# for emacs: -*- mode: sh; -*- # This file describes browser build for the micMur2 # Assembly Name: Mmur_2.0 # Organism name: Microcebus murinus # Taxid: 30608 # Submitter: The Genome Sequencing Platform, The Genome Assembly Team # Date: 2015-5-1 # BioSample: SAMN03121888 # Assembly type: haploid # Release type: major # Assembly level: Scaffold # Genome representation: full # GenBank Assembly Accession: GCA_000165445.2 (species-representative latest) ############################################################################# # fetch sequence from new style download directory (DONE - 2015-05-04 - Hiram) # NCBI has redesigned their FTP download site, new type of address # and naming scheme mkdir -p /hive/data/genomes/micMur2/genbank cd /hive/data/genomes/micMur2/genbank rsync -L -a -P rsync://ftp.ncbi.nlm.nih.gov/genomes/genbank/vertebrate_mammalian/Microcebus_murinus/all_assembly_versions/GCA_000165445.2_Mmur_2.0/ ./ # measure what we have here: faSize GCA_000165445.2_Mmur_2.0_genomic.fna.gz # 2438804424 bases (61028711 N's 2377775713 real 1652977497 upper # 724798216 lower) in 10311 sequences in 1 files # Total size: mean 236524.5 sd 1059982.8 min 271 (ABDC02048825.1) # max 23116325 (KQ053299.1) median 4628 # %29.72 masked total, %30.48 masked real ############################################################################# # fixup to UCSC naming scheme (DONE - 2015-05-04 - Hiram) mkdir /hive/data/genomes/micMur2/ucsc cd /hive/data/genomes/micMur2/ucsc # this one is unusually simple because there is just one file of # contigs, and one AGP file. Even the contig names are already # almost in UCSC format. Just need to substitute v where . is in # the accession suffix zcat ../genbank/GCA_000165445.2_Mmur_2.0_assembly_structure/Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fna.gz \ | sed -e 's/ Microcebus murinus.*//;' \ | sed -e 's/\./v/;' | gzip -c > micMur2.fa.gz zcat ../genbank/GCA_000165445.2_Mmur_2.0_assembly_structure/Primary_Assembly/unplaced_scaffolds/AGP/unplaced.scaf.agp.gz \ | sed -e 's/\./v/;' > micMur2.agp # verify nothing lost compared to the genbank source: faSize micMur2.fa.gz # 2438804424 bases (61028711 N's 2377775713 real 2377775713 upper # 0 lower) in 10311 sequences in 1 files # Total size: mean 236524.5 sd 1059982.8 min 271 (ABDC02048825v1) # max 23116325 (KQ053299v1) median 4628 # same numbers as above. ############################################################################# # Initial database build (DONE - 2015-05-04 - Hiram) cd /hive/data/genomes/micMur2 cat << '_EOF_' > micMur2.config.ra # Config parameters for makeGenomeDb.pl: db micMur2 clade mammal # genomeCladePriority 35 scientificName Microcebus murinus commonName Mouse lemur assemblyDate May 2015 assemblyLabel Broad Institute and Baylor College of Medicine Mmur_2.0 assemblyShortLabel Mouse lemur orderKey 13679 # no mito sequence found at genbank mitoAcc none fastaFiles /cluster/data/micMur2/ucsc/micMur2.fa.gz agpFiles /cluster/data/micMur2/ucsc/micMur2.agp # qualFiles none dbDbSpeciesDir mouseLemur photoCreditURL http://pin.primate.wisc.edu/factsheets/image/623 photoCreditName Verena Behringer, Primate Info Net ncbiGenomeId 777 ncbiAssemblyId 329621 ncbiAssemblyName Mmur_2.0 ncbiBioProject 19967 genBankAccessionID GCA_000165445.2 taxId 30608 '_EOF_' # << happy emacs # verify sequence and AGP are OK: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev -fileServer=hgwdev \ -stop=agp micMur2.config.ra) > agp.log 2>&1 # real 2m20.732s # then finish it off: time (makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=db micMur2.config.ra) > db.log 2>&1 # there was a problem with the last step trackDb, fixed makeGenomeDb.pl # and completed: time nice -n +19 makeGenomeDb.pl -workhorse=hgwdev -dbHost=hgwdev \ -fileServer=hgwdev -continue=trackDb micMur2.config.ra > trackDb.log 2>&1 # real 19m47.412s ########################################################################## # running repeat masker (DONE - 2015-05-04 - Hiram) mkdir /hive/data/genomes/micMur2/bed/repeatMasker cd /hive/data/genomes/micMur2/bed/repeatMasker time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -smallClusterHub=ku micMur2) > do.log 2>&1 & # real 467m4.998s # two bad records: RepeatMasker bug?: Undefined id, line 2869570 of input: 486 35.2 2.4 4.3 KQ057404v1 1319743 1320042 (2310255) C L1MA8 LINE/L1 (6499) 1002 709 RepeatMasker bug?: Undefined id, line 2869572 of input: 1383 18.3 0.0 0.8 KQ057404v1 1320464 1320700 (2309597) C L1MA8 LINE/L1 (6789) 712 478 # take them out: egrep -v "KQ057404v1 1319743 1320042|KQ057404v1 1320464 1320700" \ micMur2.sorted.fa.out > cleaned.sorted.fa.out mv micMur2.sorted.fa.out broken.sorted.fa.out mv cleaned.sorted.fa.out micMur2.sorted.fa.out /cluster/bin/scripts/extractNestedRepeats.pl micMur2.sorted.fa.out \ | sort -k1,1 -k2,2n > micMur2.nestedRepeats.bed time (doRepeatMasker.pl -buildDir=`pwd` \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ -continue=mask -smallClusterHub=ku micMur2) > mask.log 2>&1 # real 22m7.705s cat faSize.rmsk.txt # 2438804424 bases (61028711 N's 2377775713 real 1399954637 upper # 977821076 lower) in 10311 sequences in 1 files # Total size: mean 236524.5 sd 1059982.8 min 271 (ABDC02048825v1) # max 23116325 (KQ053299v1) median 4628 # %40.09 masked total, %41.12 masked real egrep -i "versi|relea" do.log # January 31 2015 (open-4-0-5) version of RepeatMasker # CC RELEASE 20140131; time featureBits -countGaps micMur2 rmsk # 978209828 bases of 2438804424 (40.110%) in intersection # real 0m31.168s # why is it different than the faSize above ? # because rmsk masks out some N's as well as bases, the count above # separates out the N's from the bases, it doesn't show lower case N's ########################################################################## # running simple repeat (DONE 2015-05-04 - Hiram) mkdir /hive/data/genomes/micMur2/bed/simpleRepeat cd /hive/data/genomes/micMur2/bed/simpleRepeat time (doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=ku \ -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=ku \ micMur2) > do.log 2>&1 # real 5m55.131s cat fb.simpleRepeat # 35966576 bases of 2377775715 (1.513%) in intersection # add to rmsk after it is done: cd /hive/data/genomes/micMur2 twoBitMask micMur2.rmsk.2bit \ -add bed/simpleRepeat/trfMask.bed micMur2.2bit # you can safely ignore the warning about fields >= 13 twoBitToFa micMur2.2bit stdout | faSize stdin > faSize.micMur2.2bit.txt cat faSize.micMur2.2bit.txt # 2438804424 bases (61028711 N's 2377775713 real 1398397964 upper # 979377749 lower) in 10311 sequences in 1 files # Total size: mean 236524.5 sd 1059982.8 min 271 (ABDC02048825v1) # max 23116325 (KQ053299v1) median 4628 # %40.16 masked total, %41.19 masked real rm /gbdb/micMur2/micMur2.2bit ln -s `pwd`/micMur2.2bit /gbdb/micMur2/micMur2.2bit ########################################################################## # CREATE MICROSAT TRACK (DONE - 2015-06-22 - Hiram) ssh hgwdev mkdir /cluster/data/micMur2/bed/microsat cd /cluster/data/micMur2/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 micMur2 microsat microsat.bed # Read 27836 elements of size 4 from microsat.bed ########################################################################## ## WINDOWMASKER (DONE - 2015-05-04 - Hiram) mkdir /hive/data/genomes/micMur2/bed/windowMasker cd /hive/data/genomes/micMur2/bed/windowMasker time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -dbHost=hgwdev micMur2) > do.log 2>&1 # real 161m47.327s # got done before rmsk, finish up here: time featureBits -countGaps micMur2 rmsk windowmaskerSdust \ > fb.micMur2.rmsk.windowmaskerSdust.txt 2>&1 # real 2m1.625s # Masking statistics cat faSize.micMur2.cleanWMSdust.txt # 2438804424 bases (61028711 N's 2377775713 real 1636436909 upper # 741338804 lower) in 10311 sequences in 1 files # Total size: mean 236524.5 sd 1059982.8 min 271 (ABDC02048825v1) # max 23116325 (KQ053299v1) median 4628 # %30.40 masked total, %31.18 masked real cat fb.micMur2.rmsk.windowmaskerSdust.txt # 470271114 bases of 2438804424 (19.283%) in intersection time (doWindowMasker.pl -buildDir=`pwd` -workhorse=hgwdev \ -continue=cleanup -dbHost=hgwdev micMur2) > cleanup.log 2>&1 # real 1m41.675s ########################################################################## # cpgIslands - (DONE - 2015-05-05 - Hiram) mkdir /hive/data/genomes/micMur2/bed/cpgIslands cd /hive/data/genomes/micMur2/bed/cpgIslands time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku \ -workhorse=hgwdev -smallClusterHub=ku micMur2) > do.log 2>&1 & # real 7m51.794s cat fb.micMur2.cpgIslandExt.txt # 52444688 bases of 2377775715 (2.206%) in intersection ############################################################################## # cpgIslands on UNMASKED sequence (DONE - 2015-05-04 - Hiram) mkdir /hive/data/genomes/micMur2/bed/cpgIslandsUnmasked cd /hive/data/genomes/micMur2/bed/cpgIslandsUnmasked # run stepwise so the loading can be done in a different table time (doCpgIslands.pl -dbHost=hgwdev -bigClusterHub=ku -buildDir=`pwd` \ -tableName=cpgIslandExtUnmasked \ -maskedSeq=/hive/data/genomes/micMur2/micMur2.unmasked.2bit \ -workhorse=hgwdev -smallClusterHub=ku micMur2) > do.log 2>&1 # real 33m50.982s cat fb.micMur2.cpgIslandExtUnmasked.txt # 63436634 bases of 2377775715 (2.668%) in intersection ############################################################################# # cytoBandIdeo - (DONE - 2015-05-04 - Hiram) mkdir /hive/data/genomes/micMur2/bed/cytoBand cd /hive/data/genomes/micMur2/bed/cytoBand makeCytoBandIdeo.csh micMur2 ######################################################################### # genscan - (DONE - 2015-05-05 - Hiram) mkdir /hive/data/genomes/micMur2/bed/genscan cd /hive/data/genomes/micMur2/bed/genscan time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -bigClusterHub=ku micMur2) > do.log 2>&1 & # real 11m8.384s # two broken jobs finished manually with 2,000,000 window, continuing: time (doGenscan.pl -buildDir=`pwd` -workhorse=hgwdev -dbHost=hgwdev \ -continue=makeBed -bigClusterHub=ku micMur2) > makeBed.log 2>&1 & # real 2m32.761s cat fb.micMur2.genscan.txt # 61639735 bases of 2377775715 (2.592%) in intersection cat fb.micMur2.genscanSubopt.txt # 59114180 bases of 2377775715 (2.486%) in intersection ######################################################################## # Create kluster run files (DONE - 2015-05-05 - Hiram) # numerator is micMur2 gapless bases "real" as reported by: head -1 faSize.micMur2.2bit.txt # 2438804424 bases (61028711 N's 2377775713 real 1398397964 upper # 979377749 lower) in 10311 sequences in 1 files # 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 \( 2377775713 / 2861349177 \) \* 1024 # ( 2377775713 / 2861349177 ) * 1024 = 850.942048 # ==> use -repMatch=850 according to size scaled down from 1024 for human. # and rounded down to nearest 50 cd /hive/data/genomes/micMur2 time blat micMur2.2bit \ /dev/null /dev/null -tileSize=11 -makeOoc=jkStuff/micMur2.11.ooc \ -repMatch=850 # Wrote 23327 overused 11-mers to jkStuff/micMur2.11.ooc # real 0m39.187s # there are no non-bridged gaps, don't need this # check non-bridged gaps to see what the typical size is: hgsql -N -e 'select * from gap where bridge="no" order by size;' \ micMur2 | ave -col=7 stdin # this does nothing # gapToLift -verbose=2 -minGap=100 bosTau7 jkStuff/nonBridged.lft \ # -bedFile=jkStuff/nonBridged.bed ######################################################################## # GENBANK AUTO UPDATE (DONE - 2015-05-28 - Hiram) ssh hgwdev cd $HOME/kent/src/hg/makeDb/genbank git pull # /cluster/data/genbank/data/organism.lst shows: # #organism mrnaCnt estCnt refSeqCnt # Microcebus murinus 59 1 0 # edit etc/genbank.conf to add micMur2 just before micMur1 # micMur2 (Mouse lemur) micMur2.serverGenome = /hive/data/genomes/micMur2/micMur2.2bit micMur2.clusterGenome = /hive/data/genomes/micMur2/micMur2.2bit micMur2.ooc = /hive/data/genomes/micMur2/jkStuff/micMur2.11.ooc micMur2.lift = no micMur2.refseq.mrna.native.pslCDnaFilter = ${ordered.refseq.mrna.native.pslCDnaFilter} micMur2.refseq.mrna.xeno.pslCDnaFilter = ${ordered.refseq.mrna.xeno.pslCDnaFilter} micMur2.genbank.mrna.native.pslCDnaFilter = ${ordered.genbank.mrna.native.pslCDnaFilter} micMur2.genbank.mrna.xeno.pslCDnaFilter = ${ordered.genbank.mrna.xeno.pslCDnaFilter} micMur2.genbank.est.native.pslCDnaFilter = ${ordered.genbank.est.native.pslCDnaFilter} micMur2.refseq.mrna.native.load = no micMur2.refseq.mrna.xeno.load = yes # DO NOT NEED genbank.mrna.xeno except for human, mouse micMur2.genbank.mrna.xeno.load = no micMur2.genbank.est.native.load = no micMur2.downloadDir = micMur2 micMur2.perChromTables = no git commit -m "Added micMur2; refs #15292" etc/genbank.conf git push # update /cluster/data/genbank/: make etc-update # Edit src/lib/gbGenome.c to add new species. Skipped screen # control this business with a screen since it takes a while cd /cluster/data/genbank time ./bin/gbAlignStep -initial micMur2 # logFile: var/build/logs/2015.05.05-08:46:58.micMur2.initalign.log # real 1605m25.150s # To re-do, rm the dir first: # /cluster/data/genbank/work/initial.micMur2 # load database when finished ssh hgwdev cd /cluster/data/genbank time ./bin/gbDbLoadStep -drop -initialLoad micMur2 # logFile: var/dbload/hgwdev/logs/2015.05.22-14:24:50.micMur2.dbload.log # something failed in this, Brian got it loaded # enable daily alignment and update of hgwdev cd ~/kent/src/hg/makeDb/genbank git pull # add micMur2 to: # vi etc/align.dbs etc/hgwdev.dbs git commit -m "Added micMur2 - Mouse lemur refs #15292" etc/align.dbs etc/hgwdev.dbs git push make etc-update ######################################################################### # ucscToINSDC table/track (DONE - 2015-05-05 - Hiram) mkdir /hive/data/genomes/micMur2/bed/ucscToINSDC cd /hive/data/genomes/micMur2/bed/ucscToINSDC # check for chrM in assembly: grep chrM ../../micMur2.agp # (there is no chrM) # use the accession name from there in this command (blank if none) ~/kent/src/hg/utils/automation/ucscToINSDC.sh \ ../../genbank/GCA_*assembly_structure/Primary_Assembly awk '{printf "%s\t0\t%d\n", $1,$2}' ../../chrom.sizes \ | sort > name.coordinate.tab join name.coordinate.tab ucscToINSDC.txt | tr '[ ]' '[\t]' \ > ucscToINSDC.bed # should all be the same line count: wc -l * # 10311 name.coordinate.tab # 10311 ucscToINSDC.bed # 10311 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 micMur2 ucscToINSDC stdin ucscToINSDC.bed checkTableCoords micMur2 # should cover %100 entirely: featureBits -countGaps micMur2 ucscToINSDC # 3453864774 bases of 3453864774 (100.000%) in intersection ######################################################################### # fixup search rule for assembly track/gold table (DONE - 2014-05-01 - Hiram) hgsql -N -e "select frag from gold;" micMur2 | sort | head -1 ABDC02000001.1 hgsql -N -e "select frag from gold;" micMur2 | sort | tail -2 ABDC02056280.1 ABDC02056281.1 # verify this rule will find them all or eliminate them all: hgsql -N -e "select frag from gold;" micMur2 | wc -l # 56281 hgsql -N -e "select frag from gold;" micMur2 \ | egrep -e 'ABDC020[0-9]+(\.1)?' | wc -l # 56281 hgsql -N -e "select frag from gold;" micMur2 \ | egrep -v -e 'ABDC020[0-9]+(\.1)?' | wc -l # 0 # hence, add to trackDb/tarsier/micMur2/trackDb.ra searchTable gold shortCircuit 1 termRegex ABDC020[0-9]+(\.1)? query select chrom,chromStart,chromEnd,frag from %s where frag like '%s%%' searchPriority 8 ######################################################################### # LIFTOVER TO micMur1 (DONE - 2015-05-05 - Hiram) ssh hgwdev mkdir /hive/data/genomes/micMur2/bed/blat.micMur1.2015-05-05 cd /hive/data/genomes/micMur2/bed/blat.micMur1.2015-05-05 time (doSameSpeciesLiftOver.pl -verbose=2 -buildDir=`pwd` \ -ooc=/hive/data/genomes/micMur2/jkStuff/micMur2.11.ooc \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ micMur2 micMur1) > do.log 2>&1 # real 595m2.431s # verify the convert link on the test browser is now active from micMur2 to # micMur1 ######################################################################### # LIFTOVER TO micMur3 (DONE - 2017-03-03 - Hiram) ssh hgwdev mkdir /hive/data/genomes/micMur2/bed/blat.micMur3.2017-03-03 cd /hive/data/genomes/micMur2/bed/blat.micMur3.2017-03-03 time (doSameSpeciesLiftOver.pl -verbose=2 -buildDir=`pwd` \ -ooc=/hive/data/genomes/micMur2/jkStuff/micMur2.11.ooc \ -bigClusterHub=ku -dbHost=hgwdev -workhorse=hgwdev \ micMur2 micMur3) > do.log 2>&1 & # real 3321m59.899s # verify the convert link on the test browser is now active from micMur2 to # micMur3 ######################################################################### # all.joiner update, downloads and in pushQ - (DONE 2014-12-29 - Hiram) cd $HOME/kent/src/hg/makeDb/schema # fixup all.joiner until this is a clean output joinerCheck -database=micMur2 -tableCoverage all.joiner joinerCheck -database=micMur2 -times all.joiner joinerCheck -database=micMur2 -keys all.joiner # about 2 minutes cd /hive/data/genomes/micMur2 time makeDownloads.pl micMur2 > downloads.log 2>&1 # real 16m40.848s # now ready for pushQ entry mkdir /hive/data/genomes/micMur2/pushQ cd /hive/data/genomes/micMur2/pushQ time (makePushQSql.pl micMur2 > micMur2.pushQ.sql) 2> stderr.out # real 5m21.558s # check for errors in stderr.out, some are OK, e.g.: # WARNING: hgwdev does not have /gbdb/micMur2/wib/gc5Base.wib # WARNING: hgwdev does not have /gbdb/micMur2/wib/quality.wib # WARNING: hgwdev does not have /gbdb/micMur2/bbi/qualityBw/quality.bw # WARNING: micMur2 does not have seq # WARNING: micMur2 does not have estOrientInfo # copy it to hgwbeta scp -p micMur2.pushQ.sql qateam@hgwbeta:/tmp ssh qateam@hgwbeta "./bin/x86_64/hgsql qapushq < /tmp/micMur2.pushQ.sql" # in that pushQ entry walk through each entry and see if the # sizes will set properly ######################################################################### # ncbiRefSeq gene track (DONE - 2016-05-05 - Hiram) mkdir /hive/data/genomes/micMur2/bed/ncbiRefSeq cd /hive/data/genomes/micMur2/bed/ncbiRefSeq # working on this script, running step by step: time (/cluster/home/hiram/kent/src/hg/utils/automation/doNcbiRefSeq.pl \ -stop=load -buildDir=`pwd` -bigClusterHub=ku \ -fileServer=hgwdev -workhorse=hgwdev -smallClusterHub=ku -dbHost=hgwdev \ refseq vertebrate_mammalian Microcebus_murinus \ GCF_000165445.1_Mmur_2.0 micMur2) > load.log 2>&1 # real 12m36.320s time (/cluster/home/hiram/kent/src/hg/utils/automation/doNcbiRefSeq.pl \ -continue=process -stop=load -buildDir=`pwd` -bigClusterHub=ku \ -fileServer=hgwdev -workhorse=hgwdev -smallClusterHub=ku -dbHost=hgwdev \ refseq vertebrate_mammalian Microcebus_murinus \ GCF_000165445.1_Mmur_2.0 micMur2) > load.log 2>&1 # real 5m29.483s cat fb.ncbiRefSeq.micMur2.txt # 90551422 bases of 2377775715 (3.808%) in intersection ##############################################################################