c856e12e25aa93d1ec5095fbb720b6b564f4a550 hiram Wed Jul 3 11:35:38 2019 -0700 ready to start genbank run refs #23771 diff --git src/hg/makeDb/doc/rheMac10/initialBuild.txt src/hg/makeDb/doc/rheMac10/initialBuild.txt new file mode 100644 index 0000000..9e6c0d4 --- /dev/null +++ src/hg/makeDb/doc/rheMac10/initialBuild.txt @@ -0,0 +1,1002 @@ +# for emacs: -*- mode: sh; -*- + +# This file describes browser build for the rheMac10 + +# Can use existing photograph (otherwise find one before starting here) + +######################################################################### +# Initial steps, reuse existing photograph (DONE - 2019-07-01 - Hiram) + +# To start this initialBuild.txt document, from a previous assembly document: + +mkdir ~/kent/src/hg/makeDb/doc/rheMac10 +cd ~/kent/src/hg/makeDb/doc/rheMac10 + +sed -e 's/galGal6/rheMac10/g; s/GalGal6/RheMac10/g; s/DONE/TBD/g;' \ + ../galGal6/initialBuild.txt > initialBuild.txt + +mkdir -p /hive/data/genomes/rheMac10/refseq +cd /hive/data/genomes/rheMac10 + +# Can use existing photograph +cp -p ../rn6/photoReference.txt ./ +cat photoReference.txt +photoCreditURL http://www.genome.gov/dmd/img.cfm?node=Photos/Animals/Rat%20(brown%20norway)&id=79111 +photoCreditName NHGRI press photos + +## download from NCBI +cd /hive/data/genomes/rheMac10/refseq + +time rsync -L -a -P --stats \ +rsync://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Macaca_mulatta/all_assembly_versions/GCF_003339765.1_Mmul_10/ ./ + +# sent 3,937 bytes received 6,369,691,679 bytes 56,619,516.59 bytes/sec +# total size is 6,368,121,788 speedup is 1.00 + +# real 1m51.641s + +# this information is from the top of +# rheMac10/refseq/*_assembly_report.txt +# (aka: rheMac10/refseq/GCF_003339765.1_Mmul_10_assembly_report.txt + +# Assembly name: Mmul_10 +# Organism name: Macaca mulatta (Rhesus monkey) +# Taxid: 9544 +# BioSample: SAMN09435472 +# BioProject: PRJNA476474 +# Submitter: The Genome Institute at Washington University School of Medicine +# Date: 2019-02-13 +# Assembly type: haploid +# Release type: major +# Assembly level: Chromosome +# Genome representation: full +# WGS project: QNVO02 +# GenBank assembly accession: GCA_003339765.3 +# RefSeq assembly accession: GCF_003339765.1 +# RefSeq assembly and GenBank assemblies identical: no +# +## Assembly-Units: +## GenBank Unit Accession RefSeq Unit Accession Assembly-Unit name +## GCA_003339795.3 GCF_003339795.1 Primary Assembly +## GCF_000005875.1 non-nuclear +# + +# check assembly size for later reference: + +faSize G*10_genomic.fna.gz +# 2971331530 bases (34438805 N's 2936892725 real 1810047126 upper +# 1126845599 lower) in 2939 sequences in 1 files +# Total size: mean 1011000.9 sd 12006707.4 min 217 (NW_021162293.1) +# max 223616942 (NC_041754.1) median 26547 +# %37.92 masked total, %38.37 masked real + +############################################################################# +# establish config.ra file (DONE - Hiram - 2019-07-01) + cd /hive/data/genomes/rheMac10 + ~/kent/src/hg/utils/automation/prepConfig.pl rheMac10 primate rhesus \ + refseq/*_assembly_report.txt > rheMac10.config.ra + + # compare with previous version to see if it is sane: + diff rheMac10.config.ra ../rheMac8/rheMac8.config.ra + + # verify it really does look sane + cat rheMac10.config.ra +# config parameters for makeGenomeDb.pl: +db rheMac10 +clade primate +genomeCladePriority 35 +scientificName Macaca mulatta +commonName rhesus +assemblyDate Feb. 2019 +assemblyLabel The Genome Institute at Washington University School of Medicine +assemblyShortLabel Mmul_10 +orderKey 18417 +# mitochondrial sequence included in refseq release +# mitoAcc NC_005943.1 +mitoAcc none +fastaFiles /hive/data/genomes/rheMac10/ucsc/*.fa.gz +agpFiles /hive/data/genomes/rheMac10/ucsc/*.agp +# qualFiles none +dbDbSpeciesDir rhesus +photoCreditURL http://www.genome.gov/dmd/img.cfm?node=Photos/Animals/Rat%20(brown%20norway)&id=79111 +photoCreditName NHGRI press photos +ncbiGenomeId 215 +ncbiAssemblyId 2345051 +ncbiAssemblyName Mmul_10 +ncbiBioProject 476474 +ncbiBioSample SAMN09435472 +genBankAccessionID GCF_003339765.1 +taxId 9544 + +############################################################################# +# setup UCSC named files (DONE - 2019-07-01 - Hiram) + + mkdir /hive/data/genomes/rheMac10/ucsc + cd /hive/data/genomes/rheMac10/ucsc + + # check for duplicate sequences: + time faToTwoBit -noMask ../refseq/G*10_genomic.fna.gz refseq.2bit + # real 0m38.595s + + 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 + + # new option required to ucscCompositeAgp.pl 2016-04-13 + time ~/kent/src/hg/utils/automation/ucscCompositeAgp.pl \ + ../refseq/G*10_genomic.fna.gz \ + ../refseq/*_assembly_structure/Primary_Assembly +# constructing refseq.2bit from ../refseq/GCF_000002315.4_Gallus_gallus-5.0_genomic.fna.gz +NC_027914.1 chrY +NC_041754.1 chr1 +NC_041755.1 chr2 +NC_041756.1 chr3 +NC_041757.1 chr4 +NC_041758.1 chr5 +NC_041759.1 chr6 +NC_041760.1 chr7 +NC_041761.1 chr8 +NC_041762.1 chr9 +NC_041763.1 chr10 +NC_041764.1 chr11 +NC_041765.1 chr12 +NC_041766.1 chr13 +NC_041767.1 chr14 +NC_041768.1 chr15 +NC_041769.1 chr16 +NC_041770.1 chr17 +NC_041771.1 chr18 +NC_041772.1 chr19 +NC_041773.1 chr20 +NC_041774.1 chrX + +real 10m33.076s + + time ~/kent/src/hg/utils/automation/unplacedWithChroms.pl \ + ../refseq/*_assembly_structure/Primary_Assembly + # processed 340 sequences into chrUn.fa.gz + # real 0m30.432s + + time ~/kent/src/hg/utils/automation/unlocalizedWithChroms.pl \ + ../refseq/*_assembly_structure/Primary_Assembly +# 11 +# 7 +# 17 +# 2 +# 1 +# 18 +# 16 +# 13 +# 6 +# X +# 3 +# 9 +# 12 +# 20 +# 14 +# 15 +# 8 +# 4 +# 19 +# 10 +# 5 +# processed 291 sequences into chr*_random.gz 21 files + +real 0m7.023s + + # bash syntax here + mitoAcc=`grep "^# mitoAcc" ../rheMac10.config.ra | awk '{print $NF}'` + printf "# mitoAcc %s\n" "$mitoAcc" +# mitoAcc NC_005943.1 + + zcat \ + ../refseq/*_assembly_structure/non-nuclear/assem*/AGP/chrMT.comp.agp.gz \ + | grep -v "^#" | sed -e "s/^$mitoAcc/chrM/;" > chrM.agp + + cat chrM.agp +# chrM 1 16564 1 O NC_005943.1 1 16564 + + + printf ">chrM\n" > chrM.fa + twoBitToFa -noMask refseq.2bit:$mitoAcc stdout | grep -v "^>" >> chrM.fa + gzip chrM.fa + + faSize chrM.fa.gz +# 16564 bases (0 N's 16564 real 16564 upper 0 lower) in 1 sequences in 1 files + + # verify fasta and AGPs agree + time faToTwoBit *.fa.gz test.2bit + # real 0m56.726s + + cat *.agp | checkAgpAndFa stdin test.2bit 2>&1 | tail -4 + # All AGP and FASTA entries agree - both files are valid + + # and no sequence lost from orginal: + twoBitToFa test.2bit stdout | faSize stdin +# 2971331530 bases (34438805 N's 2936892725 real 2936892725 upper 0 lower) +# in 2939 sequences in 1 files +# Total size: mean 1011000.9 sd 12006707.4 min 217 (chrUn_NW_021162293v1) +# max 223616942 (chr1) median 26547 + + # same numbers as above (except for upper/lower masking) +# 2971331530 bases (34438805 N's 2936892725 real 1810047126 upper +# 1126845599 lower) in 2939 sequences in 1 files +# Total size: mean 1011000.9 sd 12006707.4 min 217 (NW_021162293.1) +# max 223616942 (NC_041754.1) median 26547 + + # no longer need these temporary 2bit files + rm test.2bit refseq.2bit + +############################################################################# +# Initial database build (DONE - 2019-07-01 - Hiram) + + # verify sequence and AGP are OK: + cd /hive/data/genomes/rheMac10 + time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ + -stop=agp rheMac10.config.ra) > agp.log 2>&1 + # real 2m22.570s + + # then finish it off: + time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ + -fileServer=hgwdev -continue=db rheMac10.config.ra) > db.log 2>&1 + # real 15m24.340s + + # check in the trackDb files created in TemporaryTrackDbCheckout/ + # and add rheMac10 to trackDb/makefile + + # temporary symlink until masked sequence is available + cd /hive/data/genomes/rheMac10 + ln -s `pwd`/rheMac10.unmasked.2bit /gbdb/rheMac10/rheMac10.2bit + +############################################################################## +# cpgIslands on UNMASKED sequence (DONE - 2019-07-01 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/cpgIslandsUnmasked + cd /hive/data/genomes/rheMac10/bed/cpgIslandsUnmasked + + time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ + -tableName=cpgIslandExtUnmasked \ + -maskedSeq=/hive/data/genomes/rheMac10/rheMac10.unmasked.2bit \ + -workhorse=hgwdev -smallClusterHub=ku rheMac10) > do.log 2>&1 + # real 3m25.609s + + cat fb.rheMac10.cpgIslandExtUnmasked.txt + # 38819797 bases of 2936892733 (1.322%) in intersection + +############################################################################# +# cytoBandIdeo - (DONE - 2019-07-01 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/cytoBand + cd /hive/data/genomes/rheMac10/bed/cytoBand + makeCytoBandIdeo.csh rheMac10 + +############################################################################# +# run up idKeys files for chromAlias/ncbiRefSeq (DONE - 2019-07-01 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/idKeys + cd /hive/data/genomes/rheMac10/bed/idKeys + + time (doIdKeys.pl \ + -twoBit=/hive/data/genomes/rheMac10/rheMac10.unmasked.2bit \ + -buildDir=`pwd` rheMac10) > do.log 2>&1 & + # real 4m9.394s + + cat rheMac10.keySignature.txt + # 69c74f5fa18fe2421cabbcd0e80657e9 + +############################################################################# +# gapOverlap (DONE - 2019-07-01 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/gapOverlap + cd /hive/data/genomes/rheMac10/bed/gapOverlap + time (doGapOverlap.pl \ + -twoBit=/hive/data/genomes/rheMac10/rheMac10.unmasked.2bit rheMac10 ) \ + > do.log 2>&1 & + # real 1m3.768s + + # there is only one: + wc -l bed.tab + # 1 bed.tab + cut -f2- bed.tab +chr4 51027118 51073686 chr4:51027119-51073686 1000 + 51027118 51073686 0 2 1000,1000 0,45568 + + cat fb.rheMac10.gapOverlap.txt + # 2000 bases of 2971331530 (0.000%) in intersection + +############################################################################# +# tandemDups (DONE - 2019-07-01 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/tandemDups + cd /hive/data/genomes/rheMac10/bed/tandemDups + time (~/kent/src/hg/utils/automation/doTandemDup.pl \ + -twoBit=/hive/data/genomes/rheMac10/rheMac10.unmasked.2bit rheMac10) \ + > do.log 2>&1 & + # real 225m10.394s + + cat fb.rheMac10.tandemDups.txt + # 81485851 bases of 2971331530 (2.742%) in intersection + + bigBedInfo rheMac10.tandemDups.bb | sed -e 's/^/# /;' +# version: 4 +# fieldCount: 13 +# hasHeaderExtension: yes +# isCompressed: yes +# isSwapped: 0 +# extraIndexCount: 0 +# itemCount: 1,431,488 +# primaryDataSize: 37,737,612 +# primaryIndexSize: 176,456 +# zoomLevels: 9 +# chromCount: 2756 +# basesCovered: 1,568,894,232 +# meanDepth (of bases covered): 8.122557 +# minDepth: 1.000000 +# maxDepth: 314.000000 +# std of depth: 13.438018 + +######################################################################### +# ucscToINSDC and ucscToRefSeq table/track (DONE - 2019-07-01 - Hiram) + # construct idKeys for the refseq sequence + mkdir /hive/data/genomes/rheMac10/refseq/idKeys + cd /hive/data/genomes/rheMac10/refseq/idKeys + faToTwoBit ../GCF_003339765.1_Mmul_10_genomic.fna.gz rheMac10.refSeq.2bit + + time (doIdKeys.pl -buildDir=`pwd` \ + -twoBit=`pwd`/rheMac10.refSeq.2bit refseqRheMac10) > do.log 2>&1 & + # real 4m8.863s + + cat refseqRheMac10.keySignature.txt + # 69c74f5fa18fe2421cabbcd0e80657e9 + + # and the genbank sequence needs keys too: + mkdir /hive/data/genomes/rheMac10/refseq/idKeysGenbank + cd /hive/data/genomes/rheMac10/refseq/idKeysGenbank + faToTwoBit /hive/data/outside/ncbi/genomes/genbank/vertebrate_mammalian/Macaca_mulatta/all_assembly_versions/GCA_003339765.3_Mmul_10/GCA_003339765.3_Mmul_10_genomic.fna.gz rheMac10.genbank.2bit + + time (doIdKeys.pl -buildDir=`pwd` \ + -twoBit=`pwd`/rheMac10.genbank.2bit genbankRheMac10) > do.log 2>&1 & + # real 4m6.755s + + + cat genbankRheMac10.keySignature.txt + # 1eb34462822483c9262e98186f71a286 + + mkdir /hive/data/genomes/rheMac10/bed/chromAlias + cd /hive/data/genomes/rheMac10/bed/chromAlias + + join -t$'\t' ../idKeys/rheMac10.idKeys.txt \ + ../../refseq/idKeysGenbank/genbankRheMac10.idKeys.txt | cut -f2- \ + | sort -k1,1 | join -t$'\t' <(sort -k1,1 ../../chrom.sizes) - \ + | awk '{printf "%s\t0\t%d\t%s\n", $1, $2, $3}' \ + | sort -k1,1 -k2,2n > ucscToINSDC.bed + + join -t$'\t' ../idKeys/rheMac10.idKeys.txt \ + ../../refseq/idKeys/refseqRheMac10.idKeys.txt | cut -f2- \ + | sort -k1,1 | join -t$'\t' <(sort -k1,1 ../../chrom.sizes) - \ + | awk '{printf "%s\t0\t%d\t%s\n", $1, $2, $3}' \ + | sort -k1,1 -k2,2n > ucscToRefSeq.bed + + # should be same line counts throughout: + wc -l * ../../chrom.sizes + # 2938 ucscToINSDC.bed + # 2939 ucscToRefSeq.bed + # 2939 ../../chrom.sizes + + # need to find the accession for the INSDC equivalent to chrM: + egrep chrM * +# ucscToRefSeq.bed:chrM 0 16564 NC_005943.1 + + # lookup that accession at NCBI Entrez: AY612638.1 + # and add to ucscToINSDC.bed: + printf "chrM\t0\t16564\tAY612638.1\n" >> ucscToINSDC.bed + # verify: + grep chrM * +ucscToINSDC.bed:chrM 0 16564 AY612638.1 +ucscToRefSeq.bed:chrM 0 16564 NC_005943.1 + + export chrSize=`cut -f1 ucscToINSDC.bed | awk '{print length($0)}' | sort -n | tail -1` + echo $chrSize + # 27 + # use the $chrSize in this sed + sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ + | hgLoadSqlTab rheMac10 ucscToINSDC stdin ucscToINSDC.bed + # should be the same for ucscToRefSeq: + export chrSize=`cut -f1 ucscToRefSeq.bed | awk '{print length($0)}' | sort -n | tail -1` + echo $chrSize + # 27 + sed -e "s/21/$chrSize/" $HOME/kent/src/hg/lib/ucscToINSDC.sql \ + | sed -e 's/INSDC/RefSeq/g;' \ + | hgLoadSqlTab rheMac10 ucscToRefSeq stdin ucscToRefSeq.bed + + # should be quiet for all OK + checkTableCoords rheMac10 + + # should cover %100 entirely: + featureBits -countGaps rheMac10 ucscToINSDC + # 2971331530 bases of 2971331530 (100.000%) in intersection + featureBits -countGaps rheMac10 ucscToRefSeq + # 2971331530 bases of 2971331530 (100.000%) in intersection + +######################################################################### +# add chromAlias table (DONE - 2019-07-01 - Hiram) + + mkdir /hive/data/genomes/rheMac10/bed/chromAlias + cd /hive/data/genomes/rheMac10/bed/chromAlias + + hgsql -N -e 'select chrom,name from ucscToRefSeq;' rheMac10 \ + | sort -k1,1 > ucsc.refseq.tab + hgsql -N -e 'select chrom,name from ucscToINSDC;' rheMac10 \ + | sort -k1,1 > ucsc.genbank.tab + + ### Adding Ensembl alias with v95 release, after idKeys made: 2019-01-16 + join -t$'\t' ../idKeys/rheMac10.idKeys.txt \ + ../../ens95/ensRheMac10.idKeys.txt | cut -f2- \ + | sort -k1,1 | join -t$'\t' <(sort -k1,1 ../../chrom.sizes) - \ + | awk '{printf "%s\t0\t%d\t%s\n", $1, $2, $3}' \ + | sort -k1,1 -k2,2n > ucscToEns.bed + cut -f1,4 ucscToEns.bed | sort > ucsc.ensembl.tab + wc -l *.bed +# 2210 ucscToEns.bed +# 2211 ucscToINSDC.bed +# 2211 ucscToRefSeq.bed + + ~/kent/src/hg/utils/automation/chromAlias.pl ucsc.*.tab \ + > rheMac10.chromAlias.tab + +for t in refseq genbank +do + c0=`cat ucsc.$t.tab | wc -l` + c1=`grep $t rheMac10.chromAlias.tab | wc -l` + ok="OK" + if [ "$c0" -ne "$c1" ]; then + ok="ERROR" + fi + printf "# checking $t: $c0 =? $c1 $ok\n" +done +# checking refseq: 2939 =? 2939 OK +# checking genbank: 2939 =? 2939 OK + + # verify chrM is here properly: + grep chrM rheMac10.chromAlias.tab +# AY612638.1 chrM genbank +# NC_005943.1 chrM refseq + + hgLoadSqlTab rheMac10 chromAlias ~/kent/src/hg/lib/chromAlias.sql \ + rheMac10.chromAlias.tab + +######################################################################### +# fixup search rule for assembly track/gold table (TBD - 2019-07-01 - Hiram) + cd ~/kent/src/hg/makeDb/trackDb/rhesus/rheMac10 + # preview prefixes and suffixes: + hgsql -N -e "select frag from gold;" rheMac10 \ + | sed -e 's/[0-9][0-9]*//;' | sort | uniq -c + 2 AC.10 + 2 AC.11 + 1 AC.13 + 1 AC.17 + 4 AC.2 + 36 AC.3 + 22 AC.4 + 8 AC.5 + 2 AC.6 + 5 AC.7 + 4 AC.8 + 2 AC.9 + 1 NC_.1 + 3178 QNVO.1 + + # implies a rule: '[ANQ][CN][V0-9_][O0-9][0-9]+(\.[0-9]+)?' + + # verify this rule will find them all and eliminate them all: + hgsql -N -e "select frag from gold;" rheMac10 | wc -l + # 3268 + + hgsql -N -e "select frag from gold;" rheMac10 \ + | egrep -e '[ANQ][CN][V0-9_][O0-9][0-9]+(\.[0-9]+)?' | wc -l + # 3268 + + hgsql -N -e "select frag from gold;" rheMac10 \ + | egrep -v -e '[ANQ][CN][V0-9_][O0-9][0-9]+(\.[0-9]+)?' | wc -l + # 0 + + # hence, add to trackDb/rhesus/rheMac10/trackDb.ra +searchTable gold +shortCircuit 1 +termRegex [ANQ][CN][V0-9_][O0-9][0-9]+(\.[0-9]+)? +query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' +searchPriority 8 + +XXX - TBD + + # verify searches work in the position box + +########################################################################## +# running repeat masker (DONE - 2019-07-01 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/repeatMasker + cd /hive/data/genomes/rheMac10/bed/repeatMasker + time (doRepeatMasker.pl -buildDir=`pwd` \ + -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ + -smallClusterHub=ku rheMac10) > do.log 2>&1 + # real 500m43.380s + + cat faSize.rmsk.txt +# 2971331530 bases (34438805 N's 2936892725 real 1399379596 upper +# 1537513129 lower) in 2939 sequences in 1 files +# Total size: mean 1011000.9 sd 12006707.4 min 217 (chrUn_NW_021162293v1) +# max 223616942 (chr1) median 26547 +# %51.74 masked total, %52.35 masked real + + egrep -i "versi|relea" do.log +# RepeatMasker version development-$Id: RepeatMasker,v 1.332 2017/04/17 19:01:11 rhubley Exp $ +# February 01 2017 (open-4-0-8) 1.332 version of RepeatMasker +# CC Dfam_Consensus RELEASE 20181026; * +# CC RepBase RELEASE 20181026; + + time featureBits -countGaps rheMac10 rmsk + # 1537510421 bases of 2971331530 (51.745%) in intersection + # real 0m40.763s + + # 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;' rheMac10 \ + | bedSingleCover.pl stdin | ave -col=4 stdin | grep "^total" + # total 1537510421.000000 + # real 0m24.256s + +########################################################################## +# running simple repeat (DONE - 2019-07-01 - Hiram) + + # The '-trf409 5' is a bit smaller than human which is 6 + + mkdir /hive/data/genomes/rheMac10/bed/simpleRepeat + cd /hive/data/genomes/rheMac10/bed/simpleRepeat + time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ + -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ + -trf409=5 rheMac10) > do.log 2>&1 + # real 933m29.936s + + cat fb.simpleRepeat + # 158143361 bases of 2936892733 (5.385%) in intersection + + cd /hive/data/genomes/rheMac10 + # using the Window Masker result: + cd /hive/data/genomes/rheMac10 +# twoBitMask bed/windowMasker/rheMac10.cleanWMSdust.2bit \ +# -add bed/simpleRepeat/trfMask.bed rheMac10.2bit + # you can safely ignore the warning about fields >= 13 + + # add to rmsk after it is done: + twoBitMask rheMac10.rmsk.2bit \ + -add bed/simpleRepeat/trfMask.bed rheMac10.2bit + # you can safely ignore the warning about fields >= 13 + twoBitToFa rheMac10.2bit stdout | faSize stdin > faSize.rheMac10.2bit.txt + cat faSize.rheMac10.2bit.txt +# 2971331530 bases (34438805 N's 2936892725 real 1397548789 upper +# 1539343936 lower) in 2939 sequences in 1 files +# Total size: mean 1011000.9 sd 12006707.4 min 217 (chrUn_NW_021162293v1) +# max 223616942 (chr1) median 26547 +# %51.81 masked total, %52.41 masked real + + rm /gbdb/rheMac10/rheMac10.2bit + ln -s `pwd`/rheMac10.2bit /gbdb/rheMac10/rheMac10.2bit + +######################################################################### +# CREATE MICROSAT TRACK (DONE - 2019-07-02 - Hiram) + ssh hgwdev + mkdir /cluster/data/rheMac10/bed/microsat + cd /cluster/data/rheMac10/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 rheMac10 microsat microsat.bed + # Read 32040 elements of size 4 from microsat.bed + +########################################################################## +## WINDOWMASKER (DONE - 2019-07-01 - Hiram) + + mkdir /hive/data/genomes/rheMac10/bed/windowMasker + cd /hive/data/genomes/rheMac10/bed/windowMasker + time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ + -dbHost=hgwdev rheMac10) > do.log 2>&1 + # real 111m14.660s + + # Masking statistics + cat faSize.rheMac10.cleanWMSdust.txt +# 2971331530 bases (34438805 N's 2936892725 real 1792976685 upper +# 1143916040 lower) in 2939 sequences in 1 files +# Total size: mean 1011000.9 sd 12006707.4 min 217 (chrUn_NW_021162293v1) +# max 223616942 (chr1) median 26547 +# %38.50 masked total, %38.95 masked real + + cat fb.rheMac10.rmsk.windowmaskerSdust.txt + # 909614480 bases of 2971331530 (30.613%) in intersection + +########################################################################## +# cpgIslands - (DONE - 2019-07-02 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/cpgIslands + cd /hive/data/genomes/rheMac10/bed/cpgIslands + time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ + -workhorse=hgwdev -smallClusterHub=ku rheMac10) > do.log 2>&1 +XXX - running - Tue Jul 2 07:51:16 PDT 2019 + # real 2m5.105s + + cat fb.rheMac10.cpgIslandExt.txt + # 16395346 bases of 1055588482 (1.553%) in intersection + +############################################################################## +# genscan - (DONE - 2019-07-02 - Hiram) + mkdir /hive/data/genomes/rheMac10/bed/genscan + cd /hive/data/genomes/rheMac10/bed/genscan + time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ + -bigClusterHub=ku rheMac10) > do.log 2>&1 +XXX - running - Tue Jul 2 07:51:42 PDT 2019 + # real 88m34.900s + + cat fb.rheMac10.genscan.txt + # 23911678 bases of 1055588482 (2.265%) in intersection + + cat fb.rheMac10.genscanSubopt.txt + # 24521608 bases of 1055588482 (2.323%) in intersection + +######################################################################### +# Create kluster run files (DONE - 2019-07-01 - Hiram) + + # numerator is rheMac10 gapless bases "real" as reported by: + featureBits -noRandom -noHap rheMac10 gap + # 32449617 bases of 2821543288 (1.150%) 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 \( 2821543288 / 2861349177 \) \* 1024 + # ( 2821543288 / 2861349177 ) * 1024 = 1009.754542 + + # ==> use -repMatch=350 according to size scaled down from 1024 for human. + # and rounded down to nearest 50 + cd /hive/data/genomes/rheMac10 + blat rheMac10.2bit \ + /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/rheMac10.11.ooc \ + -repMatch=1000 + # Wrote 31768 overused 11-mers to jkStuff/rheMac10.11.ooc + # rheMac8 at repMatch=900: + # Wrote 43065 overused 11-mers to jkStuff/rheMac8.11.ooc + + # check non-bridged gaps to see what the typical size is: + hgsql -N \ + -e 'select * from gap where bridge="no" order by size;' rheMac10 \ + | sort -k7,7nr | ave -col=7 stdin + # min 100.000000 + # max 1000000.000000 + + # minimum gap size is 100 and produces a reasonable number of lifts + gapToLift -verbose=2 -minGap=100 rheMac10 jkStuff/rheMac10.nonBridged.lft \ + -bedFile=jkStuff/rheMac10.nonBridged.bed + wc -l jkStuff/rheMac10.nonBri* + # 2979 jkStuff/rheMac10.nonBridged.bed + # 2979 jkStuff/rheMac10.nonBridged.lft + +######################################################################## +# lastz/chain/net swap human/hg38 (TBD - 2018-10-12 - Hiram) + # original alignment + cd /hive/data/genomes/hg38/bed/lastzRheMac10.2018-10-12 + + cat fb.hg38.chainRheMac10Link.txt + # 154079940 bases of 3095998939 (4.977%) in intersection + cat fb.hg38.chainSynRheMac10Link.txt + # 95877644 bases of 3095998939 (3.097%) in intersection + cat fb.hg38.chainRBest.RheMac10.txt + # 106665747 bases of 3095998939 (3.445%) in intersection + + # and for the swap: + mkdir /hive/data/genomes/rheMac10/bed/blastz.hg38.swap + cd /hive/data/genomes/rheMac10/bed/blastz.hg38.swap + + time (doBlastzChainNet.pl -verbose=2 \ + /hive/data/genomes/hg38/bed/lastzRheMac10.2018-10-12/DEF \ + -swap -chainMinScore=5000 -chainLinearGap=loose \ + -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ + -syntenicNet) > swap.log 2>&1 + # real 9m45.514s + + cat fb.rheMac10.chainHg38Link.txt + # 120955955 bases of 1055588482 (11.459%) in intersection + + cat fb.rheMac10.chainSynHg38Link.txt + # 92597630 bases of 1055588482 (8.772%) in intersection + + time (doRecipBest.pl -load -workhorse=hgwdev -buildDir=`pwd` rheMac10 hg38) > rbest.log 2>&1 & + # real 139m24.408s + + cat fb.rheMac10.chainRBest.Hg38.txt + # 106294585 bases of 1055588482 (10.070%) in intersection + +######################################################################### +# lastz/chain/net swap mouse/mm10 (TBD - 2018-10-12 - Hiram) + + # original alignment + cd /hive/data/genomes/mm10/bed/lastzRheMac10.2018-10-12 + cat fb.mm10.chainRheMac10Link.txt + # 101151132 bases of 2652783500 (3.813%) in intersection + cat fb.mm10.chainSynRheMac10Link.txt + # 70707720 bases of 2652783500 (2.665%) in intersection + cat fb.mm10.chainRBest.RheMac10.txt + # 79649474 bases of 2652783500 (3.002%) in intersection + + # and for the swap: + mkdir /hive/data/genomes/rheMac10/bed/blastz.mm10.swap + cd /hive/data/genomes/rheMac10/bed/blastz.mm10.swap + + time (doBlastzChainNet.pl -verbose=2 \ + /hive/data/genomes/mm10/bed/lastzRheMac10.2018-10-12/DEF \ + -swap -chainMinScore=5000 -chainLinearGap=loose \ + -workhorse=hgwdev -smallClusterHub=ku -bigClusterHub=ku \ + -syntenicNet) > swap.log 2>&1 + # real 6m41.043s + + cat fb.rheMac10.chainMm10Link.txt + # 88539346 bases of 1055588482 (8.388%) in intersection + + time (doRecipBest.pl -load -workhorse=hgwdev -buildDir=`pwd` rheMac10 mm10) > rbest.log 2>&1 & + # real 94m11.007s + + cat fb.rheMac10.chainRBest.Mm10.txt + # 79474812 bases of 1055588482 (7.529%) in intersection + +######################################################################### +# GENBANK AUTO UPDATE (DONE - 2019-07-03 - Hiram) + ssh hgwdev + cd $HOME/kent/src/hg/makeDb/genbank + git pull + # /cluster/data/genbank/data/organism.lst shows: + # #organism mrnaCnt estCnt refSeqCnt + # Macaca mulatta 381495 60276 5820 + + # edit etc/genbank.conf to add rheMac10 just before galGal5 + +# rheMac10 Macaca mulatta (rhesus) (refseq assembly GCF_003339765.1) +rheMac10.serverGenome = /hive/data/genomes/rheMac10/rheMac10.2bit +rheMac10.ooc = /hive/data/genomes/rheMac10/jkStuff/rheMac10.11.ooc +rheMac10.lift = /hive/data/genomes/rheMac10/jkStuff/rheMac10.nonBridged.lft +rheMac10.perChromTables = no +rheMac10.refseq.mrna.native.pslCDnaFilter = ${finished.refseq.mrna.native.pslCDnaFilter} +rheMac10.refseq.mrna.xeno.pslCDnaFilter = ${finished.refseq.mrna.xeno.pslCDnaFilter} +rheMac10.genbank.mrna.native.pslCDnaFilter = ${finished.genbank.mrna.native.pslCDnaFilter} +rheMac10.genbank.mrna.xeno.pslCDnaFilter = ${finished.genbank.mrna.xeno.pslCDnaFilter} +rheMac10.genbank.est.native.pslCDnaFilter = ${finished.genbank.est.native.pslCDnaFilter} +rheMac10.genbank.est.xeno.pslCDnaFilter = ${finished.genbank.est.xeno.pslCDnaFilter} +rheMac10.downloadDir = rheMac10 +# DO NOT NEED genbank.mrna.xeno except for human, mouse(and in this case rhesus) +rheMac10.refseq.mrna.native.load = yes +rheMac10.refseq.mrna.xeno.load = yes +rheMac10.upstreamGeneTbl = refGene + + # verify the files specified exist before checking in the file: + grep ^rheMac10 etc/genbank.conf | grep hive | awk '{print $NF}' | xargs ls -og +# -rw-rw-r-- 1 127080 Jul 1 11:03 /hive/data/genomes/rheMac10/jkStuff/rheMac10.11.ooc +# -rw-rw-r-- 1 177516 Jul 1 13:16 /hive/data/genomes/rheMac10/jkStuff/rheMac10.nonBridged.lft +# -rw-rw-r-- 1 775334369 Jul 2 07:48 /hive/data/genomes/rheMac10/rheMac10.2bit + + git commit -m "Added rheMac10 rhesus/Macaca mulatta; refs #23771" etc/genbank.conf + git push + # update /cluster/data/genbank/: + make etc-update + + # enable daily alignment and update of hgwdev + cd ~/kent/src/hg/makeDb/genbank + git pull + # add rheMac10 to: + # etc/hgwdev.dbs + git add etc/align.dbs etc/hgwdev.dbs + git commit -m "Added rheMac10 - rhesus/Macaca mulatta refs #23771" etc/hgwdev.dbs + git push + make etc-update + + # wait a few days for genbank magic to take place, the tracks will + # appear + +############################################################################# +# augustus gene track (DONE - 2019-07-02 - Hiram) + + mkdir /hive/data/genomes/rheMac10/bed/augustus + cd /hive/data/genomes/rheMac10/bed/augustus + time (doAugustus.pl -buildDir=`pwd` -bigClusterHub=ku \ + -species=human -dbHost=hgwdev \ + -workhorse=hgwdev rheMac10) > do.log 2>&1 +XXX - running - Tue Jul 2 07:53:20 PDT 2019 + # real 48m48.597s + + cat fb.rheMac10.augustusGene.txt + # 25827925 bases of 1055588482 (2.447%) in intersection + +######################################################################### +# ncbiRefSeq (DONE - 2019-07-02 - Hiram) + + mkdir /hive/data/genomes/rheMac10/bed/ncbiRefSeq + cd /hive/data/genomes/rheMac10/bed/ncbiRefSeq + # running step wise just to be careful + time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ + -bigClusterHub=ku -dbHost=hgwdev \ + -stop=download -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ + refseq vertebrate_mammalian Macaca_mulatta \ + GCF_003339765.1_Mmul_10 rheMac10) > download.log 2>&1 + # real 2m3.842s + + 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_mammalian Macaca_mulatta \ + GCF_003339765.1_Mmul_10 rheMac10) > process.log 2>&1 + # real 2m6.030s + + time (~/kent/src/hg/utils/automation/doNcbiRefSeq.pl -buildDir=`pwd` \ + -continue=load -bigClusterHub=ku -dbHost=hgwdev \ + -stop=load -fileServer=hgwdev -smallClusterHub=ku -workhorse=hgwdev \ + refseq vertebrate_mammalian Macaca_mulatta \ + GCF_003339765.1_Mmul_10 rheMac10) > load.log 2>&1 + # real 0m27.175s + + cat fb.ncbiRefSeq.rheMac10.txt + # 92462016 bases of 2936892733 (3.148%) in intersection + +XXX TBD + # need to add: include ../../refSeqComposite.ra alpha + # to the rhesus/rheMac10/trackDb.ra to turn on the track in the browser + + # there was one gene that claimed to have a protein, but the + # protein sequence was not included in the protein.faa file + # discovered from joinerCheck + # manual fix to blank out this one protein, to see the entry + hgsql -e 'select * from ncbiRefSeqLink where protAcc="NP_989875.1";' rheMac10 + hgsql -e 'update ncbiRefSeqLink set protAcc="" where protAcc="NP_989875.1";' rheMac10 + # this makes the 'protein' link disappear from the gene details page + # curious that this gene is marked as a non-coding gene anyway ? + # gene: FET1 at chr4:63,102,774-63,105,516- + + featureBits -enrichment rheMac10 refGene ncbiRefSeq + # refGene 1.374%, ncbiRefSeq 8.397%, both 1.370%, cover 99.73%, enrich 11.88x + featureBits -enrichment rheMac10 ncbiRefSeq refGene + # ncbiRefSeq 8.397%, refGene 1.374%, both 1.370%, cover 16.32%, enrich 11.88x + + featureBits -enrichment rheMac10 ncbiRefSeqCurated refGene + # ncbiRefSeqCurated 1.368%, refGene 1.374%, both 1.364%, cover 99.71%, enrich 72.59x + featureBits -enrichment rheMac10 refGene ncbiRefSeqCurated + # refGene 1.374%, ncbiRefSeqCurated 1.368%, both 1.364%, cover 99.32%, enrich 72.59x + +######################################################################### +# LIFTOVER TO rheMac8 (DONE - 2019-07-02 - Hiram) + ssh hgwdev + mkdir /hive/data/genomes/rheMac10/bed/blat.rheMac8.2019-07-02 + cd /hive/data/genomes/rheMac10/bed/blat.rheMac8.2019-07-02 + doSameSpeciesLiftOver.pl -verbose=2 \ + -debug -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ + -ooc=/hive/data/genomes/rheMac10/jkStuff/rheMac10.11.ooc \ + rheMac10 rheMac8 + time (doSameSpeciesLiftOver.pl -verbose=2 \ + -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ + -ooc=/hive/data/genomes/rheMac10/jkStuff/rheMac10.11.ooc \ + rheMac10 rheMac8) > doLiftOverToRheMac8.log 2>&1 +XXX - running - Tue Jul 2 08:11:05 PDT 2019 + # real 156m30.215s + + # see if the liftOver menus function in the browser from rheMac10 to galGal5 + +######################################################################### +# BLATSERVERS ENTRY (TBD - 2018-10-12 - 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 ("rheMac10", "blat1a", "17892", "1", "0"); \ + INSERT INTO blatServers (db, host, port, isTrans, canPcr) \ + VALUES ("rheMac10", "blat1a", "17893", "0", "1");' \ + hgcentraltest + # test it with some sequence + +############################################################################ +## reset default position to MEPE gene (egg shell protein) +## (TBD - 2018-10-12 - Hiram) + + # as found from the galGal5 to rheMac10 liftOver + ssh hgwdev + hgsql -e 'update dbDb set defaultPos="chr4:45667017-45672928" + where name="rheMac10";' hgcentraltest + +######################################################################### +# crispr 10K shoulders (TBD - 2018-10-16 - Hiram) + # working on this script, adding the indexFa step: + time (~/kent/src/hg/utils/automation/doCrispr.pl \ + -stop=indexFa -buildDir=`pwd` -smallClusterHub=ku rheMac10 ncbiRefSeq) \ + > indexFa.log 2>&1 + # real 23m26.694s + + time (~/kent/src/hg/utils/automation/doCrispr.pl \ + -continue=ranges -stop=guides -buildDir=`pwd` -smallClusterHub=ku \ + rheMac10 ncbiRefSeq) > guides.log 2>&1 + # real 2m50.758s + + # adding the /dev/shm/ setup rsync for the indexed Fa + # performed manually to work out the procedure + time (~/kent/src/hg/utils/automation/doCrispr.pl \ + -continue=specScores -stop=specScores -buildDir=`pwd` \ + -smallClusterHub=ku rheMac10 ncbiRefSeq) > specScores.log + + # had about half of ku for about half this time: +# Completed: 884922 of 884922 jobs +# CPU time in finished jobs: 35872791s 597879.85m 9964.66h 415.19d 1.138 y +# IO & Wait Time: 899261s 14987.69m 249.79h 10.41d 0.029 y +# Average job time: 42s 0.69m 0.01h 0.00d +# Longest finished job: 88s 1.47m 0.02h 0.00d +# Submission to last job: 48045s 800.75m 13.35h 0.56d + + + time find tmp/outGuides -type f | xargs cut -f3-6 > ../specScores.tab + # real 236m17.220s + wc -l specScores.tab + # 66451712 specScores.tab + + time (~/kent/src/hg/utils/automation/doCrispr.pl \ + -continue=effScores -stop=load \ + -buildDir=`pwd` -smallClusterHub=ku rheMac10 ncbiRefSeq) \ + > load.log + # real 307m41.143s + +######################################################################### +# all.joiner update, downloads and in pushQ - (TBD - 2018-10-17 - Hiram) +xyz + cd $HOME/kent/src/hg/makeDb/schema + # verify all the business is done for release + ~/kent/src/hg/utils/automation/verifyBrowser.pl rheMac10 + + # fixup all.joiner until this is a clean output + joinerCheck -database=rheMac10 -tableCoverage all.joiner + joinerCheck -database=rheMac10 -times all.joiner + joinerCheck -database=rheMac10 -keys all.joiner + + # when clean, check in: + git commit -m 'adding rules for rheMac10 refs #22113' all.joiner + git push + # run up a 'make alpha' in hg/hgTables to get this all.joiner file + # into the hgwdev/genome-test system + + cd /hive/data/genomes/rheMac10 + time (makeDownloads.pl rheMac10) > downloads.log 2>&1 + # real 10m7.605s + + # now ready for pushQ entry + mkdir /hive/data/genomes/rheMac10/pushQ + cd /hive/data/genomes/rheMac10/pushQ + time (makePushQSql.pl -redmineList rheMac10) > rheMac10.pushQ.sql 2> stderr.out + # real 9m58.779s + + # remove the extra chainNet files from the listings: + sed -i -e "/etNig1/d" redmine.rheMac10.file.list + sed -i -e "/asAcu1/d" redmine.rheMac10.file.list + sed -i -e "/etNig1/d" redmine.rheMac10.table.list + sed -i -e "/onAlb1/d" redmine.rheMac10.table.list + sed -i -e "/asAcu1/d" redmine.rheMac10.table.list + sed -i -e "/Stickleback/d" redmine.rheMac10.releaseLog.txt + sed -i -e "/Tetraodon/d" redmine.rheMac10.releaseLog.txt + sed -i -e "/sparrow/d" redmine.rheMac10.releaseLog.txt + # remove the tandemDups and gapOverlap from the file list: + sed -i -e "/tandemDups/d" redmine.rheMac10.table.list + sed -i -e "/Tandem Dups/d" redmine.rheMac10.releaseLog.txt + sed -i -e "/gapOverlap/d" redmine.rheMac10.table.list + sed -i -e "/Gap Overlaps/d" redmine.rheMac10.releaseLog.txt + # real 7m21.629s + + # check for errors in stderr.out, some are OK, e.g.: + # WARNING: hgwdev does not have /gbdb/rheMac10/wib/gc5Base.wib + # WARNING: hgwdev does not have /gbdb/rheMac10/wib/quality.wib + # WARNING: hgwdev does not have /gbdb/rheMac10/bbi/quality.bw + # WARNING: rheMac10 does not have seq + # WARNING: rheMac10 does not have extFile + + # add the path names to the listing files in the redmine issue + # in the three appropriate entry boxes: + +# /hive/data/genomes/rheMac10/pushQ/redmine.rheMac10.file.list +# /hive/data/genomes/rheMac10/pushQ/redmine.rheMac10.releaseLog.txt +# /hive/data/genomes/rheMac10/pushQ/redmine.rheMac10.table.list + +#########################################################################