Commits for angie
switch to files view, user index
v361_preview2 to v361_base (2018-02-05 to 2018-02-12) v361
Show details
775c4ba0ea17856a61794426a32c2b7206e41df1 Mon Feb 5 10:48:52 2018 -0800
- Fix a few cases of assuming that some details were calculated when they weren't.
- src/hg/lib/variantProjector.c - lines changed 16, context: html, text, full: html, text
839d2f6fce40d1db8711359110dd8d83447ce70f Mon Feb 5 15:17:48 2018 -0800
- The identifier genbankAccession was not based on GenBank but rather on seq+extFile which is a catch-all and should not be related to genbank tables in general. Relating $gbd.seq to all GenBank-related tables was causing unnecessary slowness in joinerFindRoute. Rename the identifier to seqExtFile and use only where applicable.
- src/hg/makeDb/schema/all.joiner - lines changed 8, context: html, text, full: html, text
3076f02c9ea73d93de1461992f6f7d6d51c35546 Tue Feb 6 09:00:51 2018 -0800
- Limit the use of splitPrefix to the old databases that actually have split tables. We stopped making those many years ago, and splitPrefix causes a sql wildcard search when a table isn't found. This change and the previous change reduce the number of SQL queries by over 5000 when joining hg19.refSeqAli and several hgFixed genbank tables. hg19 has only one set of split tables, chr*_gl, so it is omitted from oldSplit but has special rules for gl.
- src/hg/makeDb/schema/all.joiner - lines changed 48, context: html, text, full: html, text
a0d8cb8a727257266fe5563d3e86fd502329a488 Tue Feb 6 09:54:01 2018 -0800
- Adding joinerRoute to makeDb/schema/makefile for top level build.
- src/hg/makeDb/schema/makefile - lines changed 1, context: html, text, full: html, text
d8be7e6e88c0712589dac72c1f023c84028a0d1c Wed Feb 7 11:49:17 2018 -0800
- To prevent joinerRoute from searching over all databases for routes from some genome db to hgFixed metadata tables, add a new optional arg to joinerRelate to apply joinerExclusiveCheck to the genome db in addition to the current table's database.
This reduces the time to find the route hg19.refSeqAli --> hgFixed.gbCdnaInfo --> hgFixed.cds from ~1.5s to ~0.1s. It also keeps the related-table selection in hgTables and hgIntegrator from showing >5000 related tables when starting with a genome db table and adding hgFixed.gbCdnaInfo.
refs #20949
- src/hg/hgGenome/import.c - lines changed 1, context: html, text, full: html, text
- src/hg/hgIntegrator/hgIntegrator.c - lines changed 1, context: html, text, full: html, text
- src/hg/hgTables/filterFields.c - lines changed 1, context: html, text, full: html, text
- src/hg/hgTables/hgTables.c - lines changed 1, context: html, text, full: html, text
- src/hg/hgTables/schema.c - lines changed 1, context: html, text, full: html, text
- src/hg/lib/cartTrackDb.c - lines changed 1, context: html, text, full: html, text
- src/hg/makeDb/schema/joinableFields/joinableFields.c - lines changed 1, context: html, text, full: html, text
3f06ffc10df40e72dfa18a150eeca23fda6b41df Thu Feb 8 13:42:04 2018 -0800
- annoStreamDbPslPlus internally used a json config to join in CDS and sequence with the underlying PSL -- so annoStreamDbPslPlusNew did not accept an incoming config, so hgVai's RefSeq Status option was ignored. Add jsonObjectMerge so that the config from hgVai can be merged with the internal config, and make sure that the extra columns from the internal annoStreamDb are added to annoStreamDbPslPlusNew's asObj. We can't just use the internal annoStreamDb's asObj because the field names for CDS etc. fields would differ for refSeqAli vs ncbiRefSeq and column-matching to recognize annoStreamDbPslPlus would fail.
refs #20948
- src/hg/inc/annoStreamDbPslPlus.h - lines changed 6, context: html, text, full: html, text
- src/hg/lib/annoFormatVep.c - lines changed 1, context: html, text, full: html, text
- src/hg/lib/annoStreamDbPslPlus.c - lines changed 70, context: html, text, full: html, text
7fa4a82cd43f5a4dae29815281e75d7874772176 Thu Feb 8 14:11:28 2018 -0800
- Thanks ChrisL for noticing that single-base deletion was written as a range when it should be just the pos. refs #20948
- src/hg/lib/annoFormatVep.c - lines changed 1, context: html, text, full: html, text
0a2b436d5f082becf730cea30457eb86ac1e1e3c Fri Feb 9 11:55:10 2018 -0800
- Added new column to test set ncbiRefSeqLink.tab.gz, corresponding to ncbiRefSeqLink.{as,sql} new column externalId, to avoid warning when loading the test set. refs #20215
- src/hg/utils/vcfToHgvs/tests/PGTdb/ncbiRefSeqLink.tab.gz - lines changed 0, context: html, text, full: html, text (a binary file or whitespace-only change shows no diff)
ad3a176457ad1d21a0fedc47f349ec3484e751f1 Fri Feb 9 16:27:51 2018 -0800
- Cleaning up some old ugliness about the size parameter to isAllDna and isAllNt.
hgc's printSnpAlignment code that parsed snpNNN.fa was using lineSize as length
but lineSize is length+1. Then isAllDna was written with "i<size-1" as the
loop test instead of "i < size". I didn't fix that properly when I separated
out isAllNt from isAllDna.
Later, I (re?)discovered that isAllNt needed length+1 as its size and just
added some FIXME comments. Thanks Brian R for prodding me to actually fix it.
refs #20895
- src/hg/lib/variantProjector.c - lines changed 2, context: html, text, full: html, text
- src/hg/utils/vcfToHgvs/vcfToHgvs.c - lines changed 2, context: html, text, full: html, text
a8e96a57b3485f554666c687262fcd295c0e7a6d Mon Feb 12 09:27:22 2018 -0800
- Forgot to check for NULL.
- src/hg/lib/annoStreamDbPslPlus.c - lines changed 1, context: html, text, full: html, text
switch to files view, user index