src/hg/makeDb/doc/felCatV17e.txt 1.9

1.9 2010/04/27 19:20:00 chinhli
Add multiz 6 ways support
Index: src/hg/makeDb/doc/felCatV17e.txt
===================================================================
RCS file: /projects/compbio/cvsroot/kent/src/hg/makeDb/doc/felCatV17e.txt,v
retrieving revision 1.8
retrieving revision 1.9
diff -b -B -U 1000000 -r1.8 -r1.9
--- src/hg/makeDb/doc/felCatV17e.txt	9 Apr 2010 18:41:24 -0000	1.8
+++ src/hg/makeDb/doc/felCatV17e.txt	27 Apr 2010 19:20:00 -0000	1.9
@@ -1,720 +1,845 @@
 # for emacs: -*- mode: sh; -*-
 
 #       $Id$
 
 # Felis Catus (domestic cat) --  NHGRI/GTB V17e/felCatV17e (2009-05-25)
   
 
 # file template copied from calJac3.txt
 
 
 # Felis catus (Project ID: 32759) by NHGRI/Genome Technology Branch [Draft
 #    assembly] sequence: 
 # ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/
 #       Felis_catus/catChrV17e
 #       Felis catus
 
 ##########################################################################
 # Download sequence (DONE - 2010-03-03 Chin)
     mkdir /hive/data/genomes/felCatV17e
     cd /hive/data/genomes/felCatV17e
     mkdir genbank
     cd genbank
 wget --timestamping -r --cut-dirs=6 --level=0 -nH -x \
         --no-remove-listing -np \
 "ftp://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Felis_catus/catChrV17e/*"
     # FINISHED --09:05:15--
     # Downloaded: 151 files, 1.3G in 7m 42s (2.98 MB/s)
     # Read ASSEMBLY_INFO 
 
     mkdir ucscChr
     # stay at genbank directory
     #   fixup the accession names to become UCSC chrom names
 
 S=Primary_Assembly/assembled_chromosomes
 cut -f1 ${S}/chr2acc  | while read C
 do
     ACC=`grep "${C}" ${S}/chr2acc | cut -f2`
     echo "${ACC} -> chr${C}"
     zcat ${S}/AGP/chr${C}.agp.gz \
         | sed -e "s/^${ACC}/chr${C}/" | gzip > ucscChr/chr${C}.agp.gz
 done
 
 # use component instead 
 S=Primary_Assembly/assembled_chromosomes
 cut -f1 ${S}/chr2acc  | while read C
 do
     ACC=`grep "${C}" ${S}/chr2acc | cut -f2`
     echo "${ACC} -> chr${C}"
     zcat ${S}/AGP/chr${C}.comp.agp.gz \
         | sed -e "s/^${ACC}/chr${C}/" | gzip > ucscChr/chr${C}.comp.agp.gz
 done
 
 
 S=Primary_Assembly/assembled_chromosomes
 cut -f1 ${S}/chr2acc  | while read C
 do
     ACC=`grep "${C}" ${S}/chr2acc | cut -f2`
     echo "${ACC} -> chr${C}"
     echo ">chr${C}" > ucscChr/chr${C}.fa
     zcat ${S}/FASTA/chr${C}.fa.gz | grep -v "^>" >> ucscChr/chr${C}.fa
     gzip ucscChr/chr${C}.fa &
 done
    # Check them with faSize 
    faSize Primary_Assembly/assembled_chromosomes/FASTA/chr*.fa.gz
    # 2872644707 bases (1165972091 N's 1706672616 real 1706672616 upper 0
    #       lower) in 19 sequences in 19 files
    faSize ucscChr/chr*.fa.gz
    # 2872644707 bases (1165972091 N's 1706672616 real 1706672616 upper 0
    #        lower) in 19 sequences in 19 files
 
 
    # For unplaced scalfolds, named them as chrUn_xxxxxxxx
    #   where xxxxxx is the original access id as: chrUn_ACBE01000381.1 
    # and put it into chrUn.* files 
    # copy all the comment lines (start with #)
 S=Primary_Assembly/unplaced_scaffolds
 zcat ${S}/AGP/unplaced.scaf.agp.gz | grep "^#" > ucscChr/chrUn.agp
    # append the gap records
 zcat ${S}/AGP/unplaced.scaf.agp.gz | grep -v "^#" \
         | sed -e "s/^/chrUn_/" >> ucscChr/chrUn.agp
 gzip ucscChr/chrUn.agp &
 
 zcat ${S}/FASTA/unplaced.scaf.fa.gz \
         | sed -e "s#^>.*|gb|#>chrUn_#; s#|.*##" | gzip > ucscChr/chrUn.fa.gz
    # about 104034 sequences in the unplaced
 zcat ucscChr/chrUn.fa.gz | grep "^>" | wc
    #  104034  104034 2275648
 
    # Check them with faSize 
    faSize Primary_Assembly/unplaced_scaffolds/FASTA/unplaced.scaf.fa.gz 
    # 287642232 bases (3696852 N's 283945380 real 283945380 upper 0 
    #       lower) in 104034 sequences in 1 files
    faSize ucscChr/chrUn.fa.gz
    # 287642232 bases (3696852 N's 283945380 real 283945380 upper 0
    #        lower) in 104034 sequences in 1 files
 
 
 ##########################################################################
 # Initial genome build (Done - 2010-03-03 -Chin)
 
 # decide the right order by
  hgsql -e "select orderKey,name from dbDb order by orderKey;" \
  hgcentraltest | less
     # 168     cavPor3
     # 169     cavPor2
     # 189     oryCun2
     # 190     oryCun1
     # 191     ochPri2
     # 200     eriEur1
     # 217     felCat3
  
 
     cd /hive/data/genomes/felCatV17e
 
     cat << '_EOF_' > felCatV17e.config.ra
 # Config parameters for makeGenomeDb.pl:
 db felCatV17e
 clade mammal
 genomeCladePriority 16
 scientificName Felis catus
 commonName Cat
 assemblyDate Dec. 2008
 assemblyLabel NHGRI/Genome Technology Branch (NCBI project 10703, accession ACBE0100000)
 assemblyShortLabel NHGRI/GTB V17e
 orderKey 216
 mitoAcc NC_001700
 fastaFiles /hive/data/genomes/felCatV17e/genbank/ucscChr/chr*.fa.gz
 agpFiles /hive/data/genomes/felCatV17e/genbank/ucscChr/chr*.agp.gz
 # qualFiles none
 dbDbSpeciesDir cat
 taxId 9685
 '_EOF_'
 
 
 makeGenomeDb.pl -stop seq felCatV17e.config.ra > seq.log 2>&1 &
 time makeGenomeDb.pl -continue=agp -stop=agp felCatV17e.config.ra > agp.log 2>&1 &
 #       real    0m50.486s
 time makeGenomeDb.pl -continue=db -stop=db felCatV17e.config.ra > db.log 2>&1 &
 #real    7m50.591s
 
 time makeGenomeDb.pl -continue=dbDb -stop=dbDb felCatV17e.config.ra > dbDb.log 2>&1 &
 time makeGenomeDb.pl -continue=trackDb felCatV17e.config.ra > trackDb.log 2>&1 &
 
 #### 03/15 fixing the agp using the component one
 #          manually run hgLoadGapGl (as in jkStuff/makeDb.csh) 
 #          without rerun the makeGenomeDb -continue=gap step
  
 zcat genbank/ucscChr/chr*.comp.agp.gz | grep -v '^#' \
     | sort -k1,1 -k2n,2n > felCatV17e.comp.agp
 
 zcat genbank/ucscChr/chrUn.agp.gz  | grep -v '^#' >> felCatV17e.comp.agp
 grep chrM felCatV17e.agp >> felCatV17e.comp.agp
 checkAgpAndFa felCatV17e.comp.agp felCatV17e.unmasked.2bit
     # loop: chrM, dnaOffset=0, seqSize=17009
     # chrM 1 17009 230 F NC_001700 1 17009 +
     # agpFrag->chromStart: 0, agpFrag->chromEnd: 17009, dnaOffset: 0
     # FASTA sequence entry
     # valid Fasta file entry
     # all AGP and FASTA entries agree - both files are valid
 
   dbSnoop felCatV17e before.txt 
 hgGoldGapGl -noGl felCatV17e felCatV17e.comp.agp
   dbSnoop felCatV17e after.txt
   diff before.txt after.txt # to see the change of rows in gold and gap tables
     # > 35,861,584    gold    506,133 24,249,424      11,612,160
     # > 28,192,744    gap     500,507 22,068,200      6,124,544
     # 26,27d27
     # < 858,728       gold    9,961   539,240 319,488
     # < 345,208       gap     4,335   236,664 108,544
 
 featureBits -countGaps felCatV17e gap
     # 1169658726 bases of 3160303948 (37.011%) in intersection
 hgsql -e "select type from gap;" felCatV17e | sort | uniq -c
     #	381 contig
     # 500126 fragment
     #	1 type
 
 mv felCatV17e.comp.agp felCatV17e.agp
 ##########################################################################
 # fixed up the html files (working - 2010-04-06 Chin)
 
    #  per instructions at end of makeGenomeDb.log, edit the html file
    #  and checkin the *.ra and *.html files.
    cd  /cluster/home/chinhli/kent/src/hg/makeDb/trackDb/cat/felCatV17e
 
 
 
 ##########################################################################
 # running repeat masker (DONE - 2010-03-05 - Chin)
     mkdir /hive/data/genomes/felCatV17e/bed/repeatMasker
     cd /hive/data/genomes/felCatV17e/bed/repeatMasker
     time doRepeatMasker.pl -buildDir=`pwd` -noSplit \
         -bigClusterHub=swarm -dbHost=hgwdev -workhorse=hgwdev \
         -smallClusterHub=encodek felCatV17e > do.log 2>&1 &
     # real    530m25.613s
 
     cat faSize.rmsk.txt
     # 3160303948 bases (1169668943 N's 1990635005 real 1183631401 
     # upper 807003604 lower) in 104054 sequences in 1 files
     # %25.54 masked total, %40.54 masked real
 
 ##########################################################################
 # running simple repeat (DONE - 2010-03-16 Chin)
     mkdir /hive/data/genomes/felCatV17e/bed/simpleRepeat
     cd /hive/data/genomes/felCatV17e/bed/simpleRepeat
     time doSimpleRepeat.pl -buildDir=`pwd` -bigClusterHub=swarm \
         -dbHost=hgwdev -workhorse=hgwdev -smallClusterHub=encodek \
         felCatV17e > do.log 2>&1 &
     #  reall    149m19.948s
     cat fb.simpleRepeat
     #   49107705 bases of 2690950627 (1.825%) in intersection
 
 
     cd /hive/data/genomes/felCatV17e
     twoBitMask felCatV17e.rmsk.2bit \
         -add bed/simpleRepeat/trfMask.bed felCatV17e.2bit
     #   you can safely ignore the warning about fields >= 13
 
     twoBitToFa felCatV17e.2bit stdout | faSize stdin > faSize.felCatV17e.2bit.txt
     cat faSize.felCatV17e.2bit.txt
     #  3160303948 bases (1169668943 N's 1990635005 real 1182766163 upper
     #   807868842 lower) in 104054 sequences in 1 files
     #   %25.56 masked total, %40.58 masked real
 
     # double check with featureBits
     featureBits -countGaps felCatV17e gap
     #  1169668943 bases of 3160303948 (37.011%) in intersection
 
     rm /gbdb/felCatV17e/felCatV17e.2bit
     ln -s `pwd`/felCatV17e.2bit /gbdb/felCatV17e/felCatV17e.2bit
 
 
 
 ########################################################################
 # Marking *all* gaps - they are not all in the AGP file
 #       (DONE - 2010-03-16 - Chin)
     mkdir /hive/data/genomes/felCatV17e/bed/allGaps
     cd /hive/data/genomes/felCatV17e/bed/allGaps
 
     time nice -n +19 findMotif -motif=gattaca -verbose=4 \
         -strand=+ ../../felCatV17e.rmsk.2bit > findMotif.txt 2>&1
     #   real    1m45.351s
 
     grep "^#GAP " findMotif.txt | sed -e "s/^#GAP //" > allGaps.bed
     featureBits felCatV17e -not gap -bed=notGap.bed
     #  1990645222 bases of 1990645222 (100.000%) in intersection
     featureBits felCatV17e allGaps.bed notGap.bed -bed=new.gaps.bed
     #  10217 bases of 1990645222 (0.001%) in intersection
     #    (real time 7+ hours)
     #   what is the last index in the existing gap table:
     hgsql -N -e "select ix from gap;" felCatV17e | sort -n | tail -1
     #   101050
 
     cat << '_EOF_' > mkGap.pl
 #!/usr/bin/env perl
 
 use strict;
 use warnings;
 
 my $ix=`hgsql -N -e "select ix from gap;" felCatV17e | sort -n | tail -1`;
 chomp $ix;
 
 open (FH,"<new.gaps.bed") or die "can not read new.gaps.bed";
 while (my $line = <FH>) {
     my ($chrom, $chromStart, $chromEnd, $rest) = split('\s+', $line);
     ++$ix;
     printf "%s\t%d\t%d\t%d\tN\t%d\tother\tyes\n", $chrom, $chromStart,
         $chromEnd, $ix, $chromEnd-$chromStart;
 }
 close (FH);
 '_EOF_'
     # << happy emacs
     chmod +x ./mkGap.pl
     ./mkGap.pl > other.gap
     hgLoadBed -sqlTable=$HOME/kent/src/hg/lib/gap.sql \
         -noLoad felCatV17e otherGap other.gap
     #   Reading other.gap
     #   Loaded 1294 elements of size 8
     #   Sorted
     #   Saving bed.tab
     #   No load option selected, see file: bed.tab
     wc -l bed.tab
     #   1294 bed.tab
     #   starting with this many
     hgsql -e "select count(*) from gap;" felCatV17e
     #   500507
     hgsql felCatV17e -e 'load data local infile "bed.tab" into table gap;'
     #   result count:
     hgsql -e "select count(*) from gap;" felCatV17e
     #   501801
     # == 500507 + 1294
 
 #########################################################################
 # MAKE 11.OOC FILES FOR BLAT (DONE - 2010-02-03 - Chin)
     cat faSize.felCatV17e.2bit.txt 
     #  3160303948 bases (1169668943 N's 1990635005 real 1182766163 upper 
     #       807868842 lower) in 104054 sequences in 1 files
     #       %25.56 masked total, %40.58 masked real
 
     # numerator is felCatV17e gapless bases "real" as reported by faSize 
     # denominator is hg17 gapless bases as reported by featureBits,
     # 1024 is threshold used for human -repMatch:
     calc \( 1990635005 /  2897310462 \) \* 1024
     #  ( 1990635005 / 2897310462 ) * 1024 = 703.552578
     # ==> use -repMatch=700 according to size scaled down from 1024 for human.
     #   and rounded down to nearest 50
     cd /hive/data/genomes/felCatV17e
     blat felCatV17e.2bit /dev/null /dev/null -tileSize=11 \
       -makeOoc=jkStuff/felCatV17e.11.ooc -repMatch=700
     #   Wrote 23033 overused 11-mers to jkStuff/felCatV17e.11.ooc
 
     mkdir /hive/data/staging/data/felCatV17e
     cp -p felCatV17e.2bit chrom.sizes jkStuff/felCatV17e.11.ooc \
         /hive/data/staging/data/felCatV17e
 
     gapToLift -bedFile=jkStuff/nonBridgedGaps.bed felCatV17e \
         jkStuff/felCatV17e.nonBridged.lft
 
     # Ask the admin to copy the 
     #  /hive/data/staging/data/felCatV17e directory
     #  to cluster nodes:   /scratch/data/felCatV17e
 
 ##########################################################################
 #  BLATSERVERS ENTRY (DONE - 2009-12-23 - Chin)
 #       After getting a blat server assigned by the Blat Server Gods,
 # 
 
     hgsql -e 'INSERT INTO blatServers (db, host, port, isTrans, canPcr) \
         VALUES ("felCatV17e", "blat4", "17778", "1", "0"); \
         INSERT INTO blatServers (db, host, port, isTrans, canPcr) \
         VALUES ("felCatV17e", "blat4", "17779", "0", "1");' \
             hgcentraltest
     #   test it with some sequence
 
 ############################################################################
 # chrA2:64650765-64656835
 # reset position to RHO location as found from blat of hg19 RHO gene
     hgsql -e \
 'update dbDb set defaultPos="chrA2:64650765-64656835" where 
   name="felCatV17e";' \
         hgcentraltest
 ############################################################################
 # genbank run DONE - 2010-03-16 - Chin )
     ssh hgwdev
     cd $HOME/kent/src/hg/makeDb/genbank
     # edit etc/genbank.conf to add this section just before calJac1:
 # Marmoset
 felCatV17e.serverGenome = /hive/data/genomes/felCatV17e/felCatV17e.2bit
 felCatV17e.clusterGenome = /scratch/data/felCatV17e/felCatV17e.2bit
 felCatV17e.ooc = /scratch/data/felCatV17e/felCatV17e.11.ooc
 felCatV17e.lift = no
 felCatV17e.perChromTables = no
 felCatV17e.refseq.mrna.native.pslCDnaFilter  = ${ordered.refseq.mrna.native.pslCDnaFilter}
 felCatV17e.refseq.mrna.xeno.pslCDnaFilter    = ${ordered.refseq.mrna.xeno.pslCDnaFilter}
 felCatV17e.genbank.mrna.native.pslCDnaFilter = ${ordered.genbank.mrna.native.pslCDnaFilter}
 felCatV17e.genbank.mrna.xeno.pslCDnaFilter   = ${ordered.genbank.mrna.xeno.pslCDnaFilter}
 felCatV17e.genbank.est.native.pslCDnaFilter  = ${ordered.genbank.est.native.pslCDnaFilter}
 felCatV17e.genbank.est.xeno.pslCDnaFilter    = ${ordered.genbank.est.xeno.pslCDnaFilter}
 felCatV17e.downloadDir = felCatV17e
 felCatV17e.refseq.mrna.native.load  = yes
 felCatV17e.refseq.mrna.xeno.load = yes
 felCatV17e.refseq.mrna.xeno.loadDesc  = yes
 
     cvs ci -m "adding felCatV17e" etc/genbank.conf
     make etc-update
 
     ssh genbank
     screen  # control this business with a screen since it takes a while
     cd /cluster/data/genbank
     time nice -n +19 bin/gbAlignStep -initial  felCatV17e &
     #   tail var/build/logs/2010.03.17-10:43:18.felCatV17e.initalign.log
     #      gbAlignInstall: complete: real=3.63
     #      genbank 2010.03.17-16:33:21 felCatV17e.initalign: finish
 
     ssh hgwdev
     cd /cluster/data/genbank
     time ./bin/gbDbLoadStep -drop -initialLoad felCatV17e &
     #   logFile: 
     #     tail var/dbload/hgwdev/logs/2010.03.18-15:03:58.dbload.log
     #  real    31m56.710s
 
     # enable daily alignment and update of hgwdev
     cd ~/kent/src/hg/makeDb/genbank
     cvsup
     # add felCatV17e to:
         etc/align.dbs
         etc/hgwdev.dbs
     cvs ci -m "Adding felCatV17e - Cat - Felis catus" \
         etc/align.dbs etc/hgwdev.dbs
     make etc-update
     #   DONE  - 2010-03-17 - Chin
 
 
 ############################################################################
 # running cpgIsland business (DONE - 2010-03-17 - Chin)
     mkdir /hive/data/genomes/felCatV17e/bed/cpgIsland
     cd /hive/data/genomes/felCatV17e/bed/cpgIsland
     cvs -d /projects/compbio/cvsroot checkout -P hg3rdParty/cpgIslands
     cd hg3rdParty/cpgIslands
     #   needed to fixup this source, adding include to readseq.c:
 #include "string.h"
     #   and to cpg_lh.c:
 #include "unistd.h"
 #include "stdlib.h"
     # and fixing a declaration in cpg_lh.c
     sed -e "s#\(extern char\* malloc\)#// \1#" cpg_lh.c > tmp.c
     mv tmp.c cpg_lh.c
     make
     cd ../../
     ln -s hg3rdParty/cpgIslands/cpglh.exe
     mkdir -p hardMaskedFa
     cut -f1 ../../chrom.sizes | while read C
 do
     echo ${C}
     twoBitToFa ../../felCatV17e.2bit:$C stdout \
         | maskOutFa stdin hard hardMaskedFa/${C}.fa
 done
 
     ssh swarm
     cd /hive/data/genomes/felCatV17e/bed/cpgIsland
     mkdir results
     cut -f1 ../../chrom.sizes > chr.list
     cat << '_EOF_' > template
 #LOOP
 ./runOne $(path1) {check out exists results/$(path1).cpg}
 #ENDLOOP
 '_EOF_'
     # << happy emacs
 
     #   the faCount business is to make sure there is enough sequence to
     #   work with in the fasta.  cpglh.exe does not like files with too many
     #   N's - it gets stuck
     cat << '_EOF_' > runOne
 #!/bin/csh -fe
 set C = `faCount hardMaskedFa/$1.fa | grep ^chr | awk '{print  $2 - $7 }'`
 if ( $C > 200 ) then
     ./cpglh.exe hardMaskedFa/$1.fa > /scratch/tmp/$1.$$
     mv /scratch/tmp/$1.$$ $2
 else
     touch $2
 endif
 '_EOF_'
     # << happy emacs
 
 
     gensub2 chr.list single template jobList
     para create jobList
     para try
     para check ... etc
     para time
 
 # in jobList:
 # .....
 # ./runOne chrD3 {check out exists results/chrD3.cpg}
 # wc -l jobList 
 # .....
 # 104054 jobList
 # Completed: 104054 of 104054 jobs
 # CPU time in finished jobs:        214s       3.57m     0.06h    0.00d  0.000 y
 # IO & Wait Time:                935922s   15598.69m   259.98h   10.83d  0.030 y
 # Average job time:                   9s       0.15m     0.00h    0.00d
 # Longest finished job:             174s       2.90m     0.05h    0.00d
 # Submission to last job:          1136s      18.93m     0.32h    0.01d
 
     # Transform cpglh output to bed +
     catDir results | awk '{
 $2 = $2 - 1;
 width = $3 - $2;
 printf("%s\t%d\t%s\t%s %s\t%s\t%s\t%0.0f\t%0.1f\t%s\t%s\n",
        $1, $2, $3, $5,$6, width,
        $6, width*$7*0.01, 100.0*2*$6/width, $7, $9);
 }' > cpgIsland.bed
 
     cd /hive/data/genomes/felCatV17e/bed/cpgIsland
     hgLoadBed felCatV17e cpgIslandExt -tab \
       -sqlTable=$HOME/kent/src/hg/lib/cpgIslandExt.sql cpgIsland.bed
     # Reading cpgIsland.bed
     # Loaded 57950 elements of size 10
     # Sorted
     # Creating table definition for cpgIslandExt
     # Saving bed.tab
     # Loading felCatV17e
 
     # check it with featureBits felCatV17e cpgIslandExt
     #  36646772 bases of 1990635005 (1.841%) in intersection
 
 
     #   cleanup
     rm -fr hardMaskedFa
 
 #######################################################################
 # felCatV17e Cat BLASTZ/CHAIN/NET (DONE  - 2010-03-22 - Chin)
     screen # use a screen to manage this multi-day job
     mkdir /hive/data/genomes/canFam2/bed/lastzFelCatV17e.2010-03-22
     cd /hive/data/genomes/canFam2/bed/lastzFelCatV17e.2010-03-22
 
     cat << '_EOF_' > DEF
 # dog vs. cat
 # maximum M allowed with lastz is only 254
 BLASTZ_M=254
 
 # TARGET: Dog canFan2
 SEQ1_DIR=/scratch/data/canFam2/canFam2.2bit
 SEQ1_LEN=/scratch/data/canFam2/chrom.sizes
 SEQ1_CHUNK=20000000
 SEQ1_LAP=10000
 SEQ1_LIMIT=5
 
 # QUERY: Cat (felCatV17e)
 SEQ2_DIR=/scratch/data/felCatV17e/felCatV17e.2bit
 SEQ2_LEN=/scratch/data/felCatV17e/chrom.sizes
 SEQ2_LIMIT=50
 SEQ2_CHUNK=10000000
 SEQ2_LAP=0
 
 BASE=/hive/data/genomes/canFam2/bed/lastzFelCatV17e.2010-03-22
 TMPDIR=/scratch/tmp
 '_EOF_'
     # << this line keeps emacs coloring happy
 
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
         `pwd`/DEF \
         -syntenicNet -noDbNameCheck \
         -chainMinScore=3000 -chainLinearGap=medium \
         -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
         > do.log 2>&1 &
     #   real    287m24.258s
     cat fb.canFam2.chainFelCatV17eLink.txt
     #  1481040604 bases of 2384996543 (62.098%) in intersection
 
     mkdir /hive/data/genomes/felCatV17e/bed/blastz.canFam2.swap
     cd /hive/data/genomes/felCatV17e/bed/blastz.canFam2.swap
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
         /hive/data/genomes/canFam2/bed/lastzFelCatV17e.2010-03-22/DEF \
         -swap -syntenicNet -noDbNameCheck \
         -workhorse=hgwdev -smallClusterHub=memk -bigClusterHub=swarm \
         -chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
     #   real    251m14.341s
     # *** All done !  Elapsed time: 251m14s
     # *** Make sure that goldenPath/felCatV17e/vsCanFam2/README.txt is
     # accurate.
     # *** Add {chain,net}CanFam2 tracks to trackDb.ra if necessary.
     cat fb.felCatV17e.chainCanFam2Link.txt
     # 1467506008 bases of 1990635005 (73.720%) in intersection
 
 
 
 #####################################################################
 # LASTZ Mouse Swap (DONE - 2010-03-22 - Chin)
     cd /hive/data/genomes/mm9/bed/lastzFelCatV17e.2010-03-22
     cat fb.mm9.chainFelCatV17eLink.txt
     #   637007193 bases of 2620346127 (24.310%) in intersection
     # swap
     mkdir /hive/data/genomes/felCatV17e/bed/blastz.mm9.swap
     cd /hive/data/genomes/felCatV17e/bed/blastz.mm9.swap
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
         /hive/data/genomes/mm9/bed/lastzFelCatV17e.2010-03-22/DEF \
         -swap -syntenicNet -noDbNameCheck \
         -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
         -chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
     #   real   125m37.926s 
     cat fb.felCatV17e.chainMm9Link.txt
     #   616529959 bases of 1990635005 (30.972%) in intersection
 
 
 #####################################################################
 # LASTZ Human Swap (DONE - 2010-03-22 - Chin)
     cd /hive/data/genomes/hg19/bed/lastzFelCatV17e.2010-03-22
     cat fb.hg19.chainFelCatV17eLink.txt
     #  1266003011 bases of 2897316137 (43.696%) in intersection
 
     # Swap re-run 04-07
     mkdir /hive/data/genomes/felCatV17e/bed/blastz.hg19.swap
     cd /hive/data/genomes/felCatV17e/bed/blastz.hg19.swap
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
         /hive/data/genomes/hg19/bed/lastzFelCatV17e.2010-03-22/DEF \
         -swap -syntenicNet -noDbNameCheck \
         -workhorse=hgwdev -smallClusterHub=pk -bigClusterHub=swarm \
         -chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
     #   real   200m43.245s 
     cat fb.felCatV17e.chainHg19Link.txt
 #   1211702270 bases of 1990635005 (60.870%) in intersection
 
 
 #####################################################################
 # LASTZ Panda Swap (DONE - 2010-03-22 - Chin)
     cd /hive/data/genomes/ailMel1/bed/lastzFelCatV17e.2010-03-22
     cat fb.felCatV17e.chainAilMel1Link.txt
     # 1503647735 bases of 1990635005 (75.536%) in intersection
 
+    #   re-run swap 04-26
     mkdir /hive/data/genomes/felCatV17e/bed/blastz.ailMel1.swap
     cd /hive/data/genomes/felCatV17e/bed/blastz.ailMel1.swap
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
         /hive/data/genomes/ailMel1/bed/lastzFelCatV17e.2010-03-22/DEF \
         -swap -syntenicNet -noDbNameCheck \
         -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
         -chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
     # real    155m57.566s
+    cd /hive/data/genomes/ailMel1/bed/blastz.felCatV17e.swap
     cat fb.ailMel1.chainFelCatV17eLink.txt
     # 1507273252 bases of 2245312831 (67.130%) in intersection
 
-
-
+    #   Due to the SEQ1 and SEQ2 spec'ed in
+    # /hive/data/genomes/ailMel1/bed/lastzFelCatV17e.2010-03-22/DEF
+    # need to copy the swap directory from ailMel1 to felCatv17e
+    #  also need to rename from ailMel1.felCatV17e, or 6way step will have
+    #  problem
+    #  Ping hiram for this 
 
 #####################################################################
 # LASTZ Oppsom  Swap (DONE - 2010-03-22 - Chin)
     cd /hive/data/genomes/monDom5/bed/lastzFelCatV17e.2010-03-22
     cat fb.monDom5.chainFelCatV17eLink.txt
     # 178616721 bases of 3501660299 (5.101%) in intersection
 
     # Swap
     mkdir /hive/data/genomes/felCatV17e/bed/blastz.monDom5.swap
     cd /hive/data/genomes/felCatV17e/bed/blastz.monDom5.swap
     time nice -n +19 doBlastzChainNet.pl -verbose=2 \
         /hive/data/genomes/monDom5/bed/lastzFelCatV17e.2010-03-22/DEF \
         -swap -syntenicNet -noDbNameCheck \
         -workhorse=hgwdev -smallClusterHub=encodek -bigClusterHub=swarm \
         -chainMinScore=5000 -chainLinearGap=loose > swap.log 2>&1 &
     # *** All done !  Elapsed time: 100m29s
     # *** Make sure that goldenPath/felCatV17e/vsMonDom5/README.txt is
     # accurate.
     # *** Add {chain,net}MonDom5 tracks to trackDb.ra if necessary.
 
     cat fb.felCatV17e.chainMonDom5Link.txt
     #   166499264 bases of 1990635005 (8.364%) in intersection
 
 
 ############################################################################
 # ctgPos2 track - showing clone sequence locations on chromosomes
 #       (DONE - 2010-03-30 - Chin)
     mkdir /hive/data/genomes/felCatV17e/bed/ctgPos2
     cd /hive/data/genomes/felCatV17e/bed/ctgPos2
     cat << '_EOF_' > agpToCtgPos2.pl
 #!/usr/bin/env perl
 
 use warnings;
 use strict;
 
 my $argc = scalar(@ARGV);
 
 if ($argc != 1) {
     printf STDERR "usage: zcat your.files.agp.gz | agpToCtgPos2.pl
 /dev/stdin > ctgPos2.tab\n";
     exit 255;
 }
 
 my $agpFile = shift;
 
 open (FH, "<$agpFile") or die "can not read $agpFile";
 while (my $line = <FH>) {
     next if ($line =~ m/^#/);
     chomp $line;
     my @a = split('\s+', $line);
     next if ($a[4] =~ m/^N$/);
     next if ($a[4] =~ m/^U$/);
     my $chrSize = $a[2]-$a[1]+1;
     my $ctgSize = $a[7]-$a[6]+1;
     die "sizes differ $chrSize != $ctgSize\n$line\n" if ($chrSize !=
 $ctgSize);
     printf "%s\t%d\t%s\t%d\t%d\t%s\n", $a[5], $chrSize, $a[0], $a[1]-1,
 $a[2], $a[4];
 }
 close (FH);
 '_EOF_'
     # << happy emacs
 
 chmod 775 agpToCtgPos2.pl 
 
 export S=../../genbank/Primary_Assembly/assembled_chromosomes
 cut -f2 ${S}/chr2acc |  while read ACC
 do
     C=`grep "${ACC}" ${S}/chr2acc | cut -f1`
     zcat ${S}/AGP/chr${C}.agp.gz \
         | sed -e "s/^${ACC}/chr${C}/"
 done | ./agpToCtgPos2.pl /dev/stdin > ctgPos2.tab
 
     hgLoadSqlTab felCatV17e ctgPos2 $HOME/kent/src/hg/lib/ctgPos2.sql \ 
      ctgPos2.tab
 
     # add the track ctgPos2 to src/hg/makeDb/trackDb/cat/felCatV17e/trackDb.ra
     # at src/makeDb/trackdb do "make update" or/and "make alpha"
 
 #########################################################################
 # all.joiner update, downloads and in pushQ - (DONE - 2010-04-09 Chin)
     cd $HOME/kent/src/hg/makeDb/schema
     # fixup all.joiner until this is a clean output
     joinerCheck -database=felCatV17e -all all.joiner
     
     mkdir /hive/data/genomes/felCatV17e/goldenPath
     cd /hive/data/genomes/felCatV17e/goldenPath
     makeDownloads.pl felCatV17e > do.log 2>&1
     #  vi /hive/data/genomes/felCatV17e/goldenPath/database/README.txt
     #  vi  /hive/data/genomes/felCatV17e/goldenPath/bigZips/README.txt
 
 
     #   now ready for pushQ entry
     mkdir /hive/data/genomes/felCatV17e/pushQ
     cd /hive/data/genomes/felCatV17e/pushQ
     makePushQSql.pl felCatV17e > felCatV17e.pushQ.sql 2> stderr.out
     #   check for errors in stderr.out, some are OK, e.g.:
 # WARNING: felCatV17e does not have seq
 # WARNING: felCatV17e does not have extFile
 # WARNING: felCatV17e does not have tableDescriptions
 
     #   copy it to hgwbeta
     #   use hgcat in .hg.conf.beta
     scp -p felCatV17e.pushQ.sql hgwbeta:/tmp
     ssh hgwbeta
     cd /tmp
     hgsql qapushq < felCatV17e.pushQ.sql
     #   in that pushQ entry walk through each entry and see if the
     #   sizes will set properly
 
     #   reset to hguser in .hg.conf.beta
 
 
+#####################################################################
+## 6-Way Multiz (working - 2010-04-19 - Chin)
+# use /cluster/home/chinhli/kent/src/hg/utils/phyloTrees/49way.nh 
+    mkdir /hive/data/genomes/felCatV17e/bed/multiz6way
+    cd /hive/data/genomes/felCatV17e/bed/multiz6way
+
+    /cluster/bin/phast/tree_doctor \
+      --prune-all-but=felCat3,hg19,mm9,canFam2,monDom5,ailMel1 \
+      --rename="felCat3 -> felCatV17e  " \
+/cluster/home/chinhli/kent/src/hg/utils/phyloTrees/49way.nh > 6way.nh 
+
+    #   rearrange felCatV17e to the top, get some help from tree_doctor:
+    /cluster/bin/phast/tree_doctor --name-ancestors --reroot felCatV17e \
+           --with-branch 6way.nh
+    #   edit out the ancestors, and move felCatV17e from the bottom to
+    #   the top, resulting in this tree:
+
+(((felCatV17e:0.098612,(canFam2:0.052458,ailMel1:0.050000):0.050000):0.093470,(hg19:0.144018,mm9:0.356483):0.020593):0.258392,monDom5:0.340786);
+
+    #   more rearranging after seeing what the distance table looks like
+    #   below to get them appearing as much as possible in their
+    #   distance order top to bottom: (TO_DO)
+
+    # use the phyloGif to get the tree image
+   /cluster/bin/x86_64/phyloGif -phyloGif_tree=6way.nh \
+   -phyloGif_width=260  -phyloGif_height=260 > 6way.gif
+
+    #   more rearranging after seeing what the distance table looks like
+    #   below to get them appearing as much as possible in their
+    #   distance order top to bottom: (TO_DO)
+
+
+
+    #   usee this specification in the phyloGif tool after changing the names:
+    /cluster/bin/phast/tree_doctor \
+--rename="felCatV17e -> Cat ;  hg19 -> Human ; ailMel1 -> Panda ; canFam2 -> Dog ; mm9 -> Mouse ; monDom5 -> Opossum " 6way.nh
+
+    #   http://genome.ucsc.edu/cgi-bin/phyloGif
+    #   to obtain a gif image for htdocs/images/phylo/felCatV17e_6way.gif
+   /cluster/bin/x86_64/phyloGif -phyloGif_tree=6way.nh \
+   -phyloGif_width=260  -phyloGif_height=260 > felCatV17e_6way.gif
+
+    /cluster/bin/phast/all_dists 6way.nh > 6way.distances.txt
+    # make sure all symlinks lastz.DB -> lastzDb-date
+    #   exist here and at the swap locations, the perl script expects this
+    #   in order to find featureBits numbers.
+    cd /hive/data/genomes/felCatV17e/bed]
+    ln -s blastz.hg19.swap lastz.hg19
+    ln -s blastz.canFam2.swap lastz.canFam2
+    ln -s blastz.mm9.swap lastz.mm9
+    ln -s blastz.monDom5.swap lastz.monDom5
+    # ln -s blastz.ailMel1.swap lastz.ailMel1 (we switched SEQ1 and SEQ2)
+    ln -s /hive/data/genomes/ailMel1/bed/blastz.felCatV17e.swap lastz.ailMel1
+
+
+    cd /hive/data/genomes/felCatV17e/bed/multiz6way
+    #   Use 6way.distances.txt to create the table below
+    #   with this perl script:
+    cat << '_EOF_' > sizeStats.pl
+#!/usr/bin/env perl
+use strict;
+use warnings;
+
+
+open (FH, "grep -y felCatV17e 6way.distances.txt | sort -k3,3n|") or
+        die "can not read 6way.distances.txt";
+
+my $count = 0;
+while (my $line = <FH>) {
+    chomp $line;
+    my ($felCatV17e, $D, $dist) = split('\s+', $line);
+    my $chain = "chain" . ucfirst($D);
+    my $B="/hive/data/genomes/felCatV17e/bed/lastz.$D/fb.felCatV17e." .
+        $chain . "Link.txt";
+    my $chainLinkMeasure =
+        `awk '{print \$5}' ${B} 2> /dev/null | sed -e "s/(//; s/)//"`;
+    chomp $chainLinkMeasure;
+    $chainLinkMeasure = 0.0 if (length($chainLinkMeasure) < 1);
+    $chainLinkMeasure =~ s/\%//;
+    my $swapFile="/hive/data/genomes/${D}/bed/lastz.felCatV17e/fb.${D}.chainCalJac3Link.txt";
+    my $swapMeasure = "N/A";
+    if ( -s $swapFile ) {
+        $swapMeasure =
+            `awk '{print \$5}' ${swapFile} 2> /dev/null | sed -e "s/(//; s/)//"`;
+        chomp $swapMeasure;
+        $swapMeasure = 0.0 if (length($swapMeasure) < 1);
+        $swapMeasure =~ s/\%//;
+    }
+    my $orgName=
+    `hgsql -N -e "select organism from dbDb where name='$D';" hgcentraltest`;
+    chomp $orgName;
+    if (length($orgName) < 1) {
+        $orgName="N/A";
+    }
+    ++$count;
+   ++$count;
+    if ($swapMeasure eq "N/A") {
+        printf "# %02d  %.4f - %s %s\t(%% %.3f) (%s)\n", $count, $dist,
+            $orgName, $D, $chainLinkMeasure, $swapMeasure
+    } else {
+        printf "# %02d  %.4f - %s %s\t(%% %.3f) (%% %.3f)\n", $count, $dist,
+            $orgName, $D, $chainLinkMeasure, $swapMeasure
+    }
+}
+close (FH);
+'_EOF_'
+    # << happy emacs
+    chmod +x ./sizeStats.pl
+    ./sizeStats.pl
+###### 04-27 stop here due to wrong output from sizeStats.pl 
+###### pending move/rename aelMel1/felcatV17e lastz swap results
+# 02  0.1986 - Panda ailMel1    (% 0.000) (N/A)
+# 04  0.2011 - Dog canFam2      (% 73.720) (N/A)
+# 06  0.3567 - Human hg19       (% 60.870) (N/A)
+# 08  0.5692 - Mouse mm9        (% 30.972) (N/A)
+# 10  0.7913 - Opossum monDom5  (% 8.364) (N/A)
+
+
+
 
 ####################################################################