f59b87d2038e90c7969da0fa0cc3652d41720881 markd Wed Mar 30 12:34:29 2022 -0700 create hg19/hg39 CHM13 liftover chains using NCBI names for now diff --git src/hg/makeDb/doc/hg19.txt src/hg/makeDb/doc/hg19.txt index db50aa3..2066151 100644 --- src/hg/makeDb/doc/hg19.txt +++ src/hg/makeDb/doc/hg19.txt @@ -34764,36 +34764,53 @@ ############################################################################# # chm13 liftover alignments (2022-03-29 markd) # preliminary CHM13 <-> hg19 liftOver chains until NCBI produces consensus alignments # provide by Nae-Chyun Chen mkdir -p /hive/data/genomes/hg19/bed/chm13LiftOver cd /hive/data/genomes/hg19/bed/chm13LiftOver # Obtain GRCh37 from T2T Globus: team-liftover/v1_nflo/grch37-chm13v2.chain # rename to better match UCSC convetions and compress mv grch37-chm13v2.chain hg19-chm13v2.over.chain pigz hg19-chm13v2.over.chain +# make NCBI query names as well, since chromAlias doesn't work yet for click-through +# note doc is wrong in chromToPsl PSL query is 10 + chainToPslBasic hg19-chm13v2.over.chain.gz stdout | chromToUcsc -k 10 -a /hive/data/genomes/asmHubs/genbankBuild/GCA/009/914/755/GCA_009914755.4_CHM13_T2T_v2.0/GCA_009914755.4_CHM13_T2T_v2.0.chromAlias.txt | pslToChain stdin stdout | pigz -c > hg19-chm13v2.ncbi-qnames.over.chain.gz + # build bigChain files: hgLoadChain -noBin -test none bigChain hg19-chm13v2.over.chain.gz sed 's/\.000000//' chain.tab | awk 'BEGIN {OFS="\t"} {print $2, $4, $5, $11, 1000, $8, $3, $6, $7, $9, $10, $1}' > bigChainIn.tab bedToBigBed -type=bed6+6 -as=${HOME}/kent/src/hg/lib/bigChain.as -tab bigChainIn.tab ../../chrom.sizes hg19-chm13v2.over.chain.bb tawk '{print $1, $2, $3, $5, $4}' link.tab | csort -k1,1 -k2,2n --parallel=64 > bigLinkIn.tab bedToBigBed -type=bed4+1 -as=${HOME}/kent/src/hg/lib/bigLink.as -tab bigLinkIn.tab ../../chrom.sizes hg19-chm13v2.over.link.bb +# build bigChain with NCBI names files: + hgLoadChain -noBin -test none bigChain hg19-chm13v2.ncbi-qnames.over.chain.gz + sed 's/\.000000//' chain.tab | awk 'BEGIN {OFS="\t"} {print $2, $4, $5, $11, 1000, $8, $3, $6, $7, $9, $10, $1}' > bigChainIn.tab + bedToBigBed -type=bed6+6 -as=${HOME}/kent/src/hg/lib/bigChain.as -tab bigChainIn.tab ../../chrom.sizes hg19-chm13v2.ncbi-qnames.over.chain.bb + tawk '{print $1, $2, $3, $5, $4}' link.tab | csort -k1,1 -k2,2n --parallel=64 > bigLinkIn.tab + bedToBigBed -type=bed4+1 -as=${HOME}/kent/src/hg/lib/bigLink.as -tab bigLinkIn.tab ../../chrom.sizes hg19-chm13v2.ncbi-qnames.over.link.bb + + rm *.tab + # link to gbdb mkdir -p /gbdb/hg19/bbi/chm13LiftOver ln -sf $(pwd)/*.bb /gbdb/hg19/bbi/chm13LiftOver/ # make downloads, can't add to liftOver directory due to license in that directory mkdir -p /usr/local/apache/htdocs-hgdownload/goldenPath/hg19/chm13LiftOver/ - ln -sf $(pwd)/hg19-chm13v2.over.chain.gz /usr/local/apache/htdocs-hgdownload/goldenPath/hg19/chm13LiftOver/ + ln -sf $(pwd)/hg19-chm13v2.*.chain.gz /usr/local/apache/htdocs-hgdownload/goldenPath/hg19/chm13LiftOver/ # push these files + /gbdb/hg19/bbi/chm13LiftOver/hg19-chm13v2.ncbi-qnames.over.chain.bb + /gbdb/hg19/bbi/chm13LiftOver/hg19-chm13v2.ncbi-qnames.over.link.bb /gbdb/hg19/bbi/chm13LiftOver/hg19-chm13v2.over.chain.bb /gbdb/hg19/bbi/chm13LiftOver/hg19-chm13v2.over.link.bb + /usr/local/apache/htdocs-hgdownload/goldenPath/hg19/chm13LiftOver/hg19-chm13v2.ncbi-qnames.over.chain.gz /usr/local/apache/htdocs-hgdownload/goldenPath/hg19/chm13LiftOver/hg19-chm13v2.over.chain.gz -# NOTES: chromAlias doesn't work on click through yet, so disable for now + +#############################################################################