Commits for angie
switch to files view, user index
v256_preview to v256_preview2 (2011-08-16 to 2011-08-23) v256
- If environment variable CFLAGS is set, don't override it in common.mk. Now we can compile and link for gprof using CFLAGS=-pg.
- Adding CFLAGS to hgTracks' linking.
- src/hg/hgTracks/makefile - lines changed 2, context: html, text, full: html, text
- Feature #3711 (VCF haplotype sorting display): Performance improvements.1. lib/hacTree.c: instead of doing a zillion incremental memcpy's to drop nodes,
build up a set of nodes to delete and then do the minimal memmove's afterwards.
Use memmove per memcpy's man page, which says not to use memcpy if regions overlap!
2. lib/vcf.c: avoid unnecessary calls to vcfFilePooledStr from inner loops.
Also, if ref is all nucleotides (not symbolic as for SV's), use it to set chromEnd.
It may still be overwritten if the END keyword is used in the info column.
- Feature #3711 (VCF haplotype sorting display): For now, limit the numberof variants on either side of the center variant to 20 instead of 50 --
I happened to land in a small region that had many variants that seemed
to have no linkage whatsoever in 1000 Genomes phase 1 interim VCF, e.g.
chr1:1,225,165-1,225,869 would take 42 seconds to cluster and draw!
Limiting to 20 reduces that ~0.5s (also paired w/performance improvements,
but it's mostly the number of distinct haplotypes that we pass into
hacTreeFromItems).
Also, a rounding overflow (transforming pixels to gtHapOrder index) was
sporadically causing apache-only SEGVs; check that range! :)
- src/hg/hgTracks/vcfTrack.c - lines changed 5, context: html, text, full: html, text
- Sheesh, forgot a .h file for last-minute tweak. At least I haven't pushed yet! :)
- Oops, forgot that one of my changes to hacTree (when merging root intoa node, preserve the order of its kids -- was inadvertently swapping
left kid to right when merging right kid) would require a change to
the expected output file. It's just an internal reordering in the
tree, doesn't change levels.
- src/lib/tests/expected/hacTreeTest.out - lines changed 55, context: html, text, full: html, text
- Fix for compiling without USE_BAM: bamFile.h needs to define bam_index_t.
- Feature #3707 (VCF+tabix support in hgTables):Basic hgTables support for VCF, modeled after BAM code. Tested all fields,
selected fields, bed output; filter, intersection, schema
on 3 flavors of vcfTabix:
ftp://ftp-trace.ncbi.nih.gov/1000genomes/ftp/release/20101123/interim_phase1_release/ALL.chr17.phase1.projectConsensus.genotypes.vcf.gz
ftp://ftp-trace.ncbi.nlm.nih.gov/1000genomes/ftp/release/20100804/AFR.dindel.20100804.sites.vcf.gz
ftp://ftp.ncbi.nih.gov/snp/organisms/human_9606/VCF/v4.0/ByChromosomeNoGeno/00-All.vcf.gz
- src/hg/hgTables/asFilter.c - lines changed 2, context: html, text, full: html, text
- src/hg/hgTables/asObj.c - lines changed 5, context: html, text, full: html, text
- src/hg/hgTables/bedList.c - lines changed 2, context: html, text, full: html, text
- src/hg/hgTables/filterFields.c - lines changed 25, context: html, text, full: html, text
- src/hg/hgTables/hgTables.c - lines changed 15, context: html, text, full: html, text
- src/hg/hgTables/hgTables.h - lines changed 34, context: html, text, full: html, text
- src/hg/hgTables/identifiers.c - lines changed 7, context: html, text, full: html, text
- src/hg/hgTables/joining.c - lines changed 12, context: html, text, full: html, text
- src/hg/hgTables/mainPage.c - lines changed 5, context: html, text, full: html, text
- src/hg/hgTables/makefile - lines changed 1, context: html, text, full: html, text
- src/hg/hgTables/schema.c - lines changed 4, context: html, text, full: html, text
- src/hg/hgTables/vcf.c - lines changed 492, context: html, text, full: html, text
- src/hg/lib/customFactory.c - lines changed 1, context: html, text, full: html, text
- Fixing compile-breaker: boolean TRUE and FALSE need to be uppercased.
switch to files view, user index