# for emacs: -*- mode: sh; -*- # This file describes browser build for the rhiBie1 ######################################################################### # obtain photograph: (DONE - 2017-02-28 - Hiram) # per email request to photographer, obtained photograph via email: # -rw-r--r-- 1 1371219 Feb 28 06:47 Copyright_JackyPoon_black_snub-nosed_monkey.jpg mkdir /hive/data/genomes/rhiBie1/photo cd /hive/data/genomes/rhiBie1/photo convert -quality 80 -sharpen 0 -normalize -geometry 400x400 \ Copyright_JackyPoon_black_snub-nosed_monkey.jpg Rhinopithecus_bieti.jpg # check that into the source tree kent/src/hg/htdocs/images/ git commit -m '(C) Copyright Jacky Poon, all rights reserved - http://www.jackypoon.org/mystery-monkeys refs #18785' \ Rhinopithecus_bieti.jpg # and copy to /usr/local/apache/htdocs/images/ cd /hive/data/genomes/rhiBie1 # establish photoReference.txt print 'photoCreditURL http://www.jackypoon.org/mystery-monkeys photoCreditName (C) Copyright Jacky Poon, all rights reserved ' > photoReference.txt ######################################################################### # Initial steps (DONE - 2017-02-28 - Hiram) # To start this initialBuild.txt document, from a previous assembly document: mkdir ~/kent/src/hg/makeDb/doc/rhiBie1 cd ~/kent/src/hg/makeDb/doc/rhiBie1 # best to use a most recent document since it has the latest features and # procedures: sed -e 's/macNem/rhiBie/g; s/MacNem/RhiBie/g; s/DONE/TBD/g;' \ ../macNem1/initialBuild.txt > initialBuild.txt mkdir /hive/data/genomes/rhiBie1/refseq cd /hive/data/genomes/rhiBie1/refseq rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Rhinopithecus_bieti/all_assembly_versions/GCF_001698545.1_ASM169854v1/ ./ # sent 562 bytes received 3295813018 bytes 18158752.51 bytes/sec # total size is 3295407913 speedup is 1.00 # real 3m0.443s # appears to be a scaffold-only assembly: faSize G*v1_genomic.fna.gz # 2977074946 bases (175194759 N's 2801880187 real 1797410004 upper 1004470183 lower) in 105032 sequences in 1 files # Total size: mean 28344.5 sd 282015.2 min 200 (NW_016906396.1) max 14887386 (NW_016806642.1) median 912 # %33.74 masked total, %35.85 masked real # this information is from the top of # rhiBie1/refseq/GCF_001698545.1_ASM169854v1_assembly_report.txt # Assembly name: ASM169854v1 # Organism name: Rhinopithecus bieti (black snub-nosed monkey) # Isolate: Rb0 # Sex: male # Taxid: 61621 # BioSample: SAMN02786331 # BioProject: PRJNA339282 # Submitter: Laboratory for conservation and utilization of Bio-resource # Date: 2016-8-3 # Assembly type: haploid # Release type: major # Assembly level: Scaffold # Genome representation: full # WGS project: MCGX01 # Assembly method: Newbler v. 71; Anytag v. 70; GOBOND v. 69; GapCloser v. 1.12 # Expected final version: Yes # Genome coverage: 76.6x # Sequencing technology: Illumina HiSeq # RefSeq category: Representative Genome # GenBank assembly accession: GCA_001698545.1 # RefSeq assembly accession: GCF_001698545.1 # RefSeq assembly and GenBank assemblies identical: no # ## Assembly-Units: ## GenBank Unit Accession RefSeq Unit Accession Assembly-Unit name ## GCA_001698555.1 GCF_001698555.1 Primary Assembly ## GCF_001705985.1 non-nuclear ############################################################################# # establish config.ra file (DONE - Hiram - 2017-02-28) # arguments here are: cd /hive/data/genomes/rhiBie1 $HOME/kent/src/hg/utils/automation/prepConfig.pl rhiBie1 mammal \ rhinopithecus ./refseq/*_assembly_report.txt > rhiBie1.config.ra # verify it looks sane cat rhiBie1.config.ra # config parameters for makeGenomeDb.pl: db rhiBie1 clade mammal genomeCladePriority 35 scientificName Rhinopithecus bieti commonName Black snub-nosed monkey assemblyDate Aug. 2016 assemblyLabel Laboratory for conservation and utilization of Bio-resource assemblyShortLabel ASM169854v1 orderKey 2534 # mitochondrial sequence included in refseq release # mitoAcc NC_015486.1 mitoAcc none fastaFiles /hive/data/genomes/rhiBie1/ucsc/*.fa.gz agpFiles /hive/data/genomes/rhiBie1/ucsc/*.agp # qualFiles none dbDbSpeciesDir rhinopithecus photoCreditURL http://www.jackypoon.org/mystery-monkeys photoCreditName (C) Copyright Jacky Poon, all rights reserved ncbiGenomeId 9846 ncbiAssemblyId 777531 ncbiAssemblyName ASM169854v1 ncbiBioProject 339282 ncbiBioSample SAMN02786331 genBankAccessionID GCF_001698545.1 taxId 61621 ############################################################################# # setup UCSC named files (DONE - 2017-02-28 - Hiram) mkdir /hive/data/genomes/rhiBie1/ucsc cd /hive/data/genomes/rhiBie1/ucsc # check for duplicate sequences: time faToTwoBit -noMask ../refseq/G*v1_genomic.fna.gz refseq.2bit # real 1m10.263s twoBitDup refseq.2bit # NW_016884136.1 and NW_016884133.1 are identical # no output is a good result, otherwise, would have to eliminate duplicates # the scripts creating the fasta here will be using this refseq.2bit file # place it in a list for removal echo NW_016884136.1 > exclude.list # simple unplaced_scaffolds, just transform their names: # remove the duplicate contig faSomeRecords -exclude ../refseq/G*v1_assembly_structure/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fna.gz exclude.list stdout \ | sed -e 's/.1 Rhinopithecus bieti .*/v1/;' | gzip -c > unplaced.fa.gz # real 1m21.170s zcat ../refseq/G*v1_assembly_structure/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \ | grep -v NW_016884136 | sed -e 's/.1\t/v1\t/;' > unplaced.agp # there is a chrMT in this assembly: zcat ../refseq/GCF_001698545.1_ASM169854v1_assembly_structure/non-nuclear/assembled_chromosomes/FASTA/chrMT.fna.gz \ | sed -e 's/^>.*/>chrM/;' | gzip > chrM.fa.gz zcat ../refseq/GCF_001698545.1_ASM169854v1_assembly_structure/non-nuclear/assembled_chromosomes/AGP/chrMT.comp.agp.gz \ | sed -e 's/^NC_015486.1/chrM/;' > chrM.agp # verify fasta and AGPs agree time faToTwoBit unplaced.fa.gz chrM.fa.gz test.2bit cat unplaced.agp chrM.agp \ | checkAgpAndFa stdin test.2bit 2>&1 | tail -4 # All AGP and FASTA entries agree - both files are valid # real 0m16.654s # and no sequence lost from orginal: # (well except for the one dropped contig) twoBitToFa test.2bit stdout | faSize stdin # 2977074741 bases (175194759 N's 2801879982 real 2801879982 upper 0 lower) in 105031 sequences in 1 files # Total size: mean 28344.7 sd 282016.5 min 200 (NW_016906396v1) max 14887386 (NW_016806642v1) median 912 # same numbers as above, minus the one contig (205 bases) # 2977074946 bases (175194759 N's 2801880187 real 1797410004 upper 1004470183 lower) in 105032 sequences in 1 files # Total size: mean 28344.5 sd 282015.2 min 200 (NW_016906396.1) max 14887386 (NW_016806642.1) median 912 # no longer need these temporary 2bit files rm refseq.2bit unplaced.2bit test.2bit ############################################################################# # Initial database build (DONE - 2017-02-28 - Hiram) # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp rhiBie1.config.ra) > agp.log 2>&1 # real 3m32.248s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db rhiBie1.config.ra) > db.log 2>&1 # real 23m23.025s # check in the trackDb files created in TemporaryTrackDbCheckout/ # and add rhiBie1 to trackDb/makefile # temporary symlink until masked sequence is available cd /hive/data/genomes/rhiBie1 ln -s `pwd`/rhiBie1.unmasked.2bit /gbdb/rhiBie1/rhiBie1.2bit ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2017-02-28 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/cpgIslandsUnmasked cd /hive/data/genomes/rhiBie1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/rhiBie1/rhiBie1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku rhiBie1) > do.log 2>&1 # real 17m31.456s cat fb.rhiBie1.cpgIslandExtUnmasked.txt # 35321721 bases of 2977074741 (1.186%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2017-02-28 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/cytoBand cd /hive/data/genomes/rhiBie1/bed/cytoBand makeCytoBandIdeo.csh rhiBie1 ######################################################################### # ucscToINSDC and ucscToRefSeq table/track (DONE - 2017-03-02 - Hiram) # the sequence here is working for a 'refseq' assembly # beware of a chrM situation may be specific depending upon what is # available in the assembly mkdir /hive/data/genomes/rhiBie1/bed/ucscToINSDC cd /hive/data/genomes/rhiBie1/bed/ucscToINSDC # find accession for chrM grep chrM ../../rhiBie1.agp # chrM 1 16551 1 O NC_015486.1 1 16551 + # find the genbank accession for NC_026976.1 at Entrez nucleotide # The NC_015486.1 name is the RefSeq name, the genbank name is: KP765688.1 # the assembly_report does not have this KP name since the chrM sequence # is not in the genbank assembly: grep NC_015486 ../../refseq/GCF*v1_assembly_report.txt # MT assembled-molecule MT Mitochondrion na <> NC_015486.1 non-nuclear 16551 na # if there is a chrM, use its INSDC name as a second argument: # this is a RefSeq assembly, use the chrM refSeq name: ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../refseq/GCF_*structure/Primary_Assembly NC_015486.1 # this is actually ucscToRefSeq since this is a RefSeq assembly sort -k2 ucscToINSDC.txt > ucscToRefSeq.txt awk '{printf "%s\t%s\n", $2, $1}' ucscToRefSeq.txt \ | sort > refSeqToUcsc.txt # chrM processing needs special help, fixup with the sed # extract the refseq vs. genbank names from the assembly_report # columns 5 and 7 are the INSDC and RefSeq names grep -v "^#" ../../refseq/GCF*_assembly_report.txt | cut -f5,7 \ | awk '{printf "%s\t%s\n", $2, $1}' | sed -e 's/na/KP765688.1/' \ | sort > refseq.insdc.txt awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > ucsc.coordinate.tab join -2 2 refseq.insdc.txt ucscToRefSeq.txt | tr '[ ]' '[\t]' | sort -k3 \ | join -2 3 ucsc.coordinate.tab - | tr '[ ]' '[\t]' | cut -f1-4 \ > ucscToRefSeq.bed join -2 2 refseq.insdc.txt ucscToRefSeq.txt | tr '[ ]' '[\t]' | sort -k3 \ | join -2 3 ucsc.coordinate.tab - | tr '[ ]' '[\t]' | cut -f1-3,5 \ > ucscToINSDC.bed # verify chrM is correct: grep chrM *.bed # ucscToINSDC.bed:chrM 0 16551 KP765688.1 # ucscToRefSeq.bed:chrM 0 16551 NC_015486.1 # should be same line counts throughout: # in this case one is missing in the final result due to the duplicate # contig being removed wc -l * # 105032 refSeqToUcsc.txt # 105032 refseq.insdc.txt # 105031 ucsc.coordinate.tab # 105031 ucscToINSDC.bed # 105032 ucscToINSDC.txt # 105031 ucscToRefSeq.bed # 105032 ucscToRefSeq.txt cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1 # 14 # use the 14 in this sed sed -e "s/21/14/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | hgLoadSqlTab rhiBie1 ucscToINSDC stdin ucscToINSDC.bed # should be the same for ucscToRefSeq: export chrSize=`cut -f1 ucscToRefSeq.bed | awk '{print length($0)}' | sort -n | tail -1` echo $chrSize # 14 sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ | sed -e 's/INSDC/RefSeq/g;' > ucscToRefSeq.sql hgLoadSqlTab rhiBie1 ucscToRefSeq ./ucscToRefSeq.sql ucscToRefSeq.bed # checkTableCoords should be silent checkTableCoords rhiBie1 # each should cover %100 entirely: featureBits -countGaps rhiBie1 ucscToINSDC # 2977074741 bases of 2977074741 (100.000%) in intersection featureBits -countGaps rhiBie1 ucscToRefSeq # 2977074741 bases of 2977074741 (100.000%) in intersection ######################################################################### # add chromAlias table (DONE - 2017-03-02 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/chromAlias cd /hive/data/genomes/rhiBie1/bed/chromAlias hgsql -N -e 'select chrom,name,"refseq" from ucscToRefSeq;' rhiBie1 \ > ucsc.refseq.tab hgsql -N -e 'select chrom,name,"genbank" from ucscToINSDC;' rhiBie1 \ > ucsc.genbank.tab awk '{printf "%s\t%s\t%s\n", $2,$1,$3}' ucsc.genbank.tab ucsc.refseq.tab \ | sort > rhiBie1.chromAlias.tab hgLoadSqlTab rhiBie1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ rhiBie1.chromAlias.tab cd /hive/data/genomes/rhiBie1/bed/chromAlias # add ensembl names 2017-12-14 mkdir previous mv *.tab previous join -t$'\t' ../idKeys/rhiBie1.idKeys.txt \ ../../ensembl/ensemblRhiBie1.idKeys.txt \ | cut -f2,3 | sort > ucsc.ensembl.tab # there is a duplicate in ensembl sequence # MCGX01081313.1 and MCGX01081310.1 are identical # NW_016884133v1 MCGX01081310.1 # NW_016884133v1 MCGX01081313.1 cut -f1,2 previous/ucsc.refseq.tab > ucsc.refseq.tab cut -f1,2 previous/ucsc.genbank.tab > ucsc.genbank.tab ~/kent/src/hg/utils/automation/chromAlias.pl sort -o rhiBie1.chromAlias.tab rhiBie1.chromAlias.tab for t in refseq genbank ensembl do c0=`cat ucsc.$t.tab | wc -l` c1=`grep $t rhiBie1.chromAlias.tab | wc -l` ok="OK" if [ "$c0" -ne "$c1" ]; then ok="ERROR" fi printf "# checking $t: $c0 =? $c1 $ok\n" done # checking refseq: 105031 =? 105031 OK # checking genbank: 105031 =? 105031 OK # checking ensembl: 105032 =? 105032 OK hgLoadSqlTab rhiBie1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ rhiBie1.chromAlias.tab ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2017-03-02 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/rhinopithecus/rhiBie1 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" rhiBie1 \ | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c 105030 MCGX.1 1 NC_.1 # implies a rule: '[MN]C[G_][X0-9]+(\.[0-9]+)?' # verify this rule will find them all and eliminate them all: hgsql -N -e "select frag from gold;" rhiBie1 | wc -l # 105031 hgsql -N -e "select frag from gold;" rhiBie1 \ | egrep -e '[MN]C[G_][X0-9]+(\.[0-9]+)?' | wc -l # 105031 hgsql -N -e "select frag from gold;" rhiBie1 \ | egrep -v -e '[MN]C[G_][X0-9]+(\.[0-9]+)?' | wc -l # 0 # hence, add to trackDb/chicken/rhiBie1/trackDb.ra searchTable gold shortCircuit 1 termRegex [MN]C[G_][X0-9]+(\.[0-9]+)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 # verify searches work in the position box ########################################################################## # running repeat masker (DONE - 2017-02-28 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/repeatMasker cd /hive/data/genomes/rhiBie1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku rhiBie1) > do.log 2>&1 # real 662m1.726s cat faSize.rmsk.txt # 2977074741 bases (175194759 N's 2801879982 real 1394937761 upper # 1406942221 lower) in 105031 sequences in 1 files # Total size: mean 28344.7 sd 282016.5 min 200 (NW_016906396v1) # max 14887386 (NW_016806642v1) median 912 # %47.26 masked total, %50.21 masked real egrep -i "versi|relea" do.log # RepeatMasker version open-4.0.5 # January 31 2015 (open-4-0-5) version of RepeatMasker # CC RELEASE 20140131; * time featureBits -countGaps rhiBie1 rmsk # 1407426392 bases of 2977074741 (47.275%) in intersection # real 2m3.210s # why is it different than the faSize above ? # because rmsk masks out some N's as well as bases, the faSize count above # separates out the N's from the bases, it doesn't show lower case N's # faster way to get the same result on high contig count assemblies: time hgsql -N -e 'select genoName,genoStart,genoEnd from rmsk;' rhiBie1 \ | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" # total 1407426392.000000 # real 0m41.063s ########################################################################## # running simple repeat (DONE - 2017-02-28 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/simpleRepeat cd /hive/data/genomes/rhiBie1/bed/simpleRepeat # using trf409 3 here guessing smaller genome (human == 6) time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ -trf409 6 rhiBie1) > do.log 2>&1 # real 64m29.232s cat fb.simpleRepeat # 73021659 bases of 2977074741 (2.453%) in intersection # adding this trfMask to the other masking cd /hive/data/genomes/rhiBie1 # when using the Window Masker result: # twoBitMask bed/windowMasker/rhiBie1.cleanWMSdust.2bit \ # -add bed/simpleRepeat/trfMask.bed rhiBie1.2bit # you can safely ignore the warning about fields >= 13 # when using Rmsk results, add to rmsk after it is done: twoBitMask rhiBie1.rmsk.2bit \ -add bed/simpleRepeat/trfMask.bed rhiBie1.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa rhiBie1.2bit stdout | faSize stdin > faSize.rhiBie1.2bit.txt cat faSize.rhiBie1.2bit.txt # 2977074741 bases (175194759 N's 2801879982 real 1393352740 upper # 1408527242 lower) in 105031 sequences in 1 files # Total size: mean 28344.7 sd 282016.5 min 200 (NW_016906396v1) # max 14887386 (NW_016806642v1) median 912 # %47.31 masked total, %50.27 masked real # reset the symlink rm /gbdb/rhiBie1/rhiBie1.2bit ln -s `pwd`/rhiBie1.2bit /gbdb/rhiBie1/rhiBie1.2bit ######################################################################### # CREATE MICROSAT TRACK (DONE - 2017-02-28 - Hiram) ssh hgwdev mkdir /cluster/data/rhiBie1/bed/microsat cd /cluster/data/rhiBie1/bed/microsat awk '($5==2 || $5==3) && $6 >= 15 && $8 == 100 && $9 == 0 {printf("%s\t%s\t%s\t%dx%s\n", $1, $2, $3, $6, $16);}' \ ../simpleRepeat/simpleRepeat.bed > microsat.bed hgLoadBed rhiBie1 microsat microsat.bed # Read 31859 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2017-03-01 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/windowMasker cd /hive/data/genomes/rhiBie1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev rhiBie1) > do.log 2>&1 # real 260m40.754s # Masking statistics cat faSize.rhiBie1.cleanWMSdust.txt # 2977074741 bases (175194759 N's 2801879982 real 1782047365 upper # 1019832617 lower) in 105031 sequences in 1 files # Total size: mean 28344.7 sd 282016.5 min 200 (NW_016906396v1) # max 14887386 (NW_016806642v1) median 912 # %34.26 masked total, %36.40 masked real cat fb.rhiBie1.rmsk.windowmaskerSdust.txt # 784491226 bases of 2977074741 (26.351%) in intersection ########################################################################## # run up idKeys files for ncbiRefSeq (DONE - 2017-03-01 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/idKeys cd /hive/data/genomes/rhiBie1/bed/idKeys time (doIdKeys.pl -buildDir=`pwd` rhiBie1) > do.log 2>&1 & # real 45m57.080s cat rhiBie1.keySignature.txt # ad94c2baecdb85a63bd9285d42161b91 ########################################################################## # cpgIslands - (DONE - 2017-03-01 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/cpgIslands cd /hive/data/genomes/rhiBie1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku rhiBie1) > do.log 2>&1 & # real 14m4.248s cat fb.rhiBie1.cpgIslandExt.txt # 20309988 bases of 2977074741 (0.682%) in intersection ############################################################################## # genscan - (DONE - 2017-03-01 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/genscan cd /hive/data/genomes/rhiBie1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku rhiBie1) > do.log 2>&1 & # real 64m46.377s cat fb.rhiBie1.genscan.txt # 52459334 bases of 2977074741 (1.762%) in intersection cat fb.rhiBie1.genscanSubopt.txt # 53805481 bases of 2977074741 (1.807%) in intersection ############################################################################# # augustus gene track (dONE - 2017-03-01 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/augustus cd /hive/data/genomes/rhiBie1/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=human -dbHost=hgwdev -workhorse=hgwdev rhiBie1) > do.log 2>&1 & # reel 104m18.398s cat fb.rhiBie1.augustusGene.txt # 47275232 bases of 2977074741 (1.588%) in intersection ######################################################################### # Create kluster run files (DONE - 2017-03-01 - Hiram) # numerator is rhiBie1 gapless bases "real" as reported by: featureBits -noRandom -noHap rhiBie1 gap # 0 bases of 2977074741 (0.000%) in intersection # ^^^ # denominator is hg19 gapless bases as reported by: # featureBits -noRandom -noHap hg19 gap # 234344806 bases of 2861349177 (8.190%) in intersection # 1024 is threshold used for human -repMatch: calc \( 2977074741 / 2861349177 \) \* 1024 # ( 2977074741 / 2861349177 ) * 1024 = 1065.415070 # ==> use -repMatch=1000 according to size scaled down from 1024 for human. # and rounded down to nearest 50 cd /hive/data/genomes/rhiBie1 blat rhiBie1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/rhiBie1.11.ooc \ -repMatch=1000 # Wrote 25834 overused 11-mers to jkStuff/rhiBie1.11.ooc # check non-bridged gaps to see what the typical size is: hgsql -N \ -e 'select * from gap where bridge="no" order by size;' rhiBie1 \ | sort -k7,7nr | ave -col=7 stdin # there are no gaps at all in this assembly # # all these gap sizes are 100 # # minimum gap size is 100 and produces a reasonable number of lifts # gapToLift -verbose=2 -minGap=10 rhiBie1 jkStuff/nonBridged.lft \ # -bedFile=jkStuff/nonBridged.bed ######################################################################## # GENBANK AUTO UPDATE (DONE - 2017-03-05 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # #organism mrnaCnt estCnt refSeqCnt # Rhinopithecus bieti 9 0 0 # edit etc/genbank.conf to add rhiBie1 just before macFas5 # rhiBie1 - Rhinopithecus bieti - black snub-nosed monkey taxId 61621 rhiBie1.serverGenome = /hive/data/genomes/rhiBie1/rhiBie1.2bit rhiBie1.clusterGenome = /hive/data/genomes/rhiBie1/rhiBie1.2bit rhiBie1.ooc = /hive/data/genomes/rhiBie1/jkStuff/rhiBie1.11.ooc rhiBie1.lift = no rhiBie1.perChromTables = no rhiBie1.downloadDir = rhiBie1 rhiBie1.refseq.mrna.native.pslCDnaFilter = ${ordered.refseq.mrna.native.pslCDnaFilter} rhiBie1.refseq.mrna.xeno.pslCDnaFilter = ${ordered.refseq.mrna.xeno.pslCDnaFilter} rhiBie1.genbank.mrna.native.pslCDnaFilter = ${ordered.genbank.mrna.native.pslCDnaFilter} rhiBie1.genbank.mrna.xeno.pslCDnaFilter = ${ordered.genbank.mrna.xeno.pslCDnaFilter} rhiBie1.genbank.est.native.pslCDnaFilter = ${ordered.genbank.est.native.pslCDnaFilter} # DO NOT NEED genbank.mrna.xeno except for human, mouse # defaults are fine: genbank.mrna.native refseq.mrna.native refseq.mrna.xeno yes # and genbank.est.native # rhiBie1.upstreamGeneTbl = refGene # rhiBie1.upstreamMaf = multiz7way # /hive/data/genomes/melGal4/bed/multiz7way/species.lst git commit -m "Added rhiBie1; refs #18927" etc/genbank.conf git push # update /cluster/data/genbank/: make etc-update # also add to src/lib/gbGenome.c # static char *rhiBieNames[] = {"Macaca nemestrina", NULL}; # {"rhiBie", rhiBieNames}, git commit -m "Added rhiBie1; refs #18958" src/lib/gbGenome.c make install-server cd /cluster/data/genbank time ./bin/gbAlignStep -initial rhiBie1 # logFile: var/build/logs/2017.03.05-06:48:55.rhiBie1.initalign.log # about 13 hours tail -2 2017.03.05-06:48:55.rhiBie1.initalign.log # hgwdev 2017.03.05-19:28:21 rhiBie1.initalign: Succeeded: rhiBie1 # hgwdev 2017.03.05-19:28:53 rhiBie1.initalign: finish # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.rhiBie1 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad rhiBie1 # logFile: var/dbload/hgwdev/logs/2017.03.05-19:52:47.rhiBie1.dbload.log # about 21 minutes tail -1 var/dbload/hgwdev/logs/2017.03.05-19:52:47.rhiBie1.dbload.log # hgwdev 2017.03.05-20:13:03 rhiBie1.dbload: finish # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add rhiBie1 to: # etc/align.dbs etc/hgwdev.dbs git add etc/align.dbs etc/hgwdev.dbs git commit -m 'adding rhiBie1 to the update alignments refs #18958' etc/align.dbs etc/hgwdev.dbs git push make etc-update ############################################################################# # ncbiRefSeq (TBD - 2016-05-13 - Hiram) mkdir /hive/data/genomes/rhiBie1/bed/ncbiRefSeq cd /hive/data/genomes/rhiBie1/bed/ncbiRefSeq # running step wise as this script is still under development time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev \ -stop=download -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.4_Gallus_gallus-5.0 rhiBie1) > download.log 2>&1 # real 16m29.536s time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -continue=process -bigClusterHub=ku -dbHost=hgwdev \ -stop=process -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.4_Gallus_gallus-5.0 rhiBie1) > process.log 2>&1 # real 3m58.858s time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ -continue=load -bigClusterHub=ku -dbHost=hgwdev \ -stop=load -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ refseq vertebrate_other Gallus_gallus \ GCF_000002315.4_Gallus_gallus-5.0 rhiBie1) > load.log 2>&1 # real 0m33.205s cat fb.ncbiRefSeq.rhiBie1.txt # 82563006 bases of 1218501075 (6.776%) in intersection featureBits -enrichment rhiBie1 refGene ncbiRefSeq # refGene 1.181%, ncbiRefSeq 6.776%, both 1.175%, cover 99.49%, # enrich 14.68x ######################################################################### # BLATSERVERS ENTRY (DONE - 2017-11-17 - 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 ("rhiBie1", "blat1a", "17886", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("rhiBie1", "blat1a", "17887", "0", "1");' \ hgcentraltest # test it with some sequence ############################################################################ ## reset default position to MEPE gene (egg shell protein) ## located via blat of the chicken protein ## (TBD - 2017-01-17 - Hiram) ssh hgwdev hgsql -e 'update dbDb set defaultPos="chr4:21251858-21288049" where name="rhiBie1";' hgcentraltest ######################################################################### # all.joiner update, downloads and in pushQ - (DONE - 2017-03-06 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output joinerCheck -database=rhiBie1 -tableCoverage all.joiner joinerCheck -database=rhiBie1 -times all.joiner joinerCheck -database=rhiBie1 -keys all.joiner cd /hive/data/genomes/rhiBie1 time (makeDownloads.pl -workhorse=hgwdev rhiBie1) > downloads.log 2>&1 # real 24m57.372s # now ready for pushQ entry mkdir /hive/data/genomes/rhiBie1/pushQ cd /hive/data/genomes/rhiBie1/pushQ time (makePushQSql.pl rhiBie1) > rhiBie1.pushQ.sql 2> stderr.out # real 11m29.572s # check for errors in stderr.out, some are OK, e.g.: # WARNING: rhiBie1 does not have seq # WARNING: rhiBie1 does not have extFile # WARNING: rhiBie1 does not have estOrientInfo # copy it to hgwbeta scp -p rhiBie1.pushQ.sql qateam@hgwbeta:/tmp/ ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/rhiBie1.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly #########################################################################