src/hg/makeDb/doc/canFam2.txt 1.20
1.20 2009/07/02 22:21:44 hiram
Rerun all the equCab2 chains and nets to clean up problems
Index: src/hg/makeDb/doc/canFam2.txt
===================================================================
RCS file: /projects/compbio/cvsroot/kent/src/hg/makeDb/doc/canFam2.txt,v
retrieving revision 1.19
retrieving revision 1.20
diff -b -B -U 4 -r1.19 -r1.20
--- src/hg/makeDb/doc/canFam2.txt 1 Jul 2008 16:52:59 -0000 1.19
+++ src/hg/makeDb/doc/canFam2.txt 2 Jul 2009 22:21:44 -0000 1.20
@@ -2776,5 +2776,128 @@
by a single Makefile. This is available from:
svn+ssh://hgwdev.cse.ucsc.edu/projects/compbio/usr/markd/svn/projs/transMap/tags/vertebrate.2008-06-30
see doc/builds.txt for specific details.
+
+############################################################################
+# lastz Poodle canFamPoodle1 (DONE - 2009-06-06,22 - Hiram)
+ mkdir /hive/data/genomes/canFam2/bed/lastzCanFamPoodle1.2009-06-06
+ cd /hive/data/genomes/canFam2/bed/lastzCanFamPoodle1.2009-06-06
+
+ cat << '_EOF_' > DEF
+# Tasha boxer dog vs Shadow poodle
+# parameters for very close alignment from human-chimp example
+BLASTZ_M=254
+BLASTZ_Q=/scratch/data/blastz/human_chimp.v2.q
+BLASTZ_O=600
+BLASTZ_E=150
+BLASTZ_K=4500
+BLASTZ_Y=15000
+BLASTZ_T=2
+
+# TARGET: Tasha, canFam2
+SEQ1_DIR=/scratch/data/canFam2/nib
+SEQ1_LEN=/scratch/data/canFam2/chrom.sizes
+SEQ1_CHUNK=10000000
+SEQ1_LAP=10000
+
+# QUERY: Dog CanFam2
+SEQ2_DIR=/scratch/data/canFamPoodle1/canFamPoodle1.2bit
+SEQ2_LEN=/scratch/data/canFamPoodle1/chrom.sizes
+SEQ2_CHUNK=40000000
+SEQ2_LAP=0
+SEQ2_LIMIT=800
+
+BASE=/hive/data/genomes/canFam2/bed/lastzCanFamPoodle1.2009-06-06
+TMPDIR=/scratch/tmp
+'_EOF_'
+ # << happy emacs
+
+ time nice -n +19 doBlastzChainNet.pl \
+ -verbose=2 \
+ `pwd`/DEF \
+ -noDbNameCheck -noLoadChainSplit \
+ -workhorse=hgwdev -smallClusterHub=memk -bigClusterHub=pk \
+ -chainMinScore=5000 -chainLinearGap=medium > do.log 2>&1
+ # real 8250m12.188s
+# Completed: 374825 of 374825 jobs
+# CPU time in finished jobs: 187350504s 3122508.40m 52041.81h 2168.41d 5.941 y
+# IO & Wait Time: 4127960s 68799.33m 1146.66h 47.78d 0.131 y
+# Average job time: 511s 8.51m 0.14h 0.01d
+# Longest finished job: 2339s 38.98m 0.65h 0.03d
+# Submission to last job: 494836s 8247.27m 137.45h 5.73d
+
+ # the lastz run thought it failed, but it didn't, continuing:
+ time nice -n +19 doBlastzChainNet.pl \
+ -verbose=2 \
+ `pwd`/DEF \
+ -continue=cat -noDbNameCheck -noLoadChainSplit \
+ -workhorse=hgwdev -smallClusterHub=memk -bigClusterHub=pk \
+ -chainMinScore=5000 -chainLinearGap=medium > cat.log 2>&1
+ # real 4841m48.581s <- this is actually a fb time on cavPor3
+ # this finish step was less than an hour
+ fb.canFam2.chainCanFamPoodle1Link.txt
+ # 1405528799 bases of 2384996543 (58.932%) in intersection
+
+ mkdir /hive/data/genomes/canFamPoodle1/bed/blastz.canFam2.swap
+ cd /hive/data/genomes/canFamPoodle1/bed/blastz.canFam2.swap
+
+ time nice -n +19 doBlastzChainNet.pl \
+ -verbose=2 \
+ /hive/data/genomes/canFam2/bed/lastzCanFamPoodle1.2009-06-06/DEF \
+ -swap -noDbNameCheck -noLoadChainSplit \
+ -workhorse=hgwdev -smallClusterHub=memk -bigClusterHub=pk \
+ -chainMinScore=5000 -chainLinearGap=medium > swap.log 2>&1
+ # real 8585m58.789s
+ cat fb.canFamPoodle1.chainCanFam2Link.txt
+ # 1377478896 bases of 1517497798 (90.773%) in intersection
+
+############################################################################
+# Re-Run equCab2 alignment (DONE - 2009-06-29 - Hiram
+ mkdir /hive/data/genomes/canFam2/bed/lastzEquCab2.2009-06-29
+ cd /hive/data/genomes/canFam2/bed/lastzEquCab2.2009-06-29
+
+ cat << '_EOF_' > DEF
+# Dog vs. Horse
+
+BLASTZ_M=50
+
+# TARGET: Dog canFam2
+SEQ1_DIR=/scratch/data/canFam2/nib
+SEQ1_LEN=/scratch/data/canFam2/chrom.sizes
+SEQ1_CHUNK=10000000
+SEQ1_LAP=10000
+
+# QUERY: Horse
+SEQ2_DIR=/scratch/data/equCab2/equCab2.2bit
+SEQ2_LEN=/scratch/data/equCab2/chrom.sizes
+SEQ2_CTGDIR=/hive/data/genomes/equCab2/equCab2.UnScaffolds.2bit
+SEQ2_CTGLEN=/hive/data/genomes/equCab2/equCab2.UnScaffolds.sizes
+SEQ2_LIFT=/cluster/data/equCab2/jkStuff/equCab2.chrUn.lift
+SEQ2_CHUNK=20000000
+SEQ2_LIMIT=100
+SEQ2_LAP=0
+
+BASE=/hive/data/genomes/canFam2/bed/lastzEquCab2.2009-06-29
+TMPDIR=/scratch/tmp
+'_EOF_'
+ # << happy emacs
+
+ time doBlastzChainNet.pl `pwd`/DEF \
+ -noLoadChainSplit -verbose=2 -bigClusterHub=swarm \
+ -chainMinScore=3000 -chainLinearGap=medium > do.log 2>&1 &
+ # real 338m57.973s
+ cat fb.canFam2.chainEquCab2Link.txt
+ # 1676663178 bases of 2384996543 (70.300%) in intersection
+ # this is identical to what went down before ?
+
+ mkdir /hive/data/genomes/equCab2/bed/blastz.canFam2.swap
+ cd /hive/data/genomes/equCab2/bed/blastz.canFam2.swap
+ time doBlastzChainNet.pl \
+ /hive/data/genomes/canFam2/bed/lastzEquCab2.2009-06-29/DEF \
+ -swap -noLoadChainSplit -verbose=2 -bigClusterHub=swarm \
+ -chainMinScore=3000 -chainLinearGap=medium > swap.log 2>&1 &
+ # real 286m51.658s
+ fb.equCab2.chainCanFam2Link.txt
+ # 1721407500 bases of 2428790173 (70.875%) in intersection
+
############################################################################