# for emacs: -*- mode: sh; -*- # This file describes browser build for the macNem1 ######################################################################### # obtain photograph: (DONE - 2017-02-14 - Hiram) mkdir /hive/data/genomes/macNem1/photo cd /hive/data/genomes/macNem1/photo wget -O photoFile.jpg \ 'https://upload.wikimedia.org/wikipedia/commons/7/79/Cercopithecidae_-_Macaca_nemastrina.jpg' convert -sharpen 0 -normalize -geometry 400x400 photoFile Macaca_nemestrina.png # check that into the source tree kent/src/hg/htdocs/images/ git commit -m ''WikiMedia Commons: Hectonichus https://commons.wikimedia.org/wiki/File:Cercopithecidae_-_Macaca_nemastrina.jpg refs #18785' \ Macaca_nemestrina.png # and copy to /usr/local/apache/htdocs/images/ cd /hive/data/genomes/macNem1 # establish photoReference.txt print 'photoCreditURL https://commons.wikimedia.org/wiki/File:Cercopithecidae_-_Macaca_nemastrina.jpg photoCreditName WikiMedia Commons: Hectonichus ' > photoReference.txt ######################################################################### # Initial steps, find photograph (DONE - 2017-02-23 - Hiram) # To start this initialBuild.txt document, from a previous assembly document: mkdir ~/kent/src/hg/makeDb/doc/macNem1 cd ~/kent/src/hg/makeDb/doc/macNem1 sed -e 's/melGal5/macNem1/g; s/MelGal5/MacNem1/g; s/DONE/TBD/g;' \ ../melGal5/initialBuild.txt > initialBuild.txt mkdir /hive/data/genomes/macNem1/refseq cd /hive/data/genomes/macNem1/refseq time rsync -L -a -P \ rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Macaca_nemestrina/all_assembly_versions/GCF_000956065.1_Mnem_1.0/ ./ # sent 489 bytes received 3389462340 bytes 20357134.11 bytes/sec # total size is 3389046120 speedup is 1.00 # real 2m45.892s # appears to be a scaffold-only assembly: faSize G*0_genomic.fna.gz # 2948703511 bases (110217031 N's 2838486480 real 1832124010 # upper 1006362470 lower) in 9733 sequences in 1 files # Total size: mean 302959.4 sd 2322696.7 min 353 (NW_012018190.1) # max 63123227 (NW_012016132.1) median 3915 # %34.13 masked total, %35.45 masked real # this information is from the top of # macNem1/refseq/GCF_000956065.1_Mnem_1.0_assembly_report.txt # Assembly name: Mnem_1.0 # Organism name: Macaca nemestrina (pig-tailed macaque) # Isolate: M95218 # Sex: female # Taxid: 9545 # BioSample: SAMN03121842 # BioProject: PRJNA279145 # Submitter: Baylor College of Medicine # Date: 2015-3-19 # Assembly type: haploid # Release type: major # Assembly level: Scaffold # Genome representation: full # WGS project: JZLF01 # Assembly method: AllPathsLG v. 48744; Atlas Link v. 1.1; Atlas GapFill v. 2.2; PBJelly2 v. 14.9.9 # Genome coverage: 113.1x # Sequencing technology: Illumina; PacBio RS; PacBio RSII # RefSeq category: Representative Genome # GenBank assembly accession: GCA_000956065.1 # RefSeq assembly accession: GCF_000956065.1 # RefSeq assembly and GenBank assemblies identical: yes # ## Assembly-Units: ## GenBank Unit Accession RefSeq Unit Accession Assembly-Unit name ## GCA_000956075.1 GCF_000956075.1 Primary Assembly ############################################################################# # establish config.ra file (DONE - Hiram - 2017-02-23) # arguments here are: cd /hive/data/genomes/macNem1 $HOME/kent/src/hg/utils/automation/prepConfig.pl macNem1 mammal macaca \ ./refseq/*_assembly_report.txt > macNem1.config.ra # going to need a mitoAcc ? # NCBI Entrez indicates there is a mitochondrion: NC_026976.1 # verify it looks sane cat macNem1.config.ra # config parameters for makeGenomeDb.pl: db macNem1 clade mammal genomeCladePriority 35 scientificName Macaca nemestrina commonName Pig-tailed macaque assemblyDate Mar. 2015 assemblyLabel Baylor College of Medicine assemblyShortLabel Mnem_1.0 orderKey 16322 mitoAcc NC_026976.1 fastaFiles /hive/data/genomes/macNem1/ucsc/*.fa.gz agpFiles /hive/data/genomes/macNem1/ucsc/*.agp # qualFiles none dbDbSpeciesDir macaca photoCreditURL https://commons.wikimedia.org/wiki/File:Cercopithecidae_-_Macaca_nemastrina.jpg photoCreditName WikiMedia Commons: Hectonichus ncbiGenomeId 13267 ncbiAssemblyId 313501 ncbiAssemblyName Mnem_1.0 ncbiBioProject 279145 ncbiBioSample SAMN03121842 genBankAccessionID GCF_000956065.1 taxId 9545 ############################################################################# # setup UCSC named files (DONE - 2017-02-23 - Hiram) mkdir /hive/data/genomes/macNem1/ucsc cd /hive/data/genomes/macNem1/ucsc # check for duplicate sequences: time faToTwoBit -noMask ../refseq/G*0_genomic.fna.gz refseq.2bit # real 1m18.995s twoBitDup refseq.2bit # no output is a good result, otherwise, would have to eliminate duplicates # the scripts creating the fasta here will be using this refseq.2bit file # remove it later time twoBitToFa refseq.2bit stdout | faSize stdin # 2948703511 bases (110217031 N's 2838486480 real 2838486480 upper 0 lower) # in 9733 sequences in 1 files # Total size: mean 302959.4 sd 2322696.7 min 353 (NW_012018190.1) # max 63123227 (NW_012016132.1) median 3915 # simple unplaced_scaffolds, just transform their names: zcat ../refseq/G*0_assembly_structure/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fna.gz \ | sed -e 's/.1 Macaca nemestrina .*/v1/;' | gzip -c > unplaced.fa.gz time faToTwoBit unplaced.fa.gz unplaced.2bit # real 1m21.170s zcat ../refseq/G*0_assembly_structure/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \ | sed -e 's/.1\t/v1\t/;' > unplaced.agp # verify fasta and AGPs agree time checkAgpAndFa unplaced.agp unplaced.2bit 2>&1 | tail -4 # All AGP and FASTA entries agree - both files are valid # real 0m16.654s # and no sequence lost from orginal: twoBitToFa unplaced.2bit stdout | faSize stdin # 2948703511 bases (110217031 N's 2838486480 real 2838486480 upper 0 lower) # in 9733 sequences in 1 files # Total size: mean 302959.4 sd 2322696.7 min 353 (NW_012018190v1) # max 63123227 (NW_012016132v1) median 3915 # same numbers as above (except for upper/lower masking) # 2948703511 bases (110217031 N's 2838486480 real 2838486480 upper 0 lower) # in 9733 sequences in 1 files # no longer need these temporary 2bit files rm refseq.2bit unplaced.2bit ############################################################################# # Initial database build (DONE - 2017-02-23 - Hiram) # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp macNem1.config.ra) > agp.log 2>&1 # real 2m57.235s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db macNem1.config.ra) > db.log 2>&1 # real 24m28.659s # check in the trackDb files created in TemporaryTrackDbCheckout/ # and add macNem1 to trackDb/makefile # temporary symlink until masked sequence is available cd /hive/data/genomes/macNem1 ln -s `pwd`/macNem1.unmasked.2bit /gbdb/macNem1/macNem1.2bit ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2017-02-23 - Hiram) mkdir /hive/data/genomes/macNem1/bed/cpgIslandsUnmasked cd /hive/data/genomes/macNem1/bed/cpgIslandsUnmasked time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/macNem1/macNem1.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku macNem1) > do.log 2>&1 # real 187m20.754s cat fb.macNem1.cpgIslandExtUnmasked.txt # 18554530 bases of 1093044709 (1.698%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2017-02-23 - Hiram) mkdir /hive/data/genomes/macNem1/bed/cytoBand cd /hive/data/genomes/macNem1/bed/cytoBand makeCytoBandIdeo.csh macNem1 ######################################################################### # ucscToINSDC table/track (DONE - 2017-02-23 - Hiram) # the sequence here is working for a 'refseq' assembly with a chrM # situation may be specific depending upon what is available in the assembly mkdir /hive/data/genomes/macNem1/bed/ucscToINSDC cd /hive/data/genomes/macNem1/bed/ucscToINSDC # find accession for chrM grep chrM ../../macNem1.agp # chrM 1 16560 8 F NC_026976.1 1 16560 + # find the genbank accession for NC_026976.1 at Entrez nucleotide # The NC_026976.1 name is the RefSeq name, the KP765688.1 is the INSDC name ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../refseq/GCF_*structure/Primary_Assembly KP765688.1 # this is actually ucscToRefSeq since this is a RefSeq assembly awk '{printf "%s\t%s\n", $2, $1}' ucscToINSDC.txt | sort > insdcToUcsc.txt # chrM processing needs special help printf "NC_026976.1\tKP765688.1\n" > refseq.insdc.txt # 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}' | sort >> refseq.insdc.txt awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > name.coordinate.tab sed -e 's/\.1/v1/; s/NC_026976.1/chrM/;' refseq.insdc.txt | sort \ | join - ucscToINSDC.txt | tr '[ ]' '[\t]' | sort -k1 \ | join name.coordinate.tab - | tr '[ ]' '[\t]' | cut -f1-4 \ > ucscToINSDC.bed # verify chrM is correct: grep chrM *.bed # chrM 0 16560 KP765688.1 # should be same line counts throughout: wc -l * # 9734 insdcToUcsc.txt # 9734 name.coordinate.tab # 9734 refseq.insdc.txt # 9734 ucscToINSDC.bed # 9734 ucscToINSDC.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 macNem1 ucscToINSDC stdin ucscToINSDC.bed # checkTableCoords should be silent checkTableCoords macNem1 # should cover %100 entirely: featureBits -countGaps macNem1 ucscToINSDC # 2948720071 bases of 2948720071 (100.000%) in intersection ######################################################################### # UCSC to RefSeq name correspondence (DONE - 2017-02-23 - Hiram) mkdir /hive/data/genomes/macNem1/bed/ucscToRefSeq cd /hive/data/genomes/macNem1/bed/ucscToRefSeq ln -s ../../refseq/GCF_000146605.2_Turkey_5.0_assembly_report.txt . # this assembly_report has "UCSC-style-name" in column 10 # but it does not name anything, they are all "na" # columns 5 and 7 are the INSDC and RefSeq names (printf "KP765688.1\tNC_026976.1\n" grep -v "^#" ../../refseq/G*0_assembly_report.txt \ | awk -F'\t' '{printf "%s\t%s\n", $5,$7}') \ | sort > insdc.refSeq.tab hgsql -N -e 'select name,chrom,chromStart,chromEnd from ucscToINSDC;' \ macNem1 | sort > insdc.ucsc.tab join insdc.ucsc.tab insdc.refSeq.tab | tr '[ ]' '[\t]' \ | cut -f2- | sort > ucsc.refSeq.tab # when working perfectly, all these tab files have the same line count: wc -l *.tab # 9734 insdc.refSeq.tab # 9734 insdc.ucsc.tab # 9734 ucsc.refSeq.tab export chrSize=`cut -f1 ucsc.refSeq.tab | 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 macNem1 ucscToRefSeq ./ucscToRefSeq.sql ucsc.refSeq.tab checkTableCoords macNem1 -table=ucscToRefSeq # should cover %100 all bases: featureBits -countGaps macNem1 ucscToRefSeq # 2948720071 bases of 2948720071 (100.000%) in intersection ######################################################################### # add chromAlias table (DONE - 2017-02-23 - Hiram) mkdir /hive/data/genomes/macNem1/bed/chromAlias cd /hive/data/genomes/macNem1/bed/chromAlias hgsql -N -e 'select chrom,name,"refseq" from ucscToRefSeq;' macNem1 \ > ucsc.refseq.tab hgsql -N -e 'select chrom,name,"genbank" from ucscToINSDC;' macNem1 \ > ucsc.genbank.tab awk '{printf "%s\t%s\t%s\n", $2,$1,$3}' ucsc.genbank.tab ucsc.refseq.tab \ | sort > macNem1.chromAlias.tab hgLoadSqlTab macNem1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ macNem1.chromAlias.tab # add ensembl names: 2017-12-14 mkdir previous mv *.tab previous join -t$'\t' ../idKeys/macNem1.idKeys.txt \ ../../ensembl/ensemblMacNem1.idKeys.txt \ | cut -f2,3 | sort > ucsc.ensembl.tab cut -f1,2 previous/ucsc.refseq.tab | sort > ucsc.refseq.tab cut -f1,2 previous/ucsc.genbank.tab | sort > ucsc.genbank.tab ~/kent/src/hg/utils/automation/chromAlias.pl sort -o macNem1.chromAlias.tab macNem1.chromAlias.tab for t in refseq genbank ensembl do c0=`cat ucsc.$t.tab | wc -l` c1=`grep $t macNem1.chromAlias.tab | wc -l` ok="OK" if [ "$c0" -ne "$c1" ]; then ok="ERROR" fi printf "# checking $t: $c0 =? $c1 $ok\n" done # checking refseq: 9734 =? 9734 OK # checking genbank: 9734 =? 9734 OK # checking ensembl: 9733 =? 9733 OK hgLoadSqlTab macNem1 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ macNem1.chromAlias.tab ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2017-02-23 - Hiram) cd ~/kent/src/hg/makeDb/trackDb/macaca/macNem1 # preview prefixes and suffixes: hgsql -N -e "select frag from gold;" macNem1 \ | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c 94057 JZLF.1 1 NC_.1 # implies a rule: '[JN][CZ][L_][F0-9]+(\.[0-9]+)?' # verify this rule will find them all and eliminate them all: hgsql -N -e "select frag from gold;" macNem1 | wc -l # 94058 hgsql -N -e "select frag from gold;" macNem1 \ | egrep -e '[JN][CZ][L_][F0-9]+(\.[0-9]+)?' | wc -l # 94058 hgsql -N -e "select frag from gold;" macNem1 \ | egrep -v -e '[JN][CZ][L_][F0-9]+(\.[0-9]+)?' | wc -l # 0 # hence, add to trackDb/chicken/macNem1/trackDb.ra searchTable gold shortCircuit 1 termRegex [JN][CZ][L_][F0-9]+(\.[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 - 2017-02-23,25 - Hiram) mkdir /hive/data/genomes/macNem1/bed/repeatMasker cd /hive/data/genomes/macNem1/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku macNem1) > do.log 2>&1 # something odd took place during this run. seven of the jobs appeared # to be running cross-match forever with no progress. Killed them and # allowed them to start over and they finished in normal time. # real 3511m26.962s cat faSize.rmsk.txt # 2948720071 bases (110217031 N's 2838503040 real 1418642754 upper # 1419860286 lower) in 9734 sequences in 1 files # Total size: mean 302929.9 sd 2322579.2 min 353 (NW_012018190v1) # max 63123227 (NW_012016132v1) median 3920 # %48.15 masked total, %50.02 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 macNem1 rmsk # 1420318490 bases of 2948720071 (48.167%) in intersection # real 0m39.024s # 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;' macNem1 \ | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" # total 1420318490.000000 # real 0m42.463s ########################################################################## # running simple repeat (DONE - 2017-02-23 - Hiram) mkdir /hive/data/genomes/macNem1/bed/simpleRepeat cd /hive/data/genomes/macNem1/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 macNem1) > do.log 2>&1 # real 6m30.668s cat fb.simpleRepeat # 69295380 bases of 2838503083 (2.441%) in intersection # adding this trfMask to the other masking cd /hive/data/genomes/macNem1 # when using the Window Masker result: # twoBitMask bed/windowMasker/macNem1.cleanWMSdust.2bit \ # -add bed/simpleRepeat/trfMask.bed macNem1.2bit # you can safely ignore the warning about fields >= 13 # when using Rmsk results, add to rmsk after it is done: twoBitMask macNem1.rmsk.2bit \ -add bed/simpleRepeat/trfMask.bed macNem1.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa macNem1.2bit stdout | faSize stdin > faSize.macNem1.2bit.txt cat faSize.macNem1.2bit.txt # 2948720071 bases (110217031 N's 2838503040 real 1417139630 upper # 1421363410 lower) in 9734 sequences in 1 files # Total size: mean 302929.9 sd 2322579.2 min 353 (NW_012018190v1) # max 63123227 (NW_012016132v1) median 3920 # %48.20 masked total, %50.07 masked real # reset the symlink rm /gbdb/macNem1/macNem1.2bit ln -s `pwd`/macNem1.2bit /gbdb/macNem1/macNem1.2bit ######################################################################### # CREATE MICROSAT TRACK (DONE - 2017-02-23 - Hiram) ssh hgwdev mkdir /cluster/data/macNem1/bed/microsat cd /cluster/data/macNem1/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 macNem1 microsat microsat.bed # Read 30599 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2017-02-26 - Hiram) mkdir /hive/data/genomes/macNem1/bed/windowMasker cd /hive/data/genomes/macNem1/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev macNem1) > do.log 2>&1 # real 206m6.903s # Masking statistics cat faSize.macNem1.cleanWMSdust.txt # 1128339136 bases (35294427 N's 1093044709 real 852254578 upper # 240790131 lower) in 231286 sequences in 1 files # Total size: mean 4878.5 sd 586734.6 min 99 (chrUn_NW_011236709v1) # max 190651702 (chr1) median 462 # %21.34 masked total, %22.03 masked real cat fb.macNem1.rmsk.windowmaskerSdust.txt # 779179867 bases of 2948720071 (26.424%) in intersection # when this completes before rmsk is done: time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -continue=cleanup -dbHost=hgwdev macNem1) > cleanup.log 2>&1 # real 2m2.076s ########################################################################## # run up idKeys files for ncbiRefSeq (DONE - 2017-02-27 - Hiram) mkdir /hive/data/genomes/macNem1/bed/idKeys cd /hive/data/genomes/macNem1/bed/idKeys time (doIdKeys.pl -buildDir=`pwd` macNem1) > do.log 2>&1 & # real 12m39.807s cat macNem1.keySignature.txt # 4565a53011bf78e2e6a489f4ed1cc8f9 ########################################################################## # cpgIslands - (DONE - 2017-02-27 - Hiram) mkdir /hive/data/genomes/macNem1/bed/cpgIslands cd /hive/data/genomes/macNem1/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku macNem1) > do.log 2>&1 & # real 6m11.419s cat fb.macNem1.cpgIslandExt.txt # 21891021 bases of 2838503083 (0.771%) in intersection ############################################################################## # genscan - (DONE - 2017-02-27 - Hiram) mkdir /hive/data/genomes/macNem1/bed/genscan cd /hive/data/genomes/macNem1/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku macNem1) > do.log 2>&1 & # real 20m34.741s # Completed: 9733 of 9734 jobs # Crashed: 1 jobs # CPU time in finished jobs: 79229s 1320.49m 22.01h 0.92d 0.003 y # IO & Wait Time: 27058s 450.96m 7.52h 0.31d 0.001 y # Average job time: 11s 0.18m 0.00h 0.00d # Longest finished job: 1563s 26.05m 0.43h 0.02d # Submission to last job: 1620s 27.00m 0.45h 0.02d # one job need window size 2000000 to finish ./runLastGsBig.csh NW_012012688v1 000 gtf/000/NW_012012688v1.gtf pep/000/NW_012012688v1.pep subopt/000/NW_012012688v1.bed # real 3m26.559s # continuing: time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -continue=makeBed -bigClusterHub=ku macNem1) > makeBed.log 2>&1 # real 3m3.487s cat fb.macNem1.genscan.txt # 52901202 bases of 2838503083 (1.864%) in intersection cat fb.macNem1.genscanSubopt.txt # 53835070 bases of 2838503083 (1.897%) in intersection ############################################################################# # augustus gene track (DONE - 2017-02-27 - Hiram) mkdir /hive/data/genomes/macNem1/bed/augustus cd /hive/data/genomes/macNem1/bed/augustus time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ -species=human -dbHost=hgwdev -workhorse=hgwdev macNem1) > do.log 2>&1 & # real 98m44.064s cat fb.macNem1.augustusGene.txt # 48184131 bases of 2838503083 (1.698%) in intersection ######################################################################### # Create kluster run files (DONE - 2017-02-27 - Hiram) # numerator is macNem1 gapless bases "real" as reported by: featureBits -noRandom -noHap macNem1 gap # 110216988 bases of 2838503083 (3.883%) 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 \( 2838503083 / 2861349177 \) \* 1024 # ( 2838503083 / 2861349177 ) * 1024 = 1015.823997 # ==> use -repMatch=1000 according to size scaled down from 1024 for human. # and rounded down to nearest 50 cd /hive/data/genomes/macNem1 blat macNem1.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/macNem1.11.ooc \ -repMatch=1000 # Wrote 26519 overused 11-mers to jkStuff/macNem1.11.ooc # check non-bridged gaps to see what the typical size is: hgsql -N \ -e 'select * from gap where bridge="no" order by size;' macNem1 \ | sort -k7,7nr | ave -col=7 stdin # there are none 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 macNem1 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 # Macaca nemestrina 1412 34762 55 # edit etc/genbank.conf to add macNem1 just before macFas5 # macNem1 (pig-tailed macaque) macNem1.serverGenome = /hive/data/genomes/macNem1/macNem1.2bit macNem1.clusterGenome = /hive/data/genomes/macNem1/macNem1.2bit macNem1.ooc = /hive/data/genomes/macNem1/jkStuff/macNem1.11.ooc macNem1.lift = no macNem1.perChromTables = no macNem1.downloadDir = macNem1 macNem1.refseq.mrna.native.pslCDnaFilter = ${ordered.refseq.mrna.native.pslCDnaFilter} macNem1.refseq.mrna.xeno.pslCDnaFilter = ${ordered.refseq.mrna.xeno.pslCDnaFilter} macNem1.genbank.mrna.native.pslCDnaFilter = ${ordered.genbank.mrna.native.pslCDnaFilter} macNem1.genbank.mrna.xeno.pslCDnaFilter = ${ordered.genbank.mrna.xeno.pslCDnaFilter} macNem1.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 # macNem1.upstreamGeneTbl = refGene # macNem1.upstreamMaf = multiz7way # /hive/data/genomes/melGal4/bed/multiz7way/species.lst git commit -m "Added macNem1; refs #18927" etc/genbank.conf git push # update /cluster/data/genbank/: make etc-update # also add to src/lib/gbGenome.c # static char *macNemNames[] = {"Macaca nemestrina", NULL}; # {"macNem", macNemNames}, git commit -m "Added macNem1; refs #18927" src/lib/gbGenome.c make install-server cd /cluster/data/genbank time ./bin/gbAlignStep -initial macNem1 # logFile: var/build/logs/2017.03.05-06:51:20.macNem1.initalign.log # about 13 hours tail var/build/logs/2017.03.05-06:51:20.macNem1.initalign.log # hgwdev 2017.03.05-19:31:30 macNem1.initalign: Succeeded: macNem1 # hgwdev 2017.03.05-19:32:03 macNem1.initalign: finish # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.macNem1 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad macNem1 # logFile: var/dbload/hgwdev/logs/2017.03.05-19:53:57.macNem1.dbload.log # about 21 minutes tail -1 var/dbload/hgwdev/logs/2017.03.05-19:53:57.macNem1.dbload.log # hgwdev 2017.03.05-20:14:56 macNem1.dbload: finish # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add macNem1 to: # etc/align.dbs etc/hgwdev.dbs git add etc/align.dbs etc/hgwdev.dbs git commit -m 'adding macNem1 to the update alignments refs #18927' etc/align.dbs etc/hgwdev.dbs git push make etc-update ############################################################################# # swap chainNet from mouse/mm10 (DONE - 2017-03-01 - Hiram) # since this is a new assembly, add the chainNet definitions to # trackDb # the original alignment cd /hive/data/genomes/mm10/bed/lastzMacNem1.2017-02-28 cat fb.mm10.chainMacNem1Link.txt # 918083212 bases of 2652783500 (34.608%) in intersection # and for the swap: mkdir /hive/data/genomes/macNem1/bed/blastz.mm10.swap cd /hive/data/genomes/macNem1/bed/blastz.mm10.swap time (doBlastzChainNet.pl -verbose=2 \ /hive/data/genomes/mm10/bed/lastzMacNem1.2017-02-28/DEF \ -swap -chainMinScore=3000 -chainLinearGap=medium \ -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ -syntenicNet) > swap.log 2>&1 # real 65m14.074s cat fb.macNem1.chainMm10Link.txt # 905682728 bases of 2838503083 (31.907%) in intersection time (doRecipBest.pl -workhorse=hgwdev -buildDir=`pwd` macNem1 mm10) \ > rbest.log 2>&1 # real 321m2.285s ############################################################################# # ncbiRefSeq (TBD - 2016-05-13 - Hiram) mkdir /hive/data/genomes/macNem1/bed/ncbiRefSeq cd /hive/data/genomes/macNem1/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 macNem1) > 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 macNem1) > 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 macNem1) > load.log 2>&1 # real 0m33.205s cat fb.ncbiRefSeq.macNem1.txt # 82563006 bases of 1218501075 (6.776%) in intersection featureBits -enrichment macNem1 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 ("macNem1", "blat1a", "17884", "1", "0"); \ INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ VALUES ("macNem1", "blat1a", "17885", "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="macNem1";' hgcentraltest ######################################################################### # all.joiner update, downloads and in pushQ - (DONE - 2017-03-07 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output joinerCheck -database=macNem1 -tableCoverage all.joiner joinerCheck -database=macNem1 -times all.joiner joinerCheck -database=macNem1 -keys all.joiner cd /hive/data/genomes/macNem1 time (makeDownloads.pl -workhorse=hgwdev macNem1) > downloads.log 2>&1 # real 25m42.531s # now ready for pushQ entry mkdir /hive/data/genomes/macNem1/pushQ cd /hive/data/genomes/macNem1/pushQ time (makePushQSql.pl macNem1) > macNem1.pushQ.sql 2> stderr.out # real 7m21.629s # check for errors in stderr.out, some are OK, e.g.: # WARNING: macNem1 does not have seq # WARNING: macNem1 does not have extFile # copy it to hgwbeta scp -p macNem1.pushQ.sql qateam@hgwbeta:/tmp/ ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/macNem1.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly #########################################################################