2b2a4e0795
Merge branch 'master' of ssh://gsa1/humgen/gsa-scr1/gsa-engineering/git/stable
2011-08-17 16:26:45 -04:00
d170187896
Disable optimization that increases marginal speed of the GATK slightly but
...
can produce data loss in a narrow corner case where the BGZF block(s) locations
and offsets in the last index bucket of contig n overlap exactly with the BGZF
block locations and offset in the last index bucket of contig n+1.
A proper fix that keeps the optimization has already been introduced into
unstable, but disabling the optimization is a low risk way to make sure that
users of stable experience no data loss.
2011-08-17 16:16:05 -04:00
9d4add3268
Merge branch 'master' of ssh://nickel.broadinstitute.org/humgen/gsa-scr1/gsa-engineering/git/stable
2011-08-16 14:18:03 -04:00
170d1ff7b6
Fix in UG for trying to call indels at IUPAC code bases when in EMIT_ALL_SITES mode
2011-08-16 14:17:46 -04:00
9f3328db53
fixing read group name collision: before writing the read into respective stream in nway-out mode we now retrieve the original rg, not the merged/modified one
2011-08-16 13:45:40 -04:00
9ddbfdcb9f
Check filtered status before applying to alt reference
2011-08-15 12:25:23 -04:00
8325cb8c26
Fixing up apparent source control/merge snafu: fix to correctly output PL ordering in multi-allelic sites by UG was only half-committed and hence not working. This completes fix
2011-08-10 15:31:49 -04:00
98a96f07c1
Updated standard deviation parameter in VQSR to our current recommended value
2011-08-04 14:06:26 -04:00
c0d4110ffd
Correcting redundant warning text.
2011-07-29 10:01:11 -04:00
1afc49a297
There are some really 'interesting' (but apparently valid) records in the Mus musculus dbSNP file. Generalized the handling of complex cases in the dbSNP adaptor to handle it all. I just grabbed the actual Mus musculus dbSNP file as a test, ran it whole genome, and confirmed that we finally produce a valid VCF on it. Should be the last commit needed on this adaptor.
2011-07-28 13:55:58 -04:00
6230315ff2
Along with my half-written commit message from earlier, I also forgot to commit the integration test updates. This is what happens when you try to do things 30 seconds before you leave for the day. To finish up from before: complex events weren't being padded with the reference base as per the VCF spec. They are now.
2011-07-27 22:51:21 -04:00
64aad67b5f
Fixing dbSNP adaptor for complex indels (wasn)
2011-07-27 16:13:45 -04:00
f50145b872
Reinitialize random seed in the bwa bindings from the fixed seed stored in the
...
BWA support files every time the support files are loaded.
2011-07-22 13:41:53 -04:00
8b8f121cfb
Merge branch 'master' of ssh://gsa3.broadinstitute.org/humgen/gsa-scr1/gsa-engineering/git/unstable
2011-07-21 23:01:11 -04:00
59eb1f4663
Memory limits changed from Int to Double.
...
Updated LSF calls to read memory units from config along with tweaks to select hosts.
Moved some common code from GridEngine and LSF to super classes.
2011-07-21 22:57:18 -04:00
2f5d10d16b
Fix bug wherein aligner could be closed prior to its being used to lowercase sequences.
2011-07-21 13:21:48 -04:00
7054c5342f
When using the BWA bindings, you have to explicitly call close() to get the
...
bindings to release memory.
It may or may not be possible to implicitly close triggered by the GC; I'll add a JIRA.
2011-07-21 12:13:29 -04:00
15610ce0c3
Per Matt's request, disabling BWA-based integration tests so he can assess bamboo memory usage.
2011-07-21 11:04:22 -04:00
0a1d2df8cb
Merged bug fix from Stable into Unstable
2011-07-20 13:19:35 -04:00
f15023b7d2
Bad bug fix: output GLs in multiallelic records were in incorred order (misread spec)
2011-07-20 12:10:48 -04:00
b9c9e0e952
Merge branch 'master' of ssh://nickel.broadinstitute.org/humgen/gsa-scr1/gsa-engineering/git/unstable
2011-07-20 10:45:16 -04:00
7140280bf6
Further bug fixes/cleanups for PrintReadsWalker
2011-07-20 10:44:37 -04:00
a2d90a3590
Bug fix: reverted logic so that default behavior skips over sample lookup
2011-07-20 10:23:10 -04:00
e8409c80fa
Further protection vs null pointers in PrintReadsWalker
2011-07-19 21:59:24 -04:00
5d706c9e92
Merge branch 'master' of ssh://chartl@tin.broadinstitute.org/humgen/gsa-scr1/gsa-engineering/git/unstable
...
Removing PSP and CSM
Conflicts:
public/java/src/org/broadinstitute/sting/gatk/walkers/sequenom/CreateSequenomMask.java
public/java/src/org/broadinstitute/sting/gatk/walkers/sequenom/PickSequenomProbes.java
2011-07-19 20:25:33 -04:00
fb2d475c22
Bug fix to prevent null pointer
2011-07-19 20:13:56 -04:00
92c7cfa1c8
BWA bindings and tests moved to public (was required for ValidationAmplicons)
...
Integration tests for ValidationAmplicons. New argument to disable BWA, lowercase letters only for repetitiveness instead.
2011-07-19 20:11:31 -04:00
baae381acb
Revert "Merge branch 'master' of ssh://gsa1/humgen/gsa-scr1/gsa-engineering/git/unstable"
...
This reverts commit 039a6bb01f345322ce2be50ae3634308bb24e77e, reversing
changes made to b9c9973d1c638dfc9f8c19b5eb845e99844f9d29.
2011-07-19 18:38:53 -04:00
07e716d23a
PickSequenomProbes2 expanded functionality: lowercasing based on sequence uniqueness, preserving reference base prior to indel (not a part of the VC as I thought it was), masking deletion bases with 'N's, flanking insertion with 'N's, output is a fasta formatted file. Renamed to ValidationAmplicons since this is really not for picking sequenom probes, but for generating amplicon sequence from which other applications (like sequenom) can choose PCR primers. Moved from private to public.
2011-07-19 15:21:47 -04:00
6181d1e4cb
Fixed integration test for VariantsToTable: now the * in REF column is not output
2011-07-19 14:42:11 -04:00
e6d306458c
Merge bug fixes
2011-07-19 14:36:20 -04:00
989dd17f95
a) Add ability in PrintReads to specify a sample file to easily subset samples, useful for IGV visualization, b) VariantsToTable is more R-friendly with Indels when printing ref/alt columns, c) Changes to SelectVariants ability to speficy a mask to randomly sample from a given AF distribution
2011-07-19 14:29:07 -04:00
8f0badc52b
Updating md5s, as the diffobjects walker now emits the summary in reverse order.
2011-07-18 15:44:21 -04:00
c05451047c
Support for multiple records at the same site. The first record gets chr:start, and subsequent records get chr:start_2, chr:start_3, etc.
2011-07-18 15:43:52 -04:00
782a05e9b5
Support for sorting the diff output in reverse order.
2011-07-18 15:43:01 -04:00
45702d3084
Now supports a mode where the primary key isn't sorted. In this case the records are displayed in the order in which they are added to to the table.
2011-07-18 15:40:15 -04:00
83ba2c066a
Making it deterministic
2011-07-18 13:59:02 -04:00
92fa410450
Check that it's a valid bam file before parsing or bad things can happen
2011-07-18 13:43:34 -04:00
80b5c5261a
CombineVariants no longer combines records of different types. So now when combining SNP and indel callsets, overlapping calls get their own records. Useful for Khalid in the pipeline. For those interested, it turns out the previous behavior was doing the wrong thing occasionally (and this was even captured in the integration tests).
2011-07-18 13:42:45 -04:00
bc8b5da698
Added docs while I was reading through the code to understand it
2011-07-18 12:25:54 -04:00
51b0dd01c3
Merge branch 'master' of ssh://gsa1/humgen/gsa-scr1/gsa-engineering/git/unstable
2011-07-18 10:47:29 -04:00
d6e2e89f99
Walker test system refactoring. All MD5DB related functions are now in MD5DB.java.
...
System has the concept of a local and a global MD5 db. The local one is like it operated previously. The global one lives in /humgen/gsa-hpprojects/GATK/data/integrationtests. If the system can find this directory then MD5s will also be read / written to this location. This means that gsabamboo will print differences as appropriate. And all users will in effect have access to a complete history of MD5 file results.
A few minor code reshuffles changed VariantRecalibration and VCFHeader test files.
2011-07-18 10:46:01 -04:00
6f26c07b85
Removed the SpecificDifference class. Now Difference classes always have the option to remember specific master and test values. This means that all summarized differences carry with them specific examples of their differences. Consequently, now even summarized differences give at least one example of the specific difference, even when the count of the difference is > 1. Unit tests updated. Added DiffObjects integrationtest. VCFDiffableReader now specifically reads the first line of the VCF file to capture the version number.
2011-07-18 10:42:35 -04:00
b2b7d27fed
Merge branch 'laptop'
2011-07-18 00:25:46 -04:00
497721a799
Added class documentation string.
2011-07-18 00:25:21 -04:00
ac9c66138d
Merge branch 'master' of ssh://copper.broadinstitute.org/humgen/gsa-scr1/gsa-engineering/git/unstable
2011-07-18 00:20:33 -04:00
824100e57f
Corrected typo in MergeAndMatchHaplotypes integration test
2011-07-17 22:50:54 -04:00
8167aba601
Moved (poorly named) MergeAndMatchHaplotypes to public. Added integration test
2011-07-17 22:47:32 -04:00
afb506e128
Added MD5s for PhaseByTransmission integration tests
2011-07-17 21:55:33 -04:00
558e197989
Integration test for PhaseByTransmission
2011-07-17 21:25:08 -04:00
Matt Hanna
Ryan Poplin
Andrey Sivachenko
Eric Banks
Guillermo del Angel
Khalid Shakir
Christopher Hartl
David Roazen
Mark DePristo
Kiran V Garimella