_Q=/scratch/data/blastz/HoxD55.q # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: zebrafish danRer10 SEQ2_DIR=/hive/data/genomes/danRer10/danRer10.2bit SEQ2_LEN=/hive/data/genomes/danRer10/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=100 BASE=/hive/data/genomes/rn6/bed/lastzDanRer10.2015-09-18 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose) > do.log 2>&1 # real 272m42.068s cat fb.rn6.chainDanRer10Link.txt # 88237547 bases of 2729860805 (3.232%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 0m50.415s time (doRecipBest.pl -buildDir=`pwd` rn6 danRer10) > rbest.log 2>&1 & # real 7m47.791s # and for the swap mkdir /hive/data/genomes/danRer10/bed/blastz.rn6.swap cd /hive/data/genomes/danRer10/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzDanRer10.2015-09-18/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 # real 20m40.003s cat fb.danRer10.chainRn6Link.txt # 79649796 bases of 1369683683 (5.815%) in intersection time (doRecipBest.pl -buildDir=`pwd` danRer10 rn6) > rbest.log 2>&1 # real 7m59.071s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzDanRer10.2015-09-18/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 0m41.726s ######################################################################### # cloneEnds (DONE - 2016-05-26 - Hiram) mkdir /hive/data/genomes/rn6/bed/cloneEnds cd /hive/data/genomes/rn6/bed/cloneEnds # fetch the NCBI INSDC name correspondence file: rsync -L -a -P rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Rattus_norvegicus/all_assembly_versions/GCF_000001895.5_Rnor_6.0/GCF_000001895.5_Rnor_6.0_assembly_report.txt ./ rsync -L -a -P rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Rattus_norvegicus/all_assembly_versions/GCF_000001895.5_Rnor_6.0/GCF_000001895.5_Rnor_6.0_genomic.fna.gz ./ faToTwoBit GCF_000001895.5_Rnor_6.0_genomic.fna.gz \ GCF_000001895.5_Rnor_6.0.2bit twoBitDup -keyList=stdout GCF_000001895.5_Rnor_6.0.2bit \ | sort > GCF_000001895.5_Rnor_6.0.idKeys.txt # fetch the clone reports mkdir reports rsync -a -P \ rsync://ftp.ncbi.nih.gov/repository/clone/reports/Rattus_norvegicus/*.GCF_000001895.5.102.*.gff \ ./reports/ # this procedure to correspond the names is a work in progress # turns out the RefSeq assembly has three identical contigs, two dups # to be removed, the same ones removed for rn6 build:: cat remove.list NW_007906462.1 NW_007906649.1 faSomeRecords -exclude GCF_000001895.5_Rnor_6.0_genomic.fna.gz \ remove.list stdout | gzip -c > noDups.GCF_000001895.5_Rnor_6.0.fa.gz faToTwoBit noDups.GCF_000001895.5_Rnor_6.0.fa.gz \ noDups.GCF_000001895.5_Rnor_6.0.2bit twoBitDup noDups.GCF_000001895.5_Rnor_6.0.2bit $HOME/kent/src/utils/twoBitDup/twoBitDup -keyList=stdout \ ../../rn6.2bit | sort -k1,1 > rn6.idKeys.txt $HOME/kent/src/utils/twoBitDup/twoBitDup -keyList=stdout \ noDups.GCF_000001895.5_Rnor_6.0.2bit | sort -k1,1 \ > GCF_000001895.5_Rnor_6.0.idKeys.txt join GCF_000001895.5_Rnor_6.0.idKeys.txt rn6.idKeys.txt \ | tr '[ ]' '[\t]' | cut -f2- > refSeq.ucscName.tab # establish full library list: ls reports/*.GCF_000001895.5.102.*.gff | sed -e 's#reports/##' \ | cut -d"." -f1 | sort -u > library.list.txt # only four libraries: cat library.list.txt CH230 RNB1 RNB2 RP32 # most unfortunate, the GFF files use a secondary accession identifier, # they do not use the identifier used in the assembly itself. cd /hive/data/genomes/rn6/bed/cloneEnds/reports grep -h -v "^#" *.gff | cut -f1 | sed -e 's/gpp.//; s/|$//;' \ | sed '/^$/d' | sort -u | sort > ../gpc.gps.identifiers.txt cd /hive/data/genomes/rn6/bed/cloneEnds # there are only 75 identifiers: wc -l gpc.gps.identifiers.txt # 75 gpc.gps.identifiers.txt # fetch corresponding identifiers from Entrez: export PATH=/hive/data/outside/ncbi/EDirect/edirect:$PATH export EMAIL="hiram@soe.ucsc.edu" for ID in `cat gpc.gps.identifiers.txt` do esearch -db nucleotide -query "${ID}" | efetch -format xml \ | xtract -pattern Seq-id_other -element Textseq-id_accession,Textseq-id_version | awk -v id=$ID '{printf "%s\t%s.%s\n", id, $1,$2}' done > gpcGps.to.INSDC.tab # real 3m30.775s # verify have all: wc -l gpcGps.to.INSDC.tab gpc.gps.identifiers.txt # 75 gpcGps.to.INSDC.tab # 75 gpc.gps.identifiers.txt # rework that refSeq.ucscName.tab file: join GCF_000001895.5_Rnor_6.0.idKeys.txt rn6.idKeys.txt \ | tr '[ ]' '[\t]' | cut -f2- > INSDC.ucscName.tab join -1 2 <(sort -k2 gpcGps.to.INSDC.tab) <(sort INSDC.ucscName.tab) \ | tr '[ ]' '[\t]' | cut -f2- > refSeq.ucscName.tab # a script to scan the GFF files, with the refSeq.ucscName.tab # name correspondence to construct bed files printf '#!/usr/bin/env perl use strict; use warnings; my $argc = scalar(@ARGV); if ($argc < 1) { printf STDERR "usage: ./rn6.pl <report.gff> [moreReports.gff]\n"; exit 255; } my %%refSeqToUcsc; # key is refSeq name, value is UCSC chrom name open (FH, "<refSeq.ucscName.tab") or die "can not read refSeq.ucscName.tab"; while (my $line = <FH>) { chomp $line; my ($refSeq, $ucsc) = split("\t", $line); $refSeqToUcsc{$refSeq} = $ucsc; } close (FH); my %%chromSizes; # key is UCSC chrom name, key is chrom size open (FH, "</hive/data/genomes/rn6/chrom.sizes") or die "can not read rn6/chrom.sizes"; while (my $line = <FH>) { chomp $line; my ($chr, $size) = split("\t", $line); $chromSizes{$chr} = $size; } close (FH); while (my $file = shift) { my %%starts; # key is parent ID, value is start end coordinates start,end my %%ends; # key is parent ID, value is end end coordinates start,end my %%parents; # key is parent ID, value is 1 to signify exists my %%endNames; # key is parent ID, value is the Name of the parent clone_insert printf STDERR "# processing $file\n"; open (FH, "<$file") or die "can not read $file"; while (my $line = <FH>) { chomp $line; next if ($line=~ m/^#/); my @a = split("\t", $line); next if (scalar(@a) < 1); my $contig = $a[0]; $contig =~ s/gpp//; $contig =~ s/\|/g/; my $ucscChr = $refSeqToUcsc{$contig}; if (!defined($ucscChr)) { printf STDERR "# ERR: contig not in refSeqToUcsc: %%s\\n", $contig; next; } next if (! exists($chromSizes{$ucscChr})); my $chromSize = $chromSizes{$ucscChr}; my $chromStart = $a[3] - 1; my $chromEnd = $a[4]; if ($chromStart > $chromSize) { printf STDERR "# warning chromStart over size $ucscChr $chromStart $chromEnd\\n"; $chromStart = $chromSize-1; } if ($chromEnd > $chromSize) { my $overRun = $chromEnd - $chromSize; printf STDERR "# warning chromEnd over size by $overRun -> $ucscChr $chromStart $chromEnd\\n"; $chromEnd = $chromSize; } my $id="notFound"; my $name="notFound"; my $parent="notFound"; my @b = split(";", $a[8]); for (my $i = 0; $i < scalar(@b); ++$i) { my ($tag, $value) = split("=", $b[$i]); if ($tag eq "ID") { $id = $value; if ($id !~ m/-/) { if (exists($parents{$id})) { printf STDERR "# WARN: duplicate parent: $id"; } else { $parents{$id} = $ucscChr; } } } elsif ($tag eq "Parent") { $parent = $value; } elsif ($tag eq "Name") { $name = $value; } } my $type="notFound"; my $insertType = $a[2]; if ($insertType =~ m/clone_insert_start/) { $type = "start"; if ($parent eq "notFound") { printf STDERR "# ERR: can not find parent for start $name Ttype $id\\n"; } else { if (!exists($parents{$parent})) { printf STDERR "# ERR: start found $name with no parent $parent declared\\n"; } elsif (exists($starts{$parent})) { printf STDERR "# ERR: duplicate start for $parent\\n"; } elsif ($ucscChr eq $parents{$parent}) { $starts{$parent} = sprintf("%%s\t%%s", $chromStart, $chromEnd); } else { printf STDERR "# ERR: start on different chrom $ucscChr than parent $parent $parents{$parent}\\n"; } } } elsif ($insertType =~ m/clone_insert_end/) { $type = "end"; if ($parent eq "notFound") { printf STDERR "# ERR: can not find parent for end $name Ttype $id\\n"; } else { if (!exists($parents{$parent})) { printf STDERR "# ERR: end found $name with no parent $parent declared\\n"; } elsif (exists($ends{$parent})) { printf STDERR "# ERR: duplicate end for $parent\\n"; } elsif ($ucscChr eq $parents{$parent}) { $ends{$parent} = sprintf("%%s\t%%s", $chromStart, $chromEnd); } else { printf STDERR "# ERR: end on different chrom $ucscChr than parent $parent $parents{$parent}\\n"; } } } elsif ($insertType =~ m/clone_insert/) { $type = "insert"; $endNames{$id} = $name; } $name =~ s/gi\|//g; $id =~ s/gi\|//g; printf STDERR "%%s\t%%d\t%%d\t%%s_%%s_%%s\t0\t%%s\\n", $ucscChr, $chromStart, $chromEnd, $name, $type, $id, $a[6]; } # while (my $line = <FH>) close (FH); foreach my $parent (keys %%parents) { if (! exists($starts{$parent}) ) { printf STDERR "# ERR: no start for $parent\\n"; } elsif (! exists($ends{$parent}) ) { printf STDERR "# ERR: no end for $parent\\n"; } else { my $strand = "+"; my $chrStart = 0; my $chrEnd = 0; my $blockStart = 0; my ($sStart, $sEnd) = split("\t", $starts{$parent}); my ($eStart, $eEnd) = split("\t", $ends{$parent}); my $startSize = $sEnd - $sStart; my $endSize = $eEnd - $eStart; if ($eStart < $sStart) { $chrStart = $eStart; $chrEnd = $sEnd; $blockStart = $sStart - $chrStart; $strand = "-"; $startSize = $eEnd - $eStart; $endSize = $sEnd - $sStart; } else { $chrStart = $sStart; $chrEnd = $eEnd; $blockStart = $eStart - $chrStart; } if ($startSize > $blockStart) { printf STDERR "# startSize > blockStart $endNames{$parent}\\n"; } else { printf "%%s\t%%d\t%%d\t%%s\t0\t%%s\t%%d\t%%d\t0\t2\t%%d,%%d\t0,%%d\\n", $parents{$parent}, $chrStart, $chrEnd, $endNames{$parent}, $strand, $chrStart, $chrEnd, $startSize, $endSize, $blockStart; } } } } ' > rn6.pl chmod +x rn6.pl # process GFF files into bed files into separateLibs/ directory for L in `cat library.list.txt` do export destDir="separateLibs/${L}" echo "working: ${L}" 1>&1 mkdir -p "${destDir}" ./rn6.pl reports/${L}.GCF_000001895.5.102.*.gff \ 2> ${destDir}/tmp.bed6 | sort -k1,1 -k2,2n > ${destDir}/rn6.${L}.bed sort -k1,1 -k2,2n ${destDir}/tmp.bed6 > ${destDir}/rn6.${L}.items.bed6 done # for human clone ends we used only the libraries with more than # 20,000 clone ends. Only lib RP32 here has a small number: 106 # so use them all # use only those libraries with more than 20,000 clone ends wc -l separateLibs/*/*.bed | sort -n | grep -v total | awk '$1 > 100' \ | sed -e 's#.*separateLibs/##; s#/.*##' > libs.over100.list # filter out bad ends, length must be <= median size times three cat libs.over100.list | while read D do if [ ! -s separateLibs/${D}/lengths.txt ]; then awk '{print $3-$2}' separateLibs/${D}/rn6.${D}.bed \ > separateLibs/${D}/lengths.txt fi median3X=`ave separateLibs/${D}/lengths.txt | grep median | awk '{printf "%d", $2*3}'` awk '($3-$2) < '$median3X'' separateLibs/${D}/rn6.${D}.bed > separateLibs/${D}/rn6.median3X.bed awk '($3-$2) >= '$median3X'' separateLibs/${D}/rn6.${D}.bed > separateLibs/${D}/rn6.badMap.bed before=`cat separateLibs/${D}/rn6.${D}.bed | wc -l` after=`cat separateLibs/${D}/rn6.median3X.bed | wc -l` dropped=`echo $before $after | awk '{print $1-$2}'` perCent=`echo $dropped $before | awk '{printf "%.2f", 100*'$dropped/$before'}'` printf "#\t$D $before - $after = $dropped -> %% $perCent dropped\n" done # RP32 106 - 105 = 1 -> % 0.94 dropped # CH230 83579 - 78516 = 5063 -> % 6.06 dropped # RNB2 96590 - 89108 = 7482 -> % 7.75 dropped # RNB1 134066 - 124987 = 9079 -> % 6.77 dropped wc -l separateLibs/*/*.median3X.bed # 78516 separateLibs/CH230/rn6.median3X.bed # 124987 separateLibs/RNB1/rn6.median3X.bed # 89108 separateLibs/RNB2/rn6.median3X.bed # 105 separateLibs/RP32/rn6.median3X.bed # 292716 total # loading the median3X files for L in `cat libs.over100.list` do echo $L 1>&2 hgLoadBed -type=bed12 rn6 cloneEnd${L} \ separateLibs/${L}/rn6.median3X.bed \ > separateLibs/loadBed.${L}.log 2>&1 done # loading the dropped ends: mkdir /hive/data/genomes/rn6/bed/cloneEnds/droppedTooBig cd /hive/data/genomes/rn6/bed/cloneEnds/droppedTooBig # link them to here cat ../libs.over100.list | while read L do ln -s ../separateLibs/${L}/rn6.badMap.bed ${L}.badMap.bed done wc -l *.bed # 5063 CH230.badMap.bed # 9079 RNB1.badMap.bed # 7482 RNB2.badMap.bed # 1 RP32.badMap.bed # 21625 total # then load hgLoadBed -type=bed12 rn6 cloneEndbadEnds *.badMap.bed # Read 21625 elements of size 12 from CH230.badMap.bed # construct multiple mapped ends: cd /hive/data/genomes/rn6/bed/cloneEnds for L in `cat libs.over100.list` do cat separateLibs/${L}/rn6.median3X.bed done | sort -k4 > allEnds.bed wc -l allEnds.bed # 292716 allEnds.bed cut -f4 allEnds.bed | sort | uniq -c | sort -rn > allEnds.names.count.txt awk '$1 > 1' allEnds.names.count.txt | awk '{print $2}' \ | sort > multiples.names.txt join -o "2.1,2.2,2.3,2.4,2.5,2.6,2.7,2.8,2.9,2.10,2.11,2.12" \ -2 4 multiples.names.txt allEnds.bed | tr '[ ]' '[\t]' \ | sort -k1,1 -k2,2n > allEnds.multiple.locations.bed hgLoadBed -type=bed12 rn6 cloneEndmultipleMaps \ allEnds.multiple.locations.bed > load.multipleMaps.log 2>&1 awk '$6 == "+"' allEnds.bed | sort -k1,1 -k2,2n \ | bedItemOverlapCount rn6 stdin > allEnds.forward.bedGraph awk '$6 == "-"' allEnds.bed | sort -k1,1 -k2,2n \ | bedItemOverlapCount rn6 stdin > allEnds.reverse.bedGraph wc -l *.bedGraph # 272228 allEnds.forward.bedGraph # 270927 allEnds.reverse.bedGraph # 543155 total bedGraphToBigWig allEnds.forward.bedGraph \ /hive/data/genomes/rn6/chrom.sizes \ cloneEndcoverageForward.bw bedGraphToBigWig allEnds.reverse.bedGraph \ /hive/data/genomes/rn6/chrom.sizes \ cloneEndcoverageReverse.bw bigWigInfo cloneEndcoverageForward.bw version: 4 isCompressed: yes isSwapped: 0 primaryDataSize: 1,359,215 primaryIndexSize: 21,052 zoomLevels: 7 chromCount: 69 basesCovered: 2,781,456,290 mean: 8.347160 min: 1.000000 max: 141.000000 std: 4.341902 bigWigInfo cloneEndcoverageReverse.bw version: 4 isCompressed: yes isSwapped: 0 primaryDataSize: 1,353,539 primaryIndexSize: 20,988 zoomLevels: 7 chromCount: 62 basesCovered: 2,781,413,397 mean: 8.322005 min: 1.000000 max: 168.000000 std: 4.457940 mkdir /gbdb/rn6/bbi/cloneEnd ln -s `pwd`/cloneEndcoverageForward.bw /gbdb/rn6/bbi/cloneEnd ln -s `pwd`/cloneEndcoverageReverse.bw /gbdb/rn6/bbi/cloneEnd hgBbiDbLink rn6 cloneEndcoverageForward \ /gbdb/rn6/bbi/cloneEnd/cloneEndcoverageForward.bw hgBbiDbLink rn6 cloneEndcoverageReverse \ /gbdb/rn6/bbi/cloneEnd/cloneEndcoverageReverse.bw ### Fixup the scores to indicate how many multiple mappings as mentioned ### in the hg19 bacEnds description page: one mapping: score = 1000 ### multiple mappings: score = 1500/count ### the sort | uniq -c | awk does this score calculation with the name ### in column 1 ### The join puts the existing table together with those scores ### DONE - 2015-06-18 - Hiram mkdir /hive/data/genomes/rn6/bed/cloneEnds/addCounts cd /hive/data/genomes/rn6/bed/cloneEnds/addCounts mkdir score withScore noScore for table in cloneEndCH230 cloneEndRNB1 cloneEndRNB2 cloneEndRP32 \ cloneEndbadEnds cloneEndmultipleMaps do hgsql -N -e "select name from $table;" rn6 | sort | uniq -c | awk '{ if (1 == $1) {printf "%s\t1000\n", $2} else {printf "%s\t%d\n", $2, 1500/$1} }' \ | sort > score/rn6.$table.score.tab hgsql -N -e "select * from $table order by name;" rn6 \ | sort -k5 > noScore/rn6.$table.tab join -1 5 noScore/rn6.$table.tab score/rn6.$table.score.tab \ | tr '[ ]' '[\t]' | awk '{printf "%d\t%s\t%d\t%d\t%s\t%d\t%s\t%d\t%d\t%d\t%d\t%s\t%s\n", $2,$3,$4,$5,$1,$14,$7,$8,$9,$10,$11,$12,$13}' \ | sort -k2,2 -k3,3n > withScore/rn6.$table.withScore.tab hgsql -e "delete from $table;" rn6 hgsql -e "load data local infile \"withScore/rn6.$table.withScore.tab\" into table $table;" rn6 done ############################################################################## # ensemblToRgd (TBD - 2016-08-17 - jcasper) # We would like to have otherOrgs links from other species be able to link to # RGD pages, but this requires RGD identifiers. Long term solution involves # incorporating that data from GenBank, medium term involves building an RGD # Genes track, but short term (this) is to cook up a table that gives RGD # IDs for the Ensembl transcript IDs that we have on rn6. screen -S ensToRgd BASEDIR=/hive/data/genomes/rn6/bed/ensToRgd mkdir -p ${BASEDIR} cd ${BASEDIR} mkdir rgdDownload cd rgdDownload wget ftp://ftp.rgd.mcw.edu/pub/data_release/GENES_RAT.txt cd ${BASEDIR} grep -v '^#' rgdDownload/GENES_RAT.txt | tail -n +2 | tawk '$38 ~ /[A-Z]/ {print $38, $1}' > ensGAndRgd.tab perl -ne '($genes, $rgd) = split; foreach $gene (split /;/, $genes) {print "$gene\t$rgd\n";}' < ensGAndRgd.tab | sort > ensGToRgd.tab # Halfway there - have Ensembl gene identifier translation, but need # transcript identifier translation. NB: RGD doesn't use version # numbers for the Ensembl gene IDs. hgsql rn6 -Ne 'select substring_index(gene, ".", 1), transcript from ensGtp' | sort > ensGToEnsT.tab # Workaround for join being awkward on tabs: TAB=$(printf "\t") join -t "$TAB" ensGToEnsT.tab ensGToRgd.tab | tawk '{print $2, $3}' > ensToRgd.tab # Use a generic name/value table structure to load the data hgLoadSqlTab rn6 ensToRgd ${HOME}/kent/src/hg/lib/knownTo.sql ensToRgd.tab # Edit hgGene/hgGeneData otherOrgs.ra files for hg38 and mm10 to grab IDs from # the ensToRgd table in rn6 and use them to link to RGD. ############################################################################## # lastz X. tropicalis xenTro7 (DONE - 2017-01-11 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S rn6XenTro7 mkdir /hive/data/genomes/rn6/bed/lastzXenTro7.2017-01-11 cd /hive/data/genomes/rn6/bed/lastzXenTro7.2017-01-11 printf "%s\n" \ '# Rat vs. X. tropicalis BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz BLASTZ_Y=3400 BLASTZ_L=6000 BLASTZ_K=2200 BLASTZ_Q=/scratch/data/blastz/HoxD55.q # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: X. tropicalis xenTro7 SEQ2_DIR=/hive/data/genomes/xenTro7/xenTro7.2bit SEQ2_LEN=/hive/data/genomes/xenTro7/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=20 BASE=/hive/data/genomes/rn6/bed/lastzXenTro7.2017-01-11 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose) > do.log 2>&1 # real 1754m46.659s (much contention for ku resources) cat fb.rn6.chainXenTro7Link.txt # 112036484 bases of 2729860805 (4.104%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -continue=syntenicNet -chainMinScore=5000 -chainLinearGap=loose \ -syntenicNet) > synNet.log 2>&1 & # real 2m23.289s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 xenTro7) \ > rbest.log 2>&1 & # real 1079m9.975s # and for the swap mkdir /hive/data/genomes/xenTro7/bed/blastz.rn6.swap cd /hive/data/genomes/xenTro7/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzXenTro7.2017-01-11/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 # real 66m47.360s cat fb.xenTro7.chainRn6Link.txt # 108092395 bases of 1365936747 (7.913%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` xenTro7 rn6) \ > rbest.log 2>&1 # real 1027m44.180s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzXenTro7.2017-01-11/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 2m23.304s ############################################################################## # lastz chimp panTro5 (DONE - 2017-01-11 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S panTro5Rn6 mkdir /hive/data/genomes/rn6/bed/lastzPanTro5.2017-01-11 cd /hive/data/genomes/rn6/bed/lastzPanTro5.2017-01-11 printf "%s\n" \ '# Rat vs. chimp BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz BLASTZ_Y=3400 BLASTZ_L=6000 BLASTZ_K=2200 BLASTZ_Q=/scratch/data/blastz/HoxD55.q # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: chimp panTro5 SEQ2_DIR=/hive/data/genomes/panTro5/panTro5.2bit SEQ2_LEN=/hive/data/genomes/panTro5/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=200 BASE=/hive/data/genomes/rn6/bed/lastzPanTro5.2017-01-11 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 2703m15.450s cat fb.rn6.chainPanTro5Link.txt # 1114217604 bases of 2729860805 (40.816%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 20m31.634s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 panTro5) \ > rbest.log 2>&1 & # real 1753m33.467s # and for the swap mkdir /hive/data/genomes/panTro5/bed/blastz.rn6.swap cd /hive/data/genomes/panTro5/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzPanTro5.2017-01-11/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 185m16.607s cat fb.panTro5.chainRn6Link.txt # 1107582383 bases of 3132620660 (35.356%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` panTro5 rn6) \ > rbest.log 2>&1 # real 1411m22.158s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzPanTro5.2017-01-11/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 21m41.393s ############################################################################## # lastz opossum monDom5 (DONE - 2017-01-11 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S monDom5Rn6 mkdir /hive/data/genomes/rn6/bed/lastzMonDom5.2017-01-11 cd /hive/data/genomes/rn6/bed/lastzMonDom5.2017-01-11 printf "%s\n" \ '# Rat vs. opossum BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz BLASTZ_M=254 BLASTZ_Y=3400 BLASTZ_L=6000 BLASTZ_K=2200 BLASTZ_Q=/scratch/data/blastz/HoxD55.q # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=5 # QUERY: opossum monDom5 SEQ2_DIR=/hive/data/genomes/monDom5/monDom5.2bit SEQ2_LEN=/hive/data/genomes/monDom5/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=30 BASE=/hive/data/genomes/rn6/bed/lastzMonDom5.2017-01-11 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose) > do.log 2>&1 # real 2258m54.815s (much contention for ku resources) cat fb.rn6.chainMonDom5Link.txt # 263494827 bases of 2729860805 (9.652%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -continue=syntenicNet -chainMinScore=5000 \ -chainLinearGap=loose) > synNet.log 2>&1 & # real 4m16.551s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 monDom5) \ > rbest.log 2>&1 & # real 2596m10.639s # and for the swap mkdir /hive/data/genomes/monDom5/bed/blastz.rn6.swap cd /hive/data/genomes/monDom5/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzMonDom5.2017-01-11/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 # real 41m5.953s cat fb.monDom5.chainRn6Link.txt # 251790387 bases of 3501660299 (7.191%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` monDom5 rn6) \ > rbest.log 2>&1 # real 2521m59.900s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzMonDom5.2017-01-11/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 5m25.250s ############################################################################## # lastz rhesus rheMac8 (DONE - 2017-01-13 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S rheMac8Rn6 mkdir /hive/data/genomes/rn6/bed/lastzRheMac8.2017-01-13 cd /hive/data/genomes/rn6/bed/lastzRheMac8.2017-01-13 printf "%s\n" \ '# Rat vs. rhesus BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz BLASTZ_Y=3400 BLASTZ_L=6000 BLASTZ_K=2200 BLASTZ_Q=/scratch/data/blastz/HoxD55.q # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: rhesus rheMac8 SEQ2_DIR=/hive/data/genomes/rheMac8/rheMac8.2bit SEQ2_LEN=/hive/data/genomes/rheMac8/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=1000 BASE=/hive/data/genomes/rn6/bed/lastzRheMac8.2017-01-13 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 1196m51.891s cat fb.rn6.chainRheMac8Link.txt # 1096456425 bases of 2729860805 (40.165%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 19m21.810s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 rheMac8) \ > rbest.log 2>&1 & # real 1569m42.488s # and for the swap mkdir /hive/data/genomes/rheMac8/bed/blastz.rn6.swap cd /hive/data/genomes/rheMac8/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzRheMac8.2017-01-13/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 205m19.557s # lost swap.log cat fb.rheMac8.chainRn6Link.txt # 1051607110 bases of 3142093174 (33.468%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rheMac8 rn6) \ > rbest.log 2>&1 # real 1601m35.848s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzRheMac8.2017-01-13/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > swap.log 2>&1 & # real 19m14.673s ############################################################################## # lastz guinea pig cavPor3 (DONE - 2017-01-13 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S cavPor3Rn6 mkdir /hive/data/genomes/rn6/bed/lastzCavPor3.2017-01-13 cd /hive/data/genomes/rn6/bed/lastzCavPor3.2017-01-13 printf "%s\n" \ '# Rat vs. guinea pig BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: guinea pig cavPor3 SEQ2_DIR=/hive/data/genomes/cavPor3/cavPor3.2bit SEQ2_LEN=/hive/data/genomes/cavPor3/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=15 BASE=/hive/data/genomes/rn6/bed/lastzCavPor3.2017-01-13 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 742m21.264s cat fb.rn6.chainCavPor3Link.txt # 759825883 bases of 2729860805 (27.834%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 12m11.706s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 cavPor3) \ > rbest.log 2>&1 & # real 535m40.395s # and for the swap mkdir /hive/data/genomes/cavPor3/bed/blastz.rn6.swap cd /hive/data/genomes/cavPor3/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzCavPor3.2017-01-13/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 54m21.781s cat fb.cavPor3.chainRn6Link.txt # 755393931 bases of 2663369733 (28.362%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` cavPor3 rn6) \ > rbest.log 2>&1 # real 564m23.551s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzCavPor3.2017-01-13/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 12m6.911s ############################################################################## # lastz dog canFam3 (DONE - 2017-01-13 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S canFam3Rn6 mkdir /hive/data/genomes/rn6/bed/lastzCanFam3.2017-01-13 cd /hive/data/genomes/rn6/bed/lastzCanFam3.2017-01-13 printf "%s\n" \ '# Rat vs. dog BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: dog canFam3 SEQ2_DIR=/hive/data/genomes/canFam3/canFam3.2bit SEQ2_LEN=/hive/data/genomes/canFam3/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=15 BASE=/hive/data/genomes/rn6/bed/lastzCanFam3.2017-01-13 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 694m15.665s cat fb.rn6.chainCanFam3Link.txt # 779099966 bases of 2729860805 (28.540%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 12m27.782s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 canFam3) \ > rbest.log 2>&1 & # real 471m35.128s # and for the swap mkdir /hive/data/genomes/canFam3/bed/blastz.rn6.swap cd /hive/data/genomes/canFam3/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzCanFam3.2017-01-13/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 47m17.732s cat fb.canFam3.chainRn6Link.txt # 738483963 bases of 2392715236 (30.864%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` canFam3 rn6) \ > rbest.log 2>&1 # real 357m54.846s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzCanFam3.2017-01-13/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 12m37.764s ############################################################################## # lastz cow bosTau8 (DONE - 2017-01-16 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S bosTau8Rn6 mkdir /hive/data/genomes/rn6/bed/lastzBosTau8.2017-01-16 cd /hive/data/genomes/rn6/bed/lastzBosTau8.2017-01-16 printf "%s\n" \ '# Rat vs. cow BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: cow bosTau8 SEQ2_DIR=/hive/data/genomes/bosTau8/bosTau8.2bit SEQ2_LEN=/hive/data/genomes/bosTau8/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=15 BASE=/hive/data/genomes/rn6/bed/lastzBosTau8.2017-01-16 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 379m56.864s cat fb.rn6.chainBosTau8Link.txt # 707100078 bases of 2729860805 (25.902%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 12m37.764s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 bosTau8) \ > rbest.log 2>&1 & # real 688m13.841s # and for the swap mkdir /hive/data/genomes/bosTau8/bed/blastz.rn6.swap cd /hive/data/genomes/bosTau8/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzBosTau8.2017-01-16/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 53m51.901s cat fb.bosTau8.chainRn6Link.txt # 671724400 bases of 2649307237 (25.355%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` bosTau8 rn6) \ > rbest.log 2>&1 # real 580m38.055s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzBosTau8.2017-01-16/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 11m19.902s ############################################################################## # lastz panda ailMel1 (DONE - 2017-01-16 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S ailMel1Rn6 mkdir /hive/data/genomes/rn6/bed/lastzAilMel1.2017-01-16 cd /hive/data/genomes/rn6/bed/lastzAilMel1.2017-01-16 printf "%s\n" \ '# Rat vs. panda BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: panda ailMel1 SEQ2_DIR=/hive/data/genomes/ailMel1/ailMel1.2bit SEQ2_LEN=/hive/data/genomes/ailMel1/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=400 BASE=/hive/data/genomes/rn6/bed/lastzAilMel1.2017-01-16 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 309m30.719s cat fb.rn6.chainAilMel1Link.txt # 829980217 bases of 2729860805 (30.404%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 12m19.496s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 ailMel1) \ > rbest.log 2>&1 & # real 405m31.270s # and for the swap mkdir /hive/data/genomes/ailMel1/bed/blastz.rn6.swap cd /hive/data/genomes/ailMel1/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzAilMel1.2017-01-16/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 53m19.878s cat fb.ailMel1.chainRn6Link.txt # 780846350 bases of 2245312831 (34.777%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` ailMel1 rn6) \ > rbest.log 2>&1 # real 324m22.903s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzAilMel1.2017-01-16/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 14m19.950s ############################################################################## # lastz prairie vole micOch1 (DONE - 2017-01-17 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S micOch1Rn6 mkdir /hive/data/genomes/rn6/bed/lastzMicOch1.2017-01-17 cd /hive/data/genomes/rn6/bed/lastzMicOch1.2017-01-17 printf "%s\n" \ '# Rat vs. prairie vole BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: prairie vole micOch1 SEQ2_DIR=/hive/data/genomes/micOch1/micOch1.2bit SEQ2_LEN=/hive/data/genomes/micOch1/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=20 BASE=/hive/data/genomes/rn6/bed/lastzMicOch1.2017-01-17 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 1009m6.065s cat fb.rn6.chainMicOch1Link.txt # 1561627848 bases of 2729860805 (57.205%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 26m19.328s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 micOch1) \ > rbest.log 2>&1 & # real 1213m36.203s # and for the swap mkdir /hive/data/genomes/micOch1/bed/blastz.rn6.swap cd /hive/data/genomes/micOch1/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzMicOch1.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 138m3.307s cat fb.micOch1.chainRn6Link.txt # 1475851323 bases of 2104321675 (70.134%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` micOch1 rn6) \ > rbest.log 2>&1 # real 1096m12.914s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzMicOch1.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 24m30.723s ############################################################################## # lastz rabbit oryCun2 (DONE - 2017-01-17 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S oryCun2Rn6 mkdir /hive/data/genomes/rn6/bed/lastzOryCun2.2017-01-17 cd /hive/data/genomes/rn6/bed/lastzOryCun2.2017-01-17 printf "%s\n" \ '# Rat vs. rabbit BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: rabbit oryCun2 SEQ2_DIR=/hive/data/genomes/oryCun2/oryCun2.2bit SEQ2_LEN=/hive/data/genomes/oryCun2/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=20 BASE=/hive/data/genomes/rn6/bed/lastzOryCun2.2017-01-17 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 722m23.231s cat fb.rn6.chainOryCun2Link.txt # 673020370 bases of 2729860805 (24.654%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 10m40.444s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 oryCun2) \ > rbest.log 2>&1 & # real 701m28.973s # and for the swap mkdir /hive/data/genomes/oryCun2/bed/blastz.rn6.swap cd /hive/data/genomes/oryCun2/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzOryCun2.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 45m34.526s cat fb.oryCun2.chainRn6Link.txt # 650509122 bases of 2604023284 (24.981%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` oryCun2 rn6) \ > rbest.log 2>&1 # real 641m54.166s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzOryCun2.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 10m40.444s ############################################################################## # lastz cat felCat8 (DONE - 2017-01-17 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S felCat8Rn6 mkdir /hive/data/genomes/rn6/bed/lastzFelCat8.2017-01-17 cd /hive/data/genomes/rn6/bed/lastzFelCat8.2017-01-17 printf "%s\n" \ '# Rat vs. cat BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: cat felCat8 SEQ2_DIR=/hive/data/genomes/felCat8/felCat8.2bit SEQ2_LEN=/hive/data/genomes/felCat8/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=1000 BASE=/hive/data/genomes/rn6/bed/lastzFelCat8.2017-01-17 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 830m52.261s cat fb.rn6.chainFelCat8Link.txt # 805848215 bases of 2729860805 (29.520%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 13m49.129s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 felCat8) \ > rbest.log 2>&1 & # real 648m55.901s # and for the swap mkdir /hive/data/genomes/felCat8/bed/blastz.rn6.swap cd /hive/data/genomes/felCat8/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzFelCat8.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 112m39.614s cat fb.felCat8.chainRn6Link.txt # 771419833 bases of 2599716822 (29.673%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` felCat8 rn6) \ > rbest.log 2>&1 # real 577m31.945s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzFelCat8.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 13m51.296s ############################################################################## # lastz turkey melGal5 (DONE - 2017-01-17 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S melGal5Rn6 mkdir /hive/data/genomes/rn6/bed/lastzMelGal5.2017-01-17 cd /hive/data/genomes/rn6/bed/lastzMelGal5.2017-01-17 printf "%s\n" \ '# Rat vs. turkey BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: turkey melGal5 SEQ2_DIR=/hive/data/genomes/melGal5/melGal5.2bit SEQ2_LEN=/hive/data/genomes/melGal5/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=600 BASE=/hive/data/genomes/rn6/bed/lastzMelGal5.2017-01-17 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose) > do.log 2>&1 # real 504m54.176s cat fb.rn6.chainMelGal5Link.txt # 66617232 bases of 2729860805 (2.440%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 0m52.732s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 melGal5) \ > rbest.log 2>&1 & # real 312m57.146s # lost the log # and for the swap mkdir /hive/data/genomes/melGal5/bed/blastz.rn6.swap cd /hive/data/genomes/melGal5/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzMelGal5.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 # real 22m31.971s cat fb.melGal5.chainRn6Link.txt # 53668356 bases of 1093044709 (4.910%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` melGal5 rn6) \ > rbest.log 2>&1 # real 243m51.163s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzMelGal5.2017-01-17/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 0m45.406s ############################################################################## # lastz Elephant shark calMil1 (DONE - 2018-01-18 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S calMil1Rn6 mkdir /hive/data/genomes/rn6/bed/lastzCalMil1.2018-01-18 cd /hive/data/genomes/rn6/bed/lastzCalMil1.2018-01-18 printf "%s\n" \ '# Rat vs. Elephant shark BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: Elephant shark calMil1 SEQ2_DIR=/hive/data/genomes/calMil1/calMil1.2bit SEQ2_LEN=/hive/data/genomes/calMil1/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=50 BASE=/hive/data/genomes/rn6/bed/lastzCalMil1.2018-01-18 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose) > do.log 2>&1 # real 579m52.297s cat fb.rn6.chainCalMil1Link.txt # 45434042 bases of 2729860805 (1.664%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 0m57.126s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 calMil1) \ > rbest.log 2>&1 & # real 193m26.630s # and for the swap mkdir /hive/data/genomes/calMil1/bed/blastz.rn6.swap cd /hive/data/genomes/calMil1/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzCalMil1.2018-01-18/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 # real 17m57.104s cat fb.calMil1.chainRn6Link.txt # 40370300 bases of 936953458 (4.309%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` calMil1 rn6) \ > rbest.log 2>&1 # real 163m30.392s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzCalMil1.2018-01-18/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 0m51.155s ############################################################################## # lastz tarsier tarSyr2 (DONE - 2017-01-18 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S tarSyr2Rn6 mkdir /hive/data/genomes/rn6/bed/lastzTarSyr2.2017-01-18 cd /hive/data/genomes/rn6/bed/lastzTarSyr2.2017-01-18 printf "%s\n" \ '# Rat vs. tarsier BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz BLASTZ_Y=3400 BLASTZ_L=6000 BLASTZ_K=2200 BLASTZ_Q=/scratch/data/blastz/HoxD55.q # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=15 # QUERY: tarsier tarSyr2 SEQ2_DIR=/hive/data/genomes/tarSyr2/tarSyr2.2bit SEQ2_LEN=/hive/data/genomes/tarSyr2/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=1000 BASE=/hive/data/genomes/rn6/bed/lastzTarSyr2.2017-01-18 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium) > do.log 2>&1 # real 803m10.196s cat fb.rn6.chainTarSyr2Link.txt # 1051238729 bases of 2729860805 (38.509%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=3000 -chainLinearGap=medium \ -continue=syntenicNet) > synNet.log 2>&1 & # real 15m30.483s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 tarSyr2) \ > rbest.log 2>&1 & # real 1075m22.381s # and for the swap mkdir /hive/data/genomes/tarSyr2/bed/blastz.rn6.swap cd /hive/data/genomes/tarSyr2/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzTarSyr2.2017-01-18/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=3000 -chainLinearGap=medium) > swap.log 2>&1 # real 306m57.784s cat fb.tarSyr2.chainRn6Link.txt # 1110521625 bases of 3405755564 (32.607%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` tarSyr2 rn6) \ > rbest.log 2>&1 # real 1247m45.887s ############################################################################## # lastz platypus ornAna2 (DONE - 2017-01-18 - Hiram) # establish a screen to control this job with a name to indicate what it is screen -S ornAna2Rn6 mkdir /hive/data/genomes/rn6/bed/lastzOrnAna2.2017-01-18 cd /hive/data/genomes/rn6/bed/lastzOrnAna2.2017-01-18 printf "%s\n" \ '# Rat vs. platypus BLASTZ=/cluster/bin/penn/lastz-distrib-1.03.66/bin/lastz BLASTZ_M=254 BLASTZ_Y=3400 BLASTZ_L=6000 BLASTZ_K=2200 BLASTZ_Q=/scratch/data/blastz/HoxD55.q # TARGET: Rat Rn6 SEQ1_DIR=/hive/data/genomes/rn6/rn6.2bit SEQ1_LEN=/hive/data/genomes/rn6/chrom.sizes SEQ1_CHUNK=20000000 SEQ1_LAP=10000 SEQ1_LIMIT=5 # QUERY: platypus ornAna2 SEQ2_DIR=/hive/data/genomes/ornAna2/ornAna2.2bit SEQ2_LEN=/hive/data/genomes/ornAna2/chrom.sizes SEQ2_CHUNK=20000000 SEQ2_LAP=0 SEQ2_LIMIT=1000 BASE=/hive/data/genomes/rn6/bed/lastzOrnAna2.2017-01-18 TMPDIR=/dev/shm' > DEF # adjust the SEQ2_LIMIT with -stop=partition to get a reasonable # number of jobs, 50,000 to something under 100,000 # when not present, SEQ2_LIMIT is a default 100 time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose) > do.log 2>&1 # real 636m10.747s cat fb.rn6.chainOrnAna2Link.txt # 174410156 bases of 2729860805 (6.389%) in intersection time (doBlastzChainNet.pl -verbose=2 `pwd`/DEF -syntenicNet \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 2m21.696s time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` rn6 ornAna2) \ > rbest.log 2>&1 & # real 801m25.537s # and for the swap mkdir /hive/data/genomes/ornAna2/bed/blastz.rn6.swap cd /hive/data/genomes/ornAna2/bed/blastz.rn6.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/rn6/bed/lastzOrnAna2.2017-01-18/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose) > swap.log 2>&1 # real 94m30.003s cat fb.ornAna2.chainRn6Link.txt # 159391765 bases of 1842252119 (8.652%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` ornAna2 rn6) \ > rbest.log 2>&1 & # real 794m22.755s time (doBlastzChainNet.pl -verbose=2 -syntenicNet \ /hive/data/genomes/rn6/bed/lastzOrnAna2.2017-01-18/DEF \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -swap -chainMinScore=5000 -chainLinearGap=loose \ -continue=syntenicNet) > synNet.log 2>&1 & # real 2m59.537s ############################################################################## ## 20-Way Multiz (DONE - 2017-01-20 - Hiram) ssh hgwdev mkdir /hive/data/genomes/rn6/bed/multiz20way cd /hive/data/genomes/rn6/bed/multiz20way # from the 191-way in the source tree, select out the 5 used here: /cluster/bin/phast/tree_doctor \ --prune-all-but ailMel1,bosTau8,calMil1,canFam3,cavPor3,danRer10,\ felCat8,galGal5,hg38,melGal5,micOch1,mm10,monDom5,ornAna2,oryCun2,panTro5,\ rheMac8,rn6,tarSyr2,xenTro7 \ /cluster/home/hiram/kent/src/hg/utils/phyloTrees/191way.nh \ > rn6.20way.nh.0 cat rn6.20way.nh.0 # (((((((((((hg38:0.006550,panTro5:0.006840):0.029424, # rheMac8:0.043601):0.073920,tarSyr2:0.111400):0.036014,(((micOch1:0.140000, # (mm10:0.084509,rn6:0.091589):0.047773):0.183142,cavPor3:0.175779):0.025746, # oryCun2:0.215690):0.015313):0.020593,(bosTau8:0.189080,(felCat8:0.098612, # (canFam3:0.052458,ailMel1:0.080000):0.020000):0.060572):0.032898):0.258392, # monDom5:0.340786):0.071664,ornAna2:0.456592):0.109504,(galGal5:0.041254, # melGal5:0.085718):0.518188):0.050000,xenTro7:0.347944):0.211354, # danRer10:1.201072):0.100000,calMil1:0.750000); # using TreeGraph2 tree editor on the Mac, rearrange to get rn6 # at the top: # what that looks like: ~/kent/src/hg/utils/phyloTrees/asciiTree.pl rn6.20way.nh | sed -e 's/^/# /;' # ((((((((((((rn6:0.091589, # mm10:0.084509):0.047773, # micOch1:0.14):0.183142, # cavPor3:0.175779):0.025746, # oryCun2:0.21569):0.015313, # (((hg38:0.00655, # panTro5:0.00684):0.029424, # rheMac8:0.043601):0.07392, # tarSyr2:0.1114):0.036014):0.020593, # (((canFam3:0.052458, # ailMel1:0.08):0.02, # felCat8:0.098612):0.060572, # bosTau8:0.18908):0.032898):0.258392, # monDom5:0.340786):0.071664, # ornAna2:0.456592):0.109504, # (galGal5:0.041254, # melGal5:0.085718):0.518188):0.05, # xenTro7:0.347944):0.211354, # danRer10:1.201072):0.1, # calMil1:0.75); # extract species list from that .nh file sed 's/[a-z][a-z]*_//g; s/:[0-9\.][0-9\.]*//g; s/;//; /^ *$/d' \ rn6.20way.nh | xargs echo | sed 's/ //g; s/,/ /g' \ | sed 's/[()]//g; s/,/ /g' | tr '[ ]' '[\n]' > species.list.txt # construct db to name translation list: cat species.list.txt | while read DB do hgsql -N -e "select name,organism from dbDb where name=\"${DB}\";" hgcentraltest done | sed -e "s/\t/->/; s/ /_/g;" | sed -e 's/$/;/' | sed -e 's/\./_/g' \ | sed -e 's/-nosed/_nosed/; s/-eating/_eating/;' > db.to.name.txt # construct a common name .nh file: /cluster/bin/phast/tree_doctor --rename \ "`cat db.to.name.txt`" rn6.20way.nh | sed -e 's/00*)/)/g; s/00*,/,/g' \ | $HOME/kent/src/hg/utils/phyloTrees/asciiTree.pl /dev/stdin \ > rn6.20way.commonNames.nh cat rn6.20way.commonNames.nh | sed -e 's/^/# /;' # ((((((((((((Rat:0.091589, # Mouse:0.084509):0.047773, # Prairie_vole:0.14):0.183142, # Guinea_pig:0.175779):0.025746, # Rabbit:0.21569):0.015313, # (((Human:0.00655, # Chimp:0.00684):0.029424, # Rhesus:0.043601):0.07392, # Tarsier:0.1114):0.036014):0.020593, # (((Dog:0.052458, # Panda:0.08):0.02, # Cat:0.098612):0.060572, # Cow:0.18908):0.032898):0.258392, # Opossum:0.340786):0.071664, # Platypus:0.456592):0.109504, # (Chicken:0.041254, # Turkey:0.085718):0.518188):0.05, # X__tropicalis:0.347944):0.211354, # Zebrafish:1.201072):0.1, # Elephant_shark:0.75); # Use this specification in the phyloGif tool: # http://genome.ucsc.edu/cgi-bin/phyloGif # to obtain a png image for src/hg/htdocs/images/phylo/rn6_20way.png ~/kent/src/hg/utils/phyloTrees/asciiTree.pl rn6.20way.nh > t.nh ~/kent/src/hg/utils/phyloTrees/scientificNames.sh t.nh \ | $HOME/kent/src/hg/utils/phyloTrees/asciiTree.pl /dev/stdin \ > rn6.20way.scientificNames.nh rm -f t.nh cat rn6.20way.scientificNames.nh | sed -e 's/^/# /;' # ((((((((((((Rattus_norvegicus:0.091589, # Mus_musculus:0.084509):0.047773, # Microtus_ochrogaster:0.14):0.183142, # Cavia_porcellus:0.175779):0.025746, # Oryctolagus_cuniculus:0.21569):0.015313, # (((Homo_sapiens:0.00655, # Pan_troglodytes:0.00684):0.029424, # Macaca_mulatta:0.043601):0.07392, # Tarsius_syrichta:0.1114):0.036014):0.020593, # (((Canis_lupus_familiaris:0.052458, # Ailuropoda_melanoleuca:0.08):0.02, # Felis_catus:0.098612):0.060572, # Bos_taurus:0.18908):0.032898):0.258392, # Monodelphis_domestica:0.340786):0.071664, # Ornithorhynchus_anatinus:0.456592):0.109504, # (Gallus_gallus:0.041254, # Meleagris_gallopavo:0.085718):0.518188):0.05, # Xenopus_tropicalis:0.347944):0.211354, # Danio_rerio:1.201072):0.1, # Callorhinchus_milii:0.75); /cluster/bin/phast/all_dists rn6.20way.nh | grep rn6 \ | sed -e "s/rn6.//" | sort -k2n > 20way.distances.txt # Use this output to create the table below cat 20way.distances.txt | sed -e 's/^/# /;' # mm10 0.176098 # micOch1 0.279362 # cavPor3 0.498283 # hg38 0.509471 # panTro5 0.509761 # tarSyr2 0.510977 # rheMac8 0.517098 # canFam3 0.550084 # oryCun2 0.563940 # felCat8 0.576238 # ailMel1 0.577626 # bosTau8 0.606134 # monDom5 0.983334 # ornAna2 1.170804 # xenTro7 1.221660 # galGal5 1.383158 # melGal5 1.427622 # calMil1 1.935070 # danRer10 2.286142 printf '#!/usr/bin/env perl use strict; use warnings; open (FH, "<20way.distances.txt") or die "can not read 20way.distances.txt"; my $count = 0; while (my $line = <FH>) { chomp $line; my ($D, $dist) = split('"'"'\\s+'"'"', $line); my $chain = "chain" . ucfirst($D); my $B="/hive/data/genomes/rn6/bed/lastz.$D/fb.rn6." . $chain . "Link.txt"; my $chainLinkMeasure = `awk '"'"'{print \\$5}'"'"' ${B} 2> /dev/null | sed -e "s/(//; s/)//"`; chomp $chainLinkMeasure; $chainLinkMeasure = 0.0 if (length($chainLinkMeasure) < 1); $chainLinkMeasure =~ s/\\%%//; my $swapFile="/hive/data/genomes/${D}/bed/lastz.rn6/fb.${D}.chainRn6Link.txt"; my $swapMeasure = "N/A"; if ( -s $swapFile ) { $swapMeasure = `awk '"'"'{print \\$5}'"'"' ${swapFile} 2> /dev/null | sed -e "s/(//; s/)//"`; chomp $swapMeasure; $swapMeasure = 0.0 if (length($swapMeasure) < 1); $swapMeasure =~ s/\\%%//; } my $orgName= `hgsql -N -e "select organism from dbDb where name='"'"'$D'"'"';" hgcentraltest`; chomp $orgName; if (length($orgName) < 1) { $orgName="N/A"; } ++$count; printf "# %%02d %%.4f (%%%% %%06.3f) (%%%% %%06.3f) - %%s %%s\\n", $count, $dist, $chainLinkMeasure, $swapMeasure, $orgName, $D; } close (FH); ' > sizeStats.pl chmod +x ./sizeStats.pl ./sizeStats.pl # If you can fill in all the numbers in this table, you are ready for # the multiple alignment procedure # featureBits chainLink measures # chainLink # N distance on rn6 on other other species # 01 0.1761 (% 71.015) (% 70.886) - Mouse mm10 # 02 0.2794 (% 57.205) (% 70.134) - Prairie vole micOch1 # 03 0.4983 (% 27.834) (% 28.362) - Guinea pig cavPor3 # 04 0.5095 (% 34.899) (% 31.077) - Human hg38 # 05 0.5098 (% 40.816) (% 35.356) - Chimp panTro5 # 06 0.5110 (% 38.509) (% 32.607) - Tarsier tarSyr2 # 07 0.5171 (% 40.165) (% 33.468) - Rhesus rheMac8 # 08 0.5501 (% 28.540) (% 30.864) - Dog canFam3 # 09 0.5639 (% 24.654) (% 24.981) - Rabbit oryCun2 # 10 0.5762 (% 29.520) (% 29.673) - Cat felCat8 # 11 0.5776 (% 30.404) (% 34.777) - Panda ailMel1 # 12 0.6061 (% 25.902) (% 25.355) - Cow bosTau8 # 13 0.9833 (% 09.652) (% 07.191) - Opossum monDom5 # 14 1.1708 (% 06.389) (% 08.652) - Platypus ornAna2 # 15 1.2217 (% 04.104) (% 07.913) - X. tropicalis xenTro7 # 16 1.3832 (% 03.983) (% 07.853) - Chicken galGal5 # 17 1.4276 (% 02.440) (% 04.910) - Turkey melGal5 # 18 1.9351 (% 01.664) (% 04.309) - Elephant shark calMil1 # 19 2.2861 (% 03.232) (% 05.815) - Zebrafish danRer10 # None of this concern for distances matters in building the first step, the # maf files. The distances will be better calibrated later. # create species list and stripped down tree for autoMZ sed 's/[a-z][a-z]*_//g; s/:[0-9\.][0-9\.]*//g; s/;//; /^ *$/d' \ rn6.20way.nh | xargs echo | sed 's/ //g; s/,/ /g' > tree.nh sed 's/[()]//g; s/,/ /g' tree.nh > species.list # rn6 mm10 micOch1 cavPor3 oryCun2 hg38 panTro5 rheMac8 tarSyr2 canFam3 # ailMel1 felCat8 bosTau8 monDom5 ornAna2 galGal5 melGal5 xenTro7 # danRer10 calMil1 # survey N50 for each for db in `cat species.list` do printf "# %7s:\t" "${db}" n50.pl /hive/data/genomes/$db/chrom.sizes 2>&1 | tail -1 done | sort -k4n # monDom5: 1817563546 3 chr3 527952102 # galGal5: 630658223 5 chrZ 82310166 # xenTro7: 774070486 5 KB021657 124127367 # melGal5: 591651097 6 chr5 59006440 # rn6: 1538522439 8 chr7 145729302 # felCat8: 1440507093 8 chrB4 142431058 # hg38: 1692529807 9 chr8 145138636 # mm10: 1442871972 9 chr10 130694993 # oryCun2: 1389219202 9 chr8 111795807 # rheMac8: 1622316098 9 chr8 144306982 # panTro5: 1725767609 10 chr10 135926727 # bosTau8: 1366174162 11 chr9 105708250 # danRer10: 738680380 12 chr17 53345113 # micOch1: 1154595365 14 chrX 61813736 # canFam3: 1241874881 15 chr13 63241923 # cavPor3: 1384780737 28 scaffold_27 27942054 # calMil1: 489260987 60 KI635914 4521921 # ornAna2: 998675786 182 chrUn_DS181001v1 991605 # ailMel1: 1150550801 521 GL192858.1 1281781 # tarSyr2: 1727089816 2384 KE946086v1 401181 # bash shell syntax here ... cd /hive/data/genomes/rn6/bed/multiz20way export H=/hive/data/genomes/rn6/bed mkdir mafLinks # good assemblies can use syntenic net: # monDom5 rn6 felCat8 hg38 mm10 oryCun2 rheMac8 # panTro5 bosTau8 micOch1 canFam3 for G in monDom5 felCat8 hg38 mm10 oryCun2 cavPor3 rheMac8 \ panTro5 bosTau8 micOch1 canFam3 do mkdir mafLinks/$G echo ln -s ${H}/lastz.$G/axtChain/rn6.${G}.synNet.maf.gz ./mafLinks/$G ln -s ${H}/lastz.$G/axtChain/rn6.${G}.synNet.maf.gz ./mafLinks/$G done # distant species, use unfiltered mafNet for G in ornAna2 calMil1 danRer10 galGal5 melGal5 xenTro7 do mkdir mafLinks/$G echo ln -s ${H}/lastz.$G/mafNet/rn6.${G}.net.maf.gz ./mafLinks/$G ln -s ${H}/lastz.$G/mafNet/rn6.${G}.net.maf.gz ./mafLinks/$G done # lower quality assemblies using recip best net: # ailMel1 tarSyr2 for G in ailMel1 tarSyr2 do mkdir mafLinks/$G echo ln -s ${H}/lastz.$G/mafRBestNet/rn6.${G}.rbest.maf.gz ./mafLinks/$G ln -s ${H}/lastz.$G/mafRBestNet/rn6.${G}.rbest.maf.gz ./mafLinks/$G done # verify the symLinks are good: ls -ogrtL mafLinks/*/* | sed -e 's/-rw-rw-r-- 1//;' | while read L do let "n++" printf "# %02d\t%s\n" $n "$L" done # 01 687455612 Feb 17 2015 mafLinks/hg38/rn6.hg38.synNet.maf.gz # 02 61987372 Sep 18 2015 mafLinks/danRer10/rn6.danRer10.net.maf.gz # 03 1192966029 Apr 8 2016 mafLinks/mm10/rn6.mm10.synNet.maf.gz # 04 80158096 May 3 2016 mafLinks/galGal5/rn6.galGal5.net.maf.gz # 05 74068999 Jan 12 17:25 mafLinks/xenTro7/rn6.xenTro7.net.maf.gz # 06 604285485 Jan 17 03:57 mafLinks/ailMel1/rn6.ailMel1.rbest.maf.gz # 07 49852699 Jan 17 18:41 mafLinks/melGal5/rn6.melGal5.net.maf.gz # 08 32487411 Jan 18 19:14 mafLinks/calMil1/rn6.calMil1.net.maf.gz # 09 126292792 Jan 18 20:27 mafLinks/ornAna2/rn6.ornAna2.net.maf.gz # 10 132321627 Jan 19 10:13 mafLinks/monDom5/rn6.monDom5.synNet.maf.gz # 11 562609533 Jan 19 10:24 mafLinks/canFam3/rn6.canFam3.synNet.maf.gz # 12 535101244 Jan 19 10:24 mafLinks/cavPor3/rn6.cavPor3.synNet.maf.gz # 13 749344283 Jan 19 10:31 mafLinks/panTro5/rn6.panTro5.synNet.maf.gz # 14 739119670 Jan 19 10:32 mafLinks/rheMac8/rn6.rheMac8.synNet.maf.gz # 15 497464725 Jan 19 10:33 mafLinks/bosTau8/rn6.bosTau8.synNet.maf.gz # 16 578966556 Jan 19 10:39 mafLinks/felCat8/rn6.felCat8.synNet.maf.gz # 17 1039380892 Jan 19 10:52 mafLinks/micOch1/rn6.micOch1.synNet.maf.gz # 18 463610104 Jan 19 11:06 mafLinks/oryCun2/rn6.oryCun2.synNet.maf.gz # 19 724617931 Jan 20 02:02 mafLinks/tarSyr2/rn6.tarSyr2.rbest.maf.gz # and note that it is the correct count 20-way minus 1 == 19 # split the maf files into a set of hashed named files # this hash named split keeps the same chr/contig names in the same # named hash file. mkdir /hive/data/genomes/rn6/bed/multiz20way/mafSplit cd /hive/data/genomes/rn6/bed/multiz20way/mafSplit time for D in `sed -e "s/rn6 //" ../species.list` do echo "${D}" mkdir $D cd $D echo "mafSplit -byTarget -useHashedName=8 /dev/null . ../../mafLinks/${D}/*.maf.gz" mafSplit -byTarget -useHashedName=8 /dev/null . \ ../../mafLinks/${D}/*.maf.gz cd .. done # real 6m51.805s # construct a list of all possible maf file names. # they do not all exist in each of the species directories find . -type f | wc -l # 2173 find . -type f | grep ".maf$" | xargs -L 1 basename | sort -u > maf.list wc -l maf.list # 227 maf.list mkdir /hive/data/genomes/rn6/bed/multiz20way/splitRun cd /hive/data/genomes/rn6/bed/multiz20way/splitRun mkdir maf run cd run mkdir penn cp -p /cluster/bin/penn/multiz.2009-01-21_patched/multiz penn cp -p /cluster/bin/penn/multiz.2009-01-21_patched/maf_project penn cp -p /cluster/bin/penn/multiz.2009-01-21_patched/autoMZ penn # verify the db and pairs settings are correct printf '#!/bin/csh -ef set db = rn6 set c = $1 set result = $2 set run = `/bin/pwd` set tmp = /dev/shm/$db/multiz.$c set pairs = /hive/data/genomes/rn6/bed/multiz20way/mafSplit /bin/rm -fr $tmp /bin/mkdir -p $tmp /bin/cp -p ../../tree.nh ../../species.list $tmp pushd $tmp > /dev/null foreach s (`/bin/sed -e "s/$db //" species.list`) set in = $pairs/$s/$c set out = $db.$s.sing.maf if (-e $in.gz) then /bin/zcat $in.gz > $out if (! -s $out) then echo "##maf version=1 scoring=autoMZ" > $out endif else if (-e $in) then /bin/ln -s $in $out else echo "##maf version=1 scoring=autoMZ" > $out endif end set path = ($run/penn $path); rehash $run/penn/autoMZ + T=$tmp E=$db "`cat tree.nh`" $db.*.sing.maf $c \ > /dev/null popd > /dev/null /bin/rm -f $result /bin/cp -p $tmp/$c $result /bin/rm -fr $tmp ' > autoMultiz.csh chmod +x autoMultiz.csh printf '#LOOP ./autoMultiz.csh $(file1) {check out line+ /hive/data/genomes/rn6/bed/multiz20way/splitRun/maf/$(root1).maf} #ENDLOOP ' > template ln -s ../../mafSplit/maf.list maf.list ssh ku cd /hive/data/genomes/rn6/bed/multiz20way/splitRun/run gensub2 maf.list single template jobList para create jobList para try ... check ... push ... etc... # Completed: 227 of 227 jobs # CPU time in finished jobs: 458979s 7649.65m 127.49h 5.31d 0.015 y # IO & Wait Time: 2311s 38.52m 0.64h 0.03d 0.000 y # Average job time: 2032s 33.87m 0.56h 0.02d # Longest finished job: 41181s 686.35m 11.44h 0.48d # Submission to last job: 41324s 688.73m 11.48h 0.48d # combine into one file (the 1>&2 redirect sends the echo to stderr) cd /hive/data/genomes/rn6/bed/multiz20way head -1 splitRun/maf/013.maf > multiz20way.maf time for F in splitRun/maf/*.maf do echo "${F}" 1>&2 egrep -v "^#" ${F} done >> multiz20way.maf # real 1m49.082s tail -1 splitRun/maf/013.maf >> multiz20way.maf # -rw-rw-r-- 1 29010594028 Jan 22 09:08 multiz20way.maf # Load into database ssh hgwdev cd /hive/data/genomes/rn6/bed/multiz20way mkdir /gbdb/rn6/multiz20way ln -s `pwd`/multiz20way.maf /gbdb/rn6/multiz20way cd /dev/shm time hgLoadMaf rn6 multiz20way # Loaded 24754572 mafs in 1 files from /gbdb/rn6/multiz20way # real 8m51.299s time hgLoadMafSummary -verbose=2 -minSize=30000 \ -mergeGap=1500 -maxSize=200000 rn6 multiz20waySummary \ /gbdb/rn6/multiz20way/multiz20way.maf # Created 3877691 summary blocks from 169823109 components and 24754572 mafs from /gbdb/rn6/multiz20way/multiz20way.maf # real 12m4.373s # -rw-rw-r-- 1 1265516402 Jan 22 09:15 multiz20way.tab # -rw-rw-r-- 1 182504514 Jan 22 09:37 multiz20waySummary.tab wc -l multiz20way*.tab # 24754572 multiz20way.tab # 3877691 multiz20waySummary.tab rm multiz20way*.tab ############################################################################## # GAP ANNOTATE MULTIZ20WAY MAF AND LOAD TABLES (DONE - 2017-01-22 - Hiram) # mafAddIRows has to be run on single chromosome maf files, it does not # function correctly when more than one reference sequence # are in a single file. Need to split of the maf file into individual # maf files mkdir -p /hive/data/genomes/rn6/bed/multiz20way/anno/mafSplit cd /hive/data/genomes/rn6/bed/multiz20way/anno/mafSplit time mafSplit -outDirDepth=2 -byTarget -useFullSequenceName \ /dev/null . ../../multiz20way.maf # real 10m33.882s find . -type f | wc -l # 624 # check for N.bed files everywhere: cd /hive/data/genomes/rn6/bed/multiz20way/anno for DB in `cat ../species.list` do if [ ! -s /hive/data/genomes/${DB}/${DB}.N.bed ]; then echo "MISS: ${DB}" # cd /hive/data/genomes/${DB} # twoBitInfo -nBed ${DB}.2bit ${DB}.N.bed else echo " OK: ${DB}" fi done cd /hive/data/genomes/rn6/bed/multiz20way/anno for DB in `cat ../species.list` do echo "${DB} " ln -s /hive/data/genomes/${DB}/${DB}.N.bed ${DB}.bed echo ${DB}.bed >> nBeds ln -s /hive/data/genomes/${DB}/chrom.sizes ${DB}.len echo ${DB}.len >> sizes done # make sure they all are successful symLinks: ls -ogrtL screen -S gapAnno # use a screen to control this longish job ssh ku cd /hive/data/genomes/rn6/bed/multiz20way/anno mkdir result find ./mafSplit -type d | sed -e 's#./mafSplit/##' | while read D do echo mkdir -p result/${D} mkdir -p result/${D} done printf '#LOOP mafAddIRows -nBeds=nBeds mafSplit/$(path1) /hive/data/genomes/rn6/rn6.2bit {check out exists+ result/$(path1)} #ENDLOOP ' > template # << happy emacs find ./mafSplit -type f | sed -e 's#^./mafSplit/##' > maf.list gensub2 maf.list single template jobList # limit jobs on a node with the ram=32g requirement because they go fast para -ram=32g create jobList para try ... check ... push ... # Completed: 624 of 624 jobs # CPU time in finished jobs: 1802s 30.03m 0.50h 0.02d 0.000 y # IO & Wait Time: 1624s 27.07m 0.45h 0.02d 0.000 y # Average job time: 5s 0.09m 0.00h 0.00d # Longest finished job: 148s 2.47m 0.04h 0.00d # Submission to last job: 177s 2.95m 0.05h 0.00d # verify all result files have some content, look for 0 size files: find ./result -type f -size 0 # should see none # or in this manner: find ./result -type f | xargs ls -og | sort -k3nr | tail # combine into one file (the 1>&2 redirect sends the echo to stderr) head -q -n 1 result/4/0/chrUn_KL568283v1.maf > rn6.20way.maf time find ./result -type f | while read F do echo "${F}" 1>&2 grep -h -v "^#" ${F} done >> rn6.20way.maf # real 2m13.643s # these maf files do not have the end marker, this does nothing: # tail -q -n 1 result/4/0/chrUn_KL568283v1.maf >> rn6.20way.maf # How about an official end marker: echo "##eof maf" >> rn6.20way.maf # it almost doubles in size ls -og # -rw-rw-r-- 1 48341214170 Jan 22 10:39 rn6.20way.maf # -rw-rw-r-- 1 29010594028 Jan 22 09:08 multiz20way.maf du -hsc rn6.20way.maf # 46G rn6.20way.maf # construct symlinks to get the individual maf files into gbdb: rm /gbdb/rn6/multiz20way/multiz20way.maf # remove previous results ln -s `pwd`/rn6.20way.maf /gbdb/rn6/multiz20way/multiz20way.maf # Load into database cd /dev/shm time hgLoadMaf -pathPrefix=/gbdb/rn6/multiz20way rn6 multiz20way # Loaded 25393063 mafs in 1 files from /gbdb/rn6/multiz20way # real 12m43.103s time hgLoadMafSummary -verbose=2 -minSize=30000 \ -mergeGap=1500 -maxSize=200000 rn6 multiz20waySummary \ /gbdb/rn6/multiz20way/multiz20way.maf # Created 3877691 summary blocks from 169823109 components # and 25393063 mafs from /gbdb/rn6/multiz20way/multiz20way.maf # real 15m39.900s # -rw-rw-r-- 1 1301989038 Jan 22 10:53 multiz20way.tab # -rw-rw-r-- 1 190259896 Jan 22 11:12 multiz20waySummary.tab rm multiz20way*.tab ###################################################################### # MULTIZ20WAY MAF FRAMES (DONE - 2017-01-22 - Hiram) ssh hgwdev mkdir /hive/data/genomes/rn6/bed/multiz20way/frames cd /hive/data/genomes/rn6/bed/multiz20way/frames # survey all the genomes to find out what kinds of gene tracks they have printf '#!/bin/csh -fe foreach db (`cat ../species.list`) printf "# ${db}: " set tables = `hgsql $db -N -e "show tables" | egrep "Gene|ncbiRefSeq"` foreach table ($tables) if ($table == "ensGene" || $table == "refGene" || \ $table == "ncbiRefSeq" || $table == "mgcGenes" || \ $table == "knownGene" || $table == "xenoRefGene" ) then set count = `hgsql $db -N -e "select count(*) from $table"` echo -n "${table}: ${count}, " endif end set orgName = `hgsql hgcentraltest -N -e \ "select scientificName from dbDb where name='"'"'$db'"'"'"` set orgId = `hgsql $db -N -e \ "select id from organism where name='"'"'$orgName'"'"'"` if ($orgId == "") then echo "Mrnas: 0" else set count = `hgsql $db -N -e "select count(*) from gbCdnaInfo where organism=$orgId"` echo "Mrnas: ${count}" endif end ' > showGenes.csh # << happy emacs chmod +x ./showGenes.csh time ./showGenes.csh # rn6: ensGene: 40459, mgcGenes: 7017, refGene: 19039, xenoRefGene: 197625, Mrnas: 1250600 # mm10: ensGene: 103734, knownGene: 63759, mgcGenes: 26777, ncbiRefSeq: 106507, refGene: 36803, xenoRefGene: 177505, Mrnas: 5364985 # micOch1: Mrnas: 19 # cavPor3: ensGene: 26129, refGene: 488, xenoRefGene: 305485, Mrnas: 21235 # oryCun2: ensGene: 24964, refGene: 1445, xenoRefGene: 313991, Mrnas: 40491 # hg38: ensGene: 208239, knownGene: 197782, mgcGenes: 35305, ncbiRefSeq: 159322, refGene: 67251, xenoRefGene: 184734, Mrnas: 11474400 # panTro5: refGene: 2897, xenoRefGene: 223650, Mrnas: 11351 # rheMac8: ensGene: 56743, refGene: 6465, xenoRefGene: 214674, Mrnas: 447197 # tarSyr2: xenoRefGene: 337666, Mrnas: 0 # canFam3: ensGene: 39074, refGene: 2291, xenoRefGene: 262153, Mrnas: 388033 # ailMel1: ensGene: 25055, xenoRefGene: 338914, Mrnas: 276 # felCat8: refGene: 433, xenoRefGene: 225635, Mrnas: 3785 # bosTau8: refGene: 14630, xenoRefGene: 297152, Mrnas: 1617801 # monDom5: ensGene: 24882, refGene: 995, xenoRefGene: 251960, Mrnas: 2814 # ornAna2: refGene: 695, xenoRefGene: 605912, Mrnas: 10740 # galGal5: ensGene: 48760, refGene: 7541, xenoRefGene: 235493, Mrnas: 638577 # melGal5: refGene: 109, xenoRefGene: 231886, Mrnas: 17894 # xenTro7: mgcGenes: 9029, refGene: 8910, xenoRefGene: 265136, Mrnas: 1298763 # danRer10: ensGene: 58274, mgcGenes: 16947, refGene: 16551, xenoRefGene: 127693, Mrnas: 1535288 # calMil1: xenoRefGene: 361187, Mrnas: 146271 # from that summary, use these gene sets: # ensGene: rn6 cavPor3 oryCun2 rheMac8 canFam3 ailMel1 monDom5 galGal5 # danRer10 # refGene: panTro5 bosTau8 xenTro7 # knownGene - hg38 mm10 # none: micOch1 tarSyr2 felCat8 ornAna2 melGal5 calMil1 mkdir genes # 1. knownGene: hg38 mm10 for DB in hg38 mm10 do hgsql -N -e "select name,chrom,strand,txStart,txEnd,cdsStart,cdsEnd,exonCount,exonStarts,exonEnds from knownGene" ${DB} \ | genePredSingleCover stdin stdout | gzip -2c \ > genes/${DB}.gp.gz printf "# ${DB}: " genePredCheck -db=${DB} genes/${DB}.gp.gz done # hg38: checked: 21375 failed: 0 # mm10: checked: 21100 failed: 0 # 2. ensGene: rn6 cavPor3 oryCun2 rheMac8 canFam3 ailMel1 # monDom5 galGal5 danRer10 for DB in rn6 cavPor3 oryCun2 rheMac8 canFam3 ailMel1 monDom5 galGal5 danRer10 do hgsql -N -e "select name,chrom,strand,txStart,txEnd,cdsStart,cdsEnd,exonCount,exonStarts,exonEnds from ensGene" ${DB} \ | genePredSingleCover stdin stdout | gzip -2c \ > /dev/shm/${DB}.tmp.gz mv /dev/shm/${DB}.tmp.gz genes/$DB.gp.gz printf "# ${DB}: " genePredCheck -db=${DB} genes/${DB}.gp.gz done # rn6: checked: 22124 failed: 0 # cavPor3: checked: 18631 failed: 0 # oryCun2: checked: 19165 failed: 0 # rheMac8: checked: 20859 failed: 0 # canFam3: checked: 19507 failed: 0 # ailMel1: checked: 19204 failed: 0 # monDom5: checked: 21033 failed: 0 # galGal5: checked: 16229 failed: 0 # danRer10: checked: 25250 failed: 0 # 3. refGene for panTro5 bosTau8 refGene for DB in panTro5 bosTau8 xenTro7 do hgsql -N -e "select * from refGene" ${DB} | cut -f2- \ | genePredSingleCover stdin stdout | gzip -2c \ > /dev/shm/${DB}.tmp.gz mv /dev/shm/${DB}.tmp.gz genes/$DB.gp.gz echo -n "# ${DB}: " genePredCheck -db=${DB} genes/${DB}.gp.gz done # panTro5: checked: 2068 failed: 0 # bosTau8: checked: 13344 failed: 0 # xenTro7: checked: 8517 failed: 0 # verify counts for genes are reasonable: for T in genes/*.gz do echo -n "# $T: " zcat $T | cut -f1 | sort | uniq -c | wc -l done # genes/ailMel1.gp.gz: 19204 # genes/bosTau8.gp.gz: 13071 # genes/canFam3.gp.gz: 19507 # genes/cavPor3.gp.gz: 18631 # genes/danRer10.gp.gz: 25250 # genes/galGal5.gp.gz: 16229 # genes/hg38.gp.gz: 21375 # genes/mm10.gp.gz: 21100 # genes/monDom5.gp.gz: 21033 # genes/oryCun2.gp.gz: 19165 # genes/panTro5.gp.gz: 1973 # genes/rheMac8.gp.gz: 20859 # genes/rn6.gp.gz: 22124 # genes/xenTro7.gp.gz: 8338 time (cat ../anno/rn6.20way.maf | genePredToMafFrames rn6 stdin stdout \ `ls genes | sed -e 's/.gp.gz//;' | xargs echo \ | sed -e "s#\([a-zA-Z0-9]*\)#\1 genes/\1.gp.gz#g;"` \ | gzip > multiz20wayFrames.bed.gz) # real 11m43.455s # verify there are frames on everything, should be 14 species: zcat multiz20wayFrames.bed.gz | awk '{print $4}' | sort | uniq -c \ | sed -e 's/^/# /;' # 234094 ailMel1 # 148818 bosTau8 # 247686 canFam3 # 216773 cavPor3 # 452415 danRer10 # 403321 galGal5 # 241462 hg38 # 232386 mm10 # 233904 monDom5 # 214841 oryCun2 # 20024 panTro5 # 251019 rheMac8 # 196117 rn6 # 198989 xenTro7 # load the resulting file ssh hgwdev cd /hive/data/genomes/rn6/bed/multiz20way/frames time hgLoadMafFrames rn6 multiz20wayFrames multiz20wayFrames.bed.gz # real 0m37.422s time featureBits -countGaps rn6 multiz20wayFrames # 51905008 bases of 2870182909 (1.808%) in intersection # real 0m21.147s # enable the trackDb entries: # frames multiz20wayFrames # irows on # appears to work OK ######################################################################### # Phylogenetic tree for 20-way (DONE - 2017-01-23 - Hiram) mkdir /hive/data/genomes/rn6/bed/multiz20way/4d cd /hive/data/genomes/rn6/bed/multiz20way/4d # using the ensGene hgsql -N -e "select * from ensGene;" rn6 \ | cut -f2- | genePredSingleCover stdin stdout > rn6.ensGene.gp genePredCheck -db=rn6 rn6.ensGene.gp # checked: 22124 failed: 0 # the annotated maf is: og ../anno/rn6.20way.maf # -rw-rw-r-- 1 48341214170 Jan 22 10:39 ../anno/rn6.20way.maf mkdir annoSplit cd annoSplit time mafSplit -verbose=2 -outDirDepth=2 -byTarget -useFullSequenceName \ /dev/null . ../../anno/rn6.20way.maf # real 20m9.735s find . -type f | wc -l # 624 ssh ku mkdir /hive/data/genomes/rn6/bed/multiz20way/4d/run cd /hive/data/genomes/rn6/bed/multiz20way/4d/run mkdir ../mfa # newer versions of msa_view have a slightly different operation # the sed of the gp file inserts the reference species in the chr name printf '#!/bin/csh -fe set PHASTBIN = /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin set GP = rn6.ensGene.gp set r = "/hive/data/genomes/rn6/bed/multiz20way" set c = $1:r set infile = $r/4d/annoSplit/$2 set outDir = $r/4d/mfa/$3:h set outfile = $r/4d/mfa/$3 /bin/mkdir -p $outDir cd /dev/shm /bin/awk -v C=$c '"'"'$2 == C {print}'"'"' $r/4d/$GP | sed -e "s/\\t$c\\t/\\trn6.$c\\t/" > $c.gp set NL=`wc -l $c.gp| gawk '"'"'{print $1}'"'"'` echo $NL if ("$NL" != "0") then $PHASTBIN/msa_view --4d --features $c.gp -i MAF $infile -o SS > $c.ss $PHASTBIN/msa_view -i SS --tuple-size 1 $c.ss > $outfile else echo "" > $outfile endif /bin/rm -f /dev/shm/$c.gp /dev/shm/$c.ss ' > 4d.csh chmod +x 4d.csh find ../annoSplit -type f | sed -e "s#../annoSplit/##" > maf.list wc -l maf.list # 624 maf.list printf '#LOOP 4d.csh $(file1) $(path1) {check out line+ ../mfa/$(dir1)/$(dir2)$(root1).mfa} #ENDLOOP ' > template gensub2 maf.list single template jobList # better to use 32g here to avoid out of memory, and they run quickly # anyway, no need to run so many at a time para -ram=32g create jobList para try ... check para time # one job ran out of memory, finish on hgwdev: time ./4d.csh chr1.maf 0/3/chr1.maf ../mfa/0/3/chr1.mfa # real 3m5.094s # 17 others have no results # Completed: 606 of 624 jobs # Crashed: 18 jobs # CPU time in finished jobs: 1241s 20.68m 0.34h 0.01d 0.000 y # IO & Wait Time: 1805s 30.09m 0.50h 0.02d 0.000 y # Average job time: 5s 0.08m 0.00h 0.00d # Longest finished job: 130s 2.17m 0.04h 0.00d # Submission to last job: 170s 2.83m 0.05h 0.00d # Not all results have contents, or finish successfully, that is OK # it is because not all contigs have genes, only gene sequences are measured # combine mfa files ssh hgwdev cd /hive/data/genomes/rn6/bed/multiz20way/4d # remove the broken empty files, size 0 and size 1: find ./mfa -type f -size 0 | xargs rm -f # sometimes this doesn't work, don't know why find ./mfa -type f -size 1 | xargs rm -f # when it doesn't, use this empty list procedure find ./mfa -type f | xargs ls -og | awk '$3 < 2' | awk '{print $NF}' \ > empty.list cat empty.list | xargs rm -f # see what is left: ls -ogrt mfa/*/*/*.mfa | sort -k3nr | wc -l # 116 # want comma-less species.list time /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/msa_view \ --aggregate "`cat ../species.list`" mfa/*/*/*.mfa | sed s/"> "/">"/ \ > 4d.all.mfa # real 0m2.356s # check they are all in there, should be 20: grep "^>" 4d.all.mfa | wc -l # 20 grep "^>" 4d.all.mfa | sed -e 's/^/# /;' # >rn6 # >mm10 # >micOch1 # >cavPor3 # >oryCun2 # >hg38 # >panTro5 # >rheMac8 # >tarSyr2 # >canFam3 # >ailMel1 # >felCat8 # >bosTau8 # >monDom5 # >ornAna2 # >galGal5 # >melGal5 # >xenTro7 # >danRer10 # >calMil1 sed 's/[a-z][a-z]*_//g; s/:[0-9\.][0-9\.]*//g; s/;//; /^ *$/d' \ ../rn6.20way.nh | xargs echo | sed -e 's/ //g' > tree_commas.nh # tree_commas.nh looks like: # ((((((((((((rn6,mm10),micOch1),cavPor3),oryCun2),(((hg38,panTro5),rheMac8), # tarSyr2)),(((canFam3,ailMel1),felCat8),bosTau8)),monDom5),ornAna2), # (galGal5,melGal5)),xenTro7),danRer10),calMil1) # use phyloFit to create tree model (output is phyloFit.mod) time /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/phyloFit \ --EM --precision MED --msa-format FASTA --subst-mod REV \ --tree tree_commas.nh 4d.all.mfa # real 31m34.749s mv phyloFit.mod all.mod grep TREE all.mod # ((((((((((((rn6:0.0895665,mm10:0.0844801):0.0651926, # micOch1:0.125658):0.158112,cavPor3:0.215578):0.0359592, # oryCun2:0.199508):0.015313,(((hg38:0.00685733,panTro5:0.0068924):0.0246446, # rheMac8:0.0361324):0.0780051,tarSyr2:0.139334):0.0250771):0.0171773, # (((canFam3:0.088804,ailMel1:0.077809):0.0216705, # felCat8:0.0896594):0.0508452,bosTau8:0.167604):0.0287444):0.244728, # monDom5:0.336766):0.0739297,ornAna2:0.491447):0.142595, # (galGal5:0.0589443,melGal5:0.0696907):0.398276):0.146609, # xenTro7:0.856129):0.151588,danRer10:1.03966):0.340195,calMil1:0.340195); # compare these calculated lengths to the tree extracted from 191way: grep TREE all.mod | sed -e 's/TREE: //' \ | /cluster/bin/phast/all_dists /dev/stdin | grep rn6 \ | sed -e "s/rn6.//;" | sort > new.dists /cluster/bin/phast/all_dists ../rn6.20way.nh | grep rn6 \ | sed -e "s/rn6.//;" | sort > old.dists # printing out the 'new', the 'old' the 'difference' and percent difference join new.dists old.dists | awk '{ printf "#\t%s\t%8.6f\t%8.6f\t%8.6f\t%8.6f\n", $1, $2, $3, $2-$3, 100*($2-$3)/$3 }' \ | sort -k3n # mm10 0.174047 0.176098 -0.002051 -1.164692 # micOch1 0.280417 0.279362 0.001055 0.377646 # hg38 0.498727 0.509471 -0.010744 -2.108854 # panTro5 0.498762 0.509761 -0.010999 -2.157678 # rheMac8 0.503358 0.517098 -0.013740 -2.657137 # cavPor3 0.528449 0.498283 0.030166 6.053989 # tarSyr2 0.528554 0.510977 0.017577 3.439881 # oryCun2 0.548338 0.563940 -0.015602 -2.766606 # felCat8 0.550570 0.576238 -0.025668 -4.454409 # ailMel1 0.560390 0.577626 -0.017236 -2.983938 # canFam3 0.571385 0.550084 0.021301 3.872318 # bosTau8 0.577669 0.606134 -0.028465 -4.696156 # monDom5 0.962815 0.983334 -0.020519 -2.086677 # ornAna2 1.191425 1.170804 0.020621 1.761268 # galGal5 1.299794 1.383158 -0.083364 -6.027077 # melGal5 1.310540 1.427622 -0.117082 -8.201191 # calMil1 1.821160 1.935070 -0.113910 -5.886609 # xenTro7 1.845311 1.221660 0.623651 51.049474 # danRer10 2.180430 2.286142 -0.105712 -4.624035 ######################################################################### # phastCons 20-way (DONE - 2017-01-22 - Hiram) # split 20way mafs into 10M chunks and generate sufficient statistics # files for # phastCons ssh ku mkdir -p /hive/data/genomes/rn6/bed/multiz20way/cons/SS cd /hive/data/genomes/rn6/bed/multiz20way/cons/SS mkdir result done printf '#!/bin/csh -ef set d = $1 set c = $2 set doneDir = done/$d set MAF = /hive/data/genomes/rn6/bed/multiz20way/anno/result/$d/$c.maf set WINDOWS = /hive/data/genomes/rn6/bed/multiz20way/cons/SS/result/$d/$c set WC = `cat $MAF | wc -l` set NL = `grep "^#" $MAF | wc -l` if ( -s $3 ) then exit 0 endif if ( -s $3.running ) then exit 0 endif /bin/mkdir -p $doneDir /bin/date >> $3.running /bin/rm -fr $WINDOWS /bin/mkdir -p $WINDOWS pushd $WINDOWS > /dev/null if ( $WC != $NL ) then /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/msa_split \\ $MAF -i MAF -o SS -r $WINDOWS/$c -w 10000000,0 -I 1000 -B 5000 endif popd > /dev/null /bin/date >> $3 /bin/rm -f $3.running ' > mkSS.csh chmod +x mkSS.csh printf '#LOOP mkSS.csh $(dir1) $(root1) {check out line+ done/$(dir1)/$(root1)} #ENDLOOP ' > template find ../../anno/result -type f | sed -e "s#../../anno/result/##" > maf.list wc -l maf.list # 624 maf.list ssh ku cd /hive/data/genomes/rn6/bed/multiz20way/cons/SS gensub2 maf.list single template jobList # beware overwhelming the cluster with these quick high I/O jobs para create jobList para try ... check ... etc para push # Completed: 624 of 624 jobs # CPU time in finished jobs: 5547s 92.45m 1.54h 0.06d 0.000 y # IO & Wait Time: 1900s 31.67m 0.53h 0.02d 0.000 y # Average job time: 12s 0.20m 0.00h 0.00d # Longest finished job: 681s 11.35m 0.19h 0.01d # Submission to last job: 733s 12.22m 0.20h 0.01d find ./result -type f | wc -l # 638 # Run phastCons # This job is I/O intensive in its output files, beware where this # takes place or do not run too many at once. ssh ku mkdir -p /hive/data/genomes/rn6/bed/multiz20way/cons/run.cons cd /hive/data/genomes/rn6/bed/multiz20way/cons/run.cons # This is setup for multiple runs based on subsets, but only running # the 'all' subset here. # It triggers off of the current working directory # $cwd:t which is the "grp" in this script. Running: # all and vertebrates printf '#!/bin/csh -fe set PHASTBIN = /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin set c = $1 set d = $2 set f = $3 set len = $4 set cov = $5 set rho = $6 set grp = $cwd:t set cons = /hive/data/genomes/rn6/bed/multiz20way/cons set tmp = $cons/tmp/${d}_${c} mkdir -p $tmp set ssSrc = $cons/SS/result set useGrp = "$grp.mod" if (-s $cons/$grp/$grp.non-inf) then ln -s $cons/$grp/$grp.mod $tmp ln -s $cons/$grp/$grp.non-inf $tmp ln -s $ssSrc/$d/$f $tmp else ln -s $ssSrc/$d/$f $tmp ln -s $cons/$grp/$grp.mod $tmp endif pushd $tmp > /dev/null if (-s $grp.non-inf) then $PHASTBIN/phastCons $f $useGrp \ --rho $rho --expected-length $len --target-coverage $cov --quiet \\ --not-informative `cat $grp.non-inf` \\ --seqname $c --idpref $c --most-conserved $c.bed --score > $c.pp else $PHASTBIN/phastCons $f $useGrp \\ --rho $rho --expected-length $len --target-coverage $cov --quiet \\ --seqname $c --idpref $c --most-conserved $c.bed --score > $c.pp endif popd > /dev/null mkdir -p pp/$d bed/$d sleep 4 touch pp/$d bed/$d rm -f pp/$d/$c.pp rm -f bed/$d/$c.bed mv $tmp/$c.pp pp/$d mv $tmp/$c.bed bed/$d rm -fr $tmp rmdir --ignore-fail-on-non-empty $cons/tmp/$d:h ' > doPhast.csh chmod +x doPhast.csh # this template will serve for all runs # root1 == chrom name, file1 == ss file name without .ss suffix printf '#LOOP ../run.cons/doPhast.csh $(root1) $(dir1) $(file1) 45 0.3 0.3 {check out line+ pp/$(dir1)/$(root1).pp} #ENDLOOP ' > template find ../SS/result -type f | sed -e "s#../SS/result/##" > ss.list wc -l ss.list # 638 ss.list # Create parasol batch and run it # run for all species cd /hive/data/genomes/rn6/bed/multiz20way/cons mkdir -p all cd all # Using the .mod tree cp -p ../../4d/all.mod ./all.mod gensub2 ../run.cons/ss.list single ../run.cons/template jobList para -ram=32g create jobList para try ... check ... para push # Completed: 638 of 638 jobs # CPU time in finished jobs: 10814s 180.24m 3.00h 0.13d 0.000 y # IO & Wait Time: 4534s 75.56m 1.26h 0.05d 0.000 y # Average job time: 24s 0.40m 0.01h 0.00d # Longest finished job: 57s 0.95m 0.02h 0.00d # Submission to last job: 101s 1.68m 0.03h 0.00d # create Most Conserved track cd /hive/data/genomes/rn6/bed/multiz20way/cons/all time cut -f1 ../../../../chrom.sizes | while read C do ls -d bed/?/?/${C} 2> /dev/null | while read D do echo ${D}/${C}*.bed 1>&2 cat ${D}/${C}*.bed done | sort -k1,1 -k2,2n \ | awk '{printf "%s\t%d\t%d\tlod=%d\t%s\n", "'${C}'", $2, $3, $5, $5;}' done > tmpMostConserved.bed # real 0m40.602s time /cluster/bin/scripts/lodToBedScore tmpMostConserved.bed \ > mostConserved.bed # real 0m23.400s # -rw-rw-r-- 1 92173490 Jan 23 09:13 mostConserved.bed # load into database ssh hgwdev cd /hive/data/genomes/rn6/bed/multiz20way/cons/all time hgLoadBed rn6 phastConsElements20way mostConserved.bed # Read 2626510 elements of size 5 from mostConserved.bed # real 0m24.079s # on human we often try for 5% overall cov, and 70% CDS cov # most bets are off here for that goal, these alignments are too few # and too far between # --rho 0.3 --expected-length 45 --target-coverage 0.3 time featureBits rn6 -enrichment ncbiRefSeq:cds phastConsElements20way # ncbiRefSeq:cds 1.217%, phastConsElements20way 3.976%, both 0.819%, # cover 67.27%, enrich 16.92x # real 2m33.330s # Create merged posterier probability file and wiggle track data files cd /hive/data/genomes/rn6/bed/multiz20way/cons/all mkdir downloads # the third sed fixes the chrom names, removing the partition extensions time (find ./pp -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \ | sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \ | sed -e 's/\.[0-9][0-9]*-[0-9][0-9]* start/ start/' \ | gzip -c > downloads/phastCons20way.wigFix.gz) # real 21m37.207s # -rw-rw-r-- 1 2339247044 Jan 23 09:36 phastCons20way.wigFix.gz # check integrity of data with wigToBigWig time (zcat downloads/phastCons20way.wigFix.gz \ | wigToBigWig -verbose=2 stdin /hive/data/genomes/rn6/chrom.sizes \ phastCons20way.bw) > bigWig.log 2>&1 egrep "real|VmPeak" bigWig.log # pid=61873: VmPeak: 23109920 kB # real 30m44.092s bigWigInfo phastCons20way.bw | sed -e 's/^/# /;' # version: 4 # isCompressed: yes # isSwapped: 0 # primaryDataSize: 3,494,424,791 # primaryIndexSize: 78,132,788 # zoomLevels: 10 # chromCount: 368 # basesCovered: 2,078,464,015 # mean: 0.135364 # min: 0.000000 # max: 1.000000 # std: 0.268417 # encode those files into wiggle data time (zcat downloads/phastCons20way.wigFix.gz \ | wigEncode stdin phastCons20way.wig phastCons20way.wib) # Converted stdin, upper limit 1.00, lower limit 0.00 # real 10m54.962s du -hsc *.wi? # 2.0G phastCons20way.wib # 233M phastCons20way.wig # Load gbdb and database with wiggle. ln -s `pwd`/phastCons20way.wib /gbdb/rn6/multiz20way/phastCons20way.wib time hgLoadWiggle -pathPrefix=/gbdb/rn6/multiz20way \ rn6 phastCons20way phastCons20way.wig # real 0m29.148s # use to set trackDb.ra entries for wiggle min and max # and verify table is loaded correctly wigTableStats.sh rn6 phastCons20way # db.table min max mean count sumData # rn6.phastCons20way 0 1 0.135364 2078464015 2.8135e+08 # 0.268417 viewLimits=0:1 # Create histogram to get an overview of all the data time hgWiggle -doHistogram -db=rn6 \ -hBinSize=0.001 -hBinCount=1000 -hMinVal=0.0 -verbose=2 \ phastCons20way > histogram.data 2>&1 # real 1m55.313s # create plot of histogram: printf 'set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff font \ "/usr/share/fonts/default/Type1/n022004l.pfb" set size 1.4, 0.8 set key left box set grid noxtics set grid ytics set title " Rat rn6 Histogram phastCons20way track" set xlabel " phastCons20way score" set ylabel " Relative Frequency" set y2label " Cumulative Relative Frequency (CRF)" set y2range [0:1] set y2tics set yrange [0:0.02] plot "histogram.data" using 2:5 title " RelFreq" with impulses, \\ "histogram.data" using 2:7 axes x1y2 title " CRF" with lines ' | gnuplot > histo.png display histo.png & ######################################################################### # phyloP for 20-way (DONE - 2016-05-09,11 - Hiram) # run phyloP with score=LRT ssh ku mkdir /cluster/data/rn6/bed/multiz20way/consPhyloP cd /cluster/data/rn6/bed/multiz20way/consPhyloP mkdir run.phyloP cd run.phyloP # Adjust model file base composition background and rate matrix to be # representative of the chromosomes in play grep BACKGROUND ../../4d/all.mod | awk '{printf "%0.3f\n", $3 + $4}' # 0.536 /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin/modFreqs \ ../../4d/all.mod 0.536 > all.mod # verify, the BACKGROUND should now be paired up: grep BACK all.mod # BACKGROUND: 0.232000 0.268000 0.268000 0.232000 printf '#!/bin/csh -fe set PHASTBIN = /cluster/bin/phast.build/cornellCVS/phast.2010-12-30/bin set f = $1 set d = $f:h set file1 = $f:t set out = $2 set cName = $f:t:r set grp = $cwd:t set cons = /hive/data/genomes/rn6/bed/multiz20way/consPhyloP set tmp = $cons/tmp/$grp/$f /bin/rm -fr $tmp /bin/mkdir -p $tmp set ssSrc = "/hive/data/genomes/rn6/bed/multiz20way/cons/SS/result/$f" set useGrp = "$grp.mod" /bin/ln -s $cons/run.phyloP/$grp.mod $tmp pushd $tmp > /dev/null $PHASTBIN/phyloP --method LRT --mode CONACC --wig-scores --chrom $cName \\ -i SS $useGrp $ssSrc.ss > $file1.wigFix popd > /dev/null /bin/mkdir -p $out:h sleep 4 /bin/touch $out:h /bin/mv $tmp/$file1.wigFix $out /bin/rm -fr $tmp /bin/rmdir --ignore-fail-on-non-empty $cons/tmp/$grp/$d /bin/rmdir --ignore-fail-on-non-empty $cons/tmp/$grp/$d:h /bin/rmdir --ignore-fail-on-non-empty $cons/tmp/$grp /bin/rmdir --ignore-fail-on-non-empty $cons/tmp ' > doPhyloP.csh chmod +x doPhyloP.csh # Create list of chunks find ../../cons/SS/result -type f | grep ".ss$" \ | sed -e "s/.ss$//; s#^../../cons/SS/result/##" > ss.list # make sure the list looks good wc -l ss.list # 638 ss.list # Create template file # file1 == $chr/$chunk/file name without .ss suffix printf '#LOOP ../run.phyloP/doPhyloP.csh $(path1) {check out line+ wigFix/$(dir1)/$(file1).wigFix} #ENDLOOP ' > template ###################### Running all species ####################### # setup run for all species mkdir /hive/data/genomes/rn6/bed/multiz20way/consPhyloP/all cd /hive/data/genomes/rn6/bed/multiz20way/consPhyloP/all rm -fr wigFix mkdir wigFix gensub2 ../run.phyloP/ss.list single ../run.phyloP/template jobList # beware overwhelming the cluster with these fast running high I/O jobs para create jobList para try ... check ... push ... etc ... para push para time > run.time # Completed: 638 of 638 jobs # CPU time in finished jobs: 277914s 4631.90m 77.20h 3.22d 0.009 y # IO & Wait Time: 8871s 147.85m 2.46h 0.10d 0.000 y # Average job time: 450s 7.49m 0.12h 0.01d # Longest finished job: 1737s 28.95m 0.48h 0.02d # Submission to last job: 3476s 57.93m 0.97h 0.04d mkdir downloads time (find ./wigFix -type f | sed -e "s#^./##; s#\.# d #g; s#-# m #;" \ | sort -k1,1 -k3,3n | sed -e "s# d #.#g; s# m #-#g;" | xargs cat \ | gzip -c > downloads/phyloP20way.wigFix.gz) # real 32m19.962s # check integrity of data with wigToBigWig time (zcat downloads/phyloP20way.wigFix.gz \ | wigToBigWig -verbose=2 stdin /hive/data/genomes/rn6/chrom.sizes \ phyloP20way.bw) > bigWig.log 2>&1 XXX - running - Mon Jan 23 19:45:33 PST 2017 egrep "real|VmPeak" bigWig.log # pid=25924: VmPeak: 23109924 kB # real 30m51.433s bigWigInfo phyloP20way.bw | sed -e 's/^/# /;' # version: 4 # isCompressed: yes # isSwapped: 0 # primaryDataSize: 4,248,984,727 # primaryIndexSize: 78,132,788 # zoomLevels: 10 # chromCount: 368 # basesCovered: 2,078,464,015 # mean: 0.108077 # min: -8.546000 # max: 3.488000 # std: 0.723885 # encode those files into wiggle data time (zcat downloads/phyloP20way.wigFix.gz \ | wigEncode stdin phyloP20way.wig phyloP20way.wib) # Converted stdin, upper limit 3.49, lower limit -8.55 # real 12m5.611s du -hsc *.wi? # 2.0G phyloP20way.wib # 239M phyloP20way.wig # Load gbdb and database with wiggle. ln -s `pwd`/phyloP20way.wib /gbdb/rn6/multiz20way/phyloP20way.wib time hgLoadWiggle -pathPrefix=/gbdb/rn6/multiz20way rn6 \ phyloP20way phyloP20way.wig # real 0m27.237s # use to set trackDb.ra entries for wiggle min and max # and verify table is loaded correctly wigTableStats.sh rn6 phyloP20way # db.table min max mean count sumData # rn6.phyloP20way -8.546 3.488 0.108077 2078464015 2.24635e+08 # stdDev viewLimits # 0.723885 viewLimits=-3.51135:3.488 # that range is: 8.546+3.488 = 12.034 for hBinSize=0.012034 # Create histogram to get an overview of all the data time hgWiggle -doHistogram \ -hBinSize=0.012034 -hBinCount=1000 -hMinVal=-8.546 -verbose=2 \ -db=rn6 phyloP20way > histogram.data 2>&1 # real 1m52.632s # find the Y range for the 2:5 graph grep -v chrom histogram.data | grep "^[0-9]" | ave -col=5 stdin \ | sed -e 's/^/# /;' # Q1 0.000000 # median 0.000045 # Q3 0.000649 # average 0.001104 # min 0.000000 # max 0.052600 # count 906 # total 0.999990 # standard deviation 0.003101 # find the X range for the 2:5 graph grep "^[0-9]" histogram.data | ave -col=2 stdin \ | sed -e 's/^/# /;' # Q1 -4.695120 # median -1.969420 # Q3 0.756282 # average -1.979713 # min -8.546000 # max 3.475970 # count 906 # total -1793.619961 # standard deviation 3.167347 # create plot of histogram: printf 'set terminal png small x000000 xffffff xc000ff x66ff66 xffff00 x00ffff font \ "/usr/share/fonts/default/Type1/n022004l.pfb" set size 1.4, 0.8 set key left box set grid noxtics set grid ytics set title " Rat rn6 Histogram phyloP20way track" set xlabel " phyloP20way score" set ylabel " Relative Frequency" set y2label " Cumulative Relative Frequency (CRF)" set y2range [0:1] set y2tics set xtics set xrange [-2:2] set yrange [0:0.053] plot "histogram.data" using 2:5 title " RelFreq" with impulses, \ "histogram.data" using 2:7 axes x1y2 title " CRF" with lines ' | gnuplot > histo.png display histo.png & ############################################################################# # hgPal downloads (DONE - 2017-01-24 - Hiram) # FASTA from 20-way for ensGene ssh hgwdev screen -S rn6HgPal mkdir /hive/data/genomes/rn6/bed/multiz20way/pal cd /hive/data/genomes/rn6/bed/multiz20way/pal cat ../species.list | tr '[ ]' '[\n]' > order.list # beware large contig count assemblies, this can take a long time (hours) export mz=multiz20way export gp=ensGene export db=rn6 export I=0 export D=0 mkdir exonAA exonNuc printf '#!/bin/sh\n' > $gp.jobs time for C in `sort -nk2 ../../../chrom.sizes | cut -f1` do I=`echo $I | awk '{print $1+1}'` D=`echo $D | awk '{print $1+1}'` dNum=`echo $D | awk '{printf "%03d", int($1/1000)}'` mkdir -p exonNuc/${dNum} > /dev/null mkdir -p exonAA/${dNum} > /dev/null echo "mafGene -chrom=$C -exons -noTrans $db $mz $gp order.list stdout | gzip -c > exonNuc/${dNum}/$C.exonNuc.fa.gz &" echo "mafGene -chrom=$C -exons $db $mz $gp order.list stdout | gzip -c > exonAA/${dNum}/$C.exonAA.fa.gz &" if [ $I -gt 16 ]; then echo "date" echo "wait" I=0 fi done >> $gp.jobs # real 0m21.771s echo "date" >> $gp.jobs echo "wait" >> $gp.jobs chmod +x $gp.jobs time (./$gp.jobs) > $gp.jobs.log 2>&1 & # real 28m28.221s export mz=multiz20way export gp=ensGene time find ./exonAA -type f | grep exonAA.fa.gz | xargs zcat \ | gzip -c > $gp.$mz.exonAA.fa.gz # real 0m29.606s time find ./exonNuc -type f | grep exonNuc.fa.gz | xargs zcat \ | gzip -c > $gp.$mz.exonNuc.fa.gz # real 2m27.672s # -rw-rw-r-- 1 133052531 Jan 24 08:13 ensGene.multiz20way.exonAA.fa.gz # -rw-rw-r-- 1 231318306 Jan 24 08:16 ensGene.multiz20way.exonNuc.fa.gz export mz=multiz20way export gp=ensGene export db=rn6 export pd=/usr/local/apache/htdocs-hgdownload/goldenPath/$db/$mz/alignments mkdir -p $pd md5sum *.fa.gz > md5sum.txt ln -s `pwd`/$gp.$mz.exonAA.fa.gz $pd/$gp.exonAA.fa.gz ln -s `pwd`/$gp.$mz.exonNuc.fa.gz $pd/$gp.exonNuc.fa.gz ln -s `pwd`/md5sum.txt $pd/ rm -rf exonAA exonNuc ############################################################################# # construct download files for 20-way (DONE - 2017-01-24 - Hiram) mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/rn6/multiz20way mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/rn6/phastCons20way mkdir /usr/local/apache/htdocs-hgdownload/goldenPath/rn6/phyloP20way mkdir /hive/data/genomes/rn6/bed/multiz20way/downloads cd /hive/data/genomes/rn6/bed/multiz20way/downloads printf '#!/bin/sh for db in `cat ../species.list` do orgName=`hgsql -N -e "select organism from dbDb where name=\\"${db}\\";" hgcentraltest` sciName=`hgsql -N -e "select scientificName from dbDb where name=\\"${db}\\";" hgcentraltest` genome=`hgsql -N -e "select genome from dbDb where name=\\"${db}\\";" hgcentraltest` descr=`hgsql -N -e "select description from dbDb where name=\\"${db}\\";" hgcentraltest` printf \"%%-15s %%-24s %%-20s\\n\" \"$genome\" \"$sciName\" \"$descr\" done ' > docInfo.sh chmod +x docInfo.sh ./docInfo.sh # Rat Rattus norvegicus Jul. 2014 (RGSC 6.0/rn6) # Mouse Mus musculus Dec. 2011 (GRCm38/mm10) # Prairie vole Microtus ochrogaster Oct 2012 (MicOch1.0/micOch1) # Guinea pig Cavia porcellus Feb. 2008 (Broad/cavPor3) # Rabbit Oryctolagus cuniculus Apr. 2009 (Broad/oryCun2) # Human Homo sapiens Dec. 2013 (GRCh38/hg38) # Chimp Pan troglodytes May 2016 (Pan_tro 3.0/panTro5) # Rhesus Macaca mulatta Nov. 2015 (BCM Mmul_8.0.1/rheMac8) # Tarsier Tarsius syrichta Sep. 2013 (Tarsius_syrichta-2.0.1/tarSyr2) # Dog Canis lupus familiaris Sep. 2011 (Broad CanFam3.1/canFam3) # Panda Ailuropoda melanoleuca Dec. 2009 (BGI-Shenzhen 1.0/ailMel1) # Cat Felis catus Nov. 2014 (ICGSC Felis_catus_8.0/felCat8) # Cow Bos taurus Jun. 2014 (Bos_taurus_UMD_3.1.1/bosTau8) # Opossum Monodelphis domestica Oct. 2006 (Broad/monDom5) # Platypus Ornithorhynchus anatinus Feb. 2007 (ASM227v2/ornAna2) # Chicken Gallus gallus Dec 2015 (Gallus_gallus-5.0/galGal5) # Turkey Meleagris gallopavo Nov. 2014 (Turkey_5.0/melGal5) # X. tropicalis Xenopus tropicalis Sep. 2012 (JGI 7.0/xenTro7) # Zebrafish Danio rerio Sep. 2014 (GRCz10/danRer10) # Elephant shark Callorhinchus milii Dec. 2013 (Callorhinchus_milii-6.1.3/calMil1) printf '#!/bin/sh for db in `cat ../species.list` do orgName=`hgsql -N -e "select organism from dbDb where name=\\"${db}\\";" hgcentraltest` sciName=`hgsql -N -e "select scientificName from dbDb where name=\\"${db}\\";" hgcentraltest` genome=`hgsql -N -e "select genome from dbDb where name=\\"${db}\\";" hgcentraltest` descr=`hgsql -N -e "select description from dbDb where name=\\"${db}\\";" hgcentraltest` printf "<tr align=left><td>%%s</td><td>%%s</td><td>\\n %%s</td><td><a href=\\"../cgi-bin/hgTracks?db=%%s\\"\\n target=_blank>%%s</a></td><td>%%s</td></tr>\\n" "$genome" "$sciName" "$descr" "${db}" "$descr" "MAF Net" done ' > descrPageInfo.sh chmod +x descrPageInfo.sh # use this to add the table in the description page # it doesn't check the maf type, that needs to be fixed mkdir multiz20way phastCons20way phyloP20way cd multiz20way time cp -p ../../anno/rn6.20way.maf . # real 1m13.217s # -rw-rw-r-- 1 48341214170 Jan 22 10:39 rn6.20way.maf du -hsc * # 46G rn6.20way.maf time gzip *.maf # real 69m38.345s # -rw-rw-r-- 1 7725464919 Jan 22 10:39 rn6.20way.maf.gz du -hsc *.maf.gz # 7.2G rn6.20way.maf.gz grep TREE ../../4d/all.mod | awk '{print $NF}' \ | ~/kent/src/hg/utils/phyloTrees/asciiTree.pl /dev/stdin \ > rn6.20way.nh ~/kent/src/hg/utils/phyloTrees/commonNames.sh rn6.20way.nh \ | ~/kent/src/hg/utils/phyloTrees/asciiTree.pl /dev/stdin \ > rn6.20way.commonNames.nh ~/kent/src/hg/utils/phyloTrees/scientificNames.sh rn6.20way.nh \ | $HOME/kent/src/hg/utils/phyloTrees/asciiTree.pl /dev/stdin \ > rn6.20way.scientificNames.nh du -hsc *.maf.gz ../../anno/rn6.20way.maf # 7.2G rn6.20way.maf.gz # 46G ../../anno/rn6.20way.maf ### construct the upstream MAF files for ensGene # bash script #!/bin/sh export geneTbl="ensGene" for S in 1000 2000 5000 do echo "making upstream${S}.maf" featureBits rn6 ${geneTbl}:upstream:${S} -fa=/dev/null -bed=stdout \ | perl -wpe 's/_up[^\t]+/\t0/' | sort -k1,1 -k2,2n \ | /cluster/bin/$MACHTYPE/mafFrags rn6 multiz20way \ stdin stdout \ -orgs=/hive/data/genomes/rn6/bed/multiz20way/species.list \ | gzip -c > upstream${S}.${geneTbl}.maf.gz echo "done upstream${S}.${geneTbl}.maf.gz" done # real 18m28.217s time md5sum *.nh *.maf.gz > md5sum.txt # real 0m26.660s # obtain the README.txt from galVar1/multiz5way and update for this # situation ln -s `pwd`/*.nh `pwd`/*.maf.gz `pwd`/*.txt \ /usr/local/apache/htdocs-hgdownload/goldenPath/rn6/multiz20way ##################################################################### cd /hive/data/genomes/rn6/bed/multiz20way/downloads/phastCons20way ln -s ../../cons/all/downloads/phastCons20way.wigFix.gz \ ./rn6.phastCons20way.wigFix.gz ln -s ../../cons/all/phastCons20way.bw ./rn6.phastCons20way.bw ln -s ../../cons/all/all.mod ./rn6.phastCons20way.mod time md5sum *.gz *.mod *.bw > md5sum.txt # real 0m20.354s # obtain the README.txt from galVar1/phastCons5way and update for this # situation ln -s `pwd`/* \ /usr/local/apache/htdocs-hgdownload/goldenPath/rn6/phastCons20way ##################################################################### cd /hive/data/genomes/rn6/bed/multiz20way/downloads/phyloP20way ln -s ../../consPhyloP/all/downloads/phyloP20way.wigFix.gz \ ./rn6.phyloP20way.wigFix.gz ln -s ../../consPhyloP/run.phyloP/all.mod rn6.phyloP20way.mod ln -s ../../consPhyloP/all/phyloP20way.bw rn6.phyloP20way.bw time md5sum *.mod *.bw *.gz > md5sum.txt # real 0m29.648s # obtain the README.txt from galVar1/phyloP5way and update for this # situation ln -s `pwd`/* \ /usr/local/apache/htdocs-hgdownload/goldenPath/rn6/phyloP20way ############################################################################# # wiki page for 20-way (DONE - 2017-01-24 - Hiram) mkdir /hive/users/hiram/bigWays/rn6.20way cd /hive/users/hiram/bigWays echo "rn6" > rn6.20way/ordered.list awk '{print $1}' /hive/data/genomes/rn6/bed/multiz20way/20way.distances.txt \ >> rn6.20way/ordered.list # sizeStats.sh catches up the cached measurements required for data # in the tables. They are usually already mostly done, only new # assemblies will have updates. ./sizeStats.sh rn6.20way/ordered.list # dbDb.sh constructs rn6.20way/Rn6_20-way_conservation_alignment.html # may need to add new assembly references to srcReference.list and # urlReference.list ./dbDb.sh rn6 20way # sizeStats.pl constructs rn6.20way/Rn6_20-way_Genome_size_statistics.html # this requires entries in coverage.list for new sequences ./sizeStats.pl rn6 20way # defCheck.pl constructs Rn6_20-way_conservation_lastz_parameters.html ./defCheck.pl rn6 20way # this constructs the html pages in rn6.20way/: # -rw-rw-r-- 1 3317 May 11 11:08 Rn6_20-way_conservation_alignment.html # -rw-rw-r-- 1 4827 May 11 11:08 Rn6_20-way_Genome_size_statistics.html # -rw-rw-r-- 1 3308 May 11 11:08 Rn6_20-way_conservation_lastz_parameters.html # add those pages to the genomewiki. Their page names are the # names of the .html files without the .html: # Rn6_20-way_conservation_alignment # Rn6_20-way_Genome_size_statistics # Rn6_20-way_conservation_lastz_parameters # when you view the first one you enter, it will have links to the # missing two. ############################################################################ # composite chainNet trackDb (DONE - 2017-01-27 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/rat/rn6 printf 'mm10 micOch1 cavPor3 oryCun2 ' > glires.list printf 'hg38 panTro5 tarSyr2 rheMac8 ' > primates.list printf 'canFam3 felCat8 ailMel1 bosTau8 ' > laurasiatheria.list printf 'monDom5 ornAna2 xenTro7 ' > sarcopterygii.list printf 'galGal5 melGal5 ' > birds.list printf 'calMil1 danRer10' > fish.list $HOME/kent/src/hg/utils/phyloTrees/chainNetCompositeTrackDb.pl \ vertebrate glires.list primates.list \ laurasiatheria.list sarcopterygii.list birds.list fish.list \ > vertebrateChainNet.ra # edit that result to fixup the shortLabel, longLabel and default # visibilities desired on some of the tracks # insert this line in the trackDb.ra to get it included: include vertebrateChainNet.ra ############################################################################ # crispr 10K track (DONE - 2018-07-02 - Hiram) mkdir -p /hive/data/genomes/rn6/bed/crispr.10K cd /hive/data/genomes/rn6/bed/crispr.10K # since this time, the indexing has been built into the primary script export PATH=/hive/data/outside/crisprTrack/crispor/tools/usrLocalBin:$PATH export TMPDIR=/dev/shm time (/hive/data/outside/crisprTrack/crispor/tools/crisprAddGenome ucscLocal \ rn6 --geneTable ensGene \ --baseDir /hive/data/outside/crisprTrack/crispor/genomes) > index.log 2>&1 # real 109m47.348s time (/cluster/home/hiram/kent/src/hg/utils/automation/doCrispr.pl \ -buildDir=`pwd` -bigClusterHub=ku -smallClusterHub=ku \ -continue=guides -stop=guides -workhorse=hgwdev rn6 ensGene) > guides.log 2>&1 # real 18m23.512s # step specScores were run manually to test fixes in the script to avoid # the bottleneck /cluster/home/hiram/kent/src/hg/utils/automation/doCrispr.pl \ -buildDir=`pwd` -bigClusterHub=ku -smallClusterHub=ku -debug \ -continue=specScores -stop=specScores -workhorse=hgwdev rn6 ensGene /cluster/home/hiram/kent/src/hg/utils/automation/doCrispr.pl -debug \ -buildDir=`pwd` -bigClusterHub=ku -smallClusterHub=ku \ -continue=effScores -stop=offTargets -workhorse=hgwdev rn6 ensGene) /cluster/home/hiram/kent/src/hg/utils/automation/doCrispr.pl -debug \ -buildDir=`pwd` -bigClusterHub=ku -smallClusterHub=ku \ -continue=load -stop=load -workhorse=hgwdev rn6 ensGene ############################################################################ +# LIFTOVER TO regenRn0 (DONE - 2019-06-29 - Hiram) + ssh hgwdev + mkdir /hive/data/genomes/rn6/bed/blat.regenRn0.2019-06-29 + cd /hive/data/genomes/rn6/bed/blat.regenRn0.2019-06-29 + doSameSpeciesLiftOver.pl -verbose=2 \ + -debug -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ + -ooc=/hive/data/genomes/rn6/jkStuff/rn6.11.ooc \ + rn6 regenRn0 + time (doSameSpeciesLiftOver.pl -verbose=2 \ + -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ + -ooc=/hive/data/genomes/rn6/jkStuff/rn6.11.ooc \ + rn6 regenRn0) > doLiftOverToRegenRn0.log 2>&1 + # real 249m30.284s + + # see if the liftOver menus function in the browser from regenRn0 to galGal5 + + # would like to see this as a track: + chainToPsl rn6ToRegenRn0.over.chain.gz ../../chrom.sizes \ + /hive/data/genomes/regenRn0/chrom.sizes ../../rn6.2bit \ + /hive/data/genomes/regenRn0/regenRn0.2bit rn6ToRegenRn0.psl + +######################################################################### +# LIFTOVER TO regenRn1 (DONE - 2019-09-09 - Hiram) + ssh hgwdev + mkdir /hive/data/genomes/rn6/bed/blat.regenRn1.2019-09-09 + cd /hive/data/genomes/rn6/bed/blat.regenRn1.2019-09-09 + doSameSpeciesLiftOver.pl -verbose=2 \ + -debug -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ + -ooc=/hive/data/genomes/rn6/jkStuff/rn6.11.ooc \ + rn6 regenRn1 + time (doSameSpeciesLiftOver.pl -verbose=2 \ + -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ + -ooc=/hive/data/genomes/rn6/jkStuff/rn6.11.ooc \ + rn6 regenRn1) > doLiftOverToRegenRn1.log 2>&1 + # real 252m38.421s + + # see if the liftOver menus function in the browser from rn6 to regenRn1 + + # would like to see this as a track: + hgLoadChain rn6 chainRegenRn1 rn6ToRegenRn1.over.chain.gz + # Loading 237485 chains into rn6.chainRegenRn1 + + chainSort rn6ToRegenRn1.over.chain.gz stdout \ + | chainPreNet stdin \ + /hive/data/genomes/rn6/chrom.sizes \ + /hive/data/genomes/regenRn1/chrom.sizes stdout \ + | chainNet stdin -minSpace=1 /hive/data/genomes/rn6/chrom.sizes \ + /hive/data/genomes/regenRn1/chrom.sizes stdout /dev/null \ + | netSyntenic stdin noClass.net + + netClass -verbose=0 -noAr noClass.net rn6 regenRn1 rn6.regenRn1.net + + netFilter -minGap=10 rn6.regenRn1.net \ + | hgLoadNet -verbose=0 rn6 netRegenRn1 stdin + +featureBits rn6 chainRegenRn1Link > fb.rn6.chainRegenRn1Link.txt 2>&1 + + cat fb.rn6.chainRegenRn1Link.txt + # 2265178557 bases of 2729860805 (82.978%) in intersection + +#########################################################################