sort by:
Revision Author Date Message Commit Date
406f61d add periodic announcement while waiting on lock 14 April 2014, 21:24:03 UTC
b5f5617 set a minimum timeout 14 April 2014, 21:24:02 UTC
f8561f3 move wait_announce_interval default up to lock_resource 14 April 2014, 21:24:01 UTC
7a8d5d0 Rely on output_file_path instead of get_vcf 14 April 2014, 21:18:15 UTC
c522ddb is_concrete is now a public method 14 April 2014, 21:08:22 UTC
0245fc2 Fixed typo in software_result class 14 April 2014, 20:49:28 UTC
8a68633 Don't crash build when vcf sample name is different from aligned-reads This happens for example in samples from TCGA because of some logic here: Genome/Model/Tools/Vcf/Convert/Base.pm 14 April 2014, 18:57:28 UTC
a9c1137 Annotation::Vep classes inherit from Annotation::Detail base classes 14 April 2014, 16:14:45 UTC
c92a10e Name class correctly and remove unused params before passing on 14 April 2014, 16:13:00 UTC
ed814f9 Factor out test_cmd_and_result_are_in_sync to TestHelpers 14 April 2014, 16:09:15 UTC
16498e5 Add Command and Result to add readcount info to a vcf_result 14 April 2014, 15:27:33 UTC
64e7621 Change inputs to AnnotateWithReadcounts We cannot rely on UR to keep is_many properties in any particular order, so we have to combine these two fields. While we're at it, since sample_names aren't in fact useful except to get at the column index in the vcf-header, lets go ahead and just require the index. 14 April 2014, 15:27:33 UTC
7ce1aa5 Fix up how to get param_names and input_names In particular, we have to be careful with params on software results. 14 April 2014, 15:27:32 UTC
77092ef File::Vcf::Writer can write bgzipped files 14 April 2014, 15:27:32 UTC
f933d61 Add sample_name accessor for Readcount::Result 14 April 2014, 15:27:32 UTC
e43910c Make base classes for Annotation commands and results 14 April 2014, 15:27:32 UTC
abc22f5 Moved command into gmt 14 April 2014, 15:27:32 UTC
bdb8d0b ClinSeq: rm redirect to dev/null for illumina_info command to investigate failure 11 April 2014, 21:35:50 UTC
f59afd4 updated jenkins submodule to 20afc03 11 April 2014, 21:33:00 UTC
b9f5fdd Added Genome::Annotation::Vep and accompanying result and tests. 11 April 2014, 20:38:38 UTC
b228d96 Add get_tabix_and_gzipped_bed_file to Genome::Featurelist. This will be used for VEP annotation. 11 April 2014, 20:38:37 UTC
67fb1b3 Added gzip_file to Genome::Sys, which compresses a file to a target path. 11 April 2014, 20:38:37 UTC
a0e4f64 added vcf.gz test 11 April 2014, 20:37:03 UTC
fe24c91 Added VCF test. Updated test directories. 11 April 2014, 20:36:48 UTC
4f26a83 Merge branch 'updated' into gb_master 11 April 2014, 19:01:08 UTC
ccbb5b4 Improve tests for Adaptor::SomaticVariation 11 April 2014, 18:36:52 UTC
6090d88 Add abstract methods to abstract class Adaptor 11 April 2014, 18:36:51 UTC
6c7aaf4 Remove unneeded temp variable $results 11 April 2014, 18:36:50 UTC
75f7c47 Command to add readcount to vcf sample fields We assume that all readcount files are sorted in the same order as the vcf file, and that they contain only variants in the vcf file. 11 April 2014, 18:36:50 UTC
0c82324 Get original string from bam-readcount entry 11 April 2014, 18:36:49 UTC
7b2e71b Catch exceptions if vcf results don't exist There may be builds without vcf results, so we can catch those exceptions and simply return undef. 11 April 2014, 18:36:49 UTC
c761754 Throw exception if files don't exist 11 April 2014, 18:36:48 UTC
d405b9b Catch exceptions if vcf result is not available 11 April 2014, 18:36:47 UTC
bd7a3be adjust UR constraint to match DB constraint 11 April 2014, 17:19:02 UTC
5d511c0 migration to widen name column 11 April 2014, 17:15:06 UTC
016d2e7 change subject->sample for somval importer 11 April 2014, 15:21:29 UTC
702ed35 Updated `gmt` tab completion. 11 April 2014, 12:18:38 UTC
da0e271 Updated `genome` tab completion. 11 April 2014, 12:17:54 UTC
c045cf0 Updated class browser cache. 11 April 2014, 12:16:21 UTC
10d3651 Revert "switch _new_style_lock to retry" This reverts commit eccbdeaf713350c4f7c80b70626a62e34ad794c2. 11 April 2014, 02:15:50 UTC
fb00a65 adding option to dump mean and median coverage to a file 10 April 2014, 23:20:45 UTC
28b1ab0 Sync/GM: do not die if reallocate fails 10 April 2014, 22:41:51 UTC
00bf30d correctly match up coordinates from assembly validation 10 April 2014, 20:18:00 UTC
1104d8d report VAF percentages (50) instead of fractions (0.50) for large indels pulled from assembly validation 10 April 2014, 19:51:06 UTC
eccbdea switch _new_style_lock to retry In order to provide periodic status updates while waiting on locks like the NFS backend did switch _new_style_lock's timeout to block_sleep and retry max_try times instead of doing one block_sleep * max_try attempt. 10 April 2014, 19:26:17 UTC
7567de7 move defaults up so they get passed to both backends 10 April 2014, 19:26:16 UTC
569f0fa Don't die when variants-directory isn't found. This behavior is relied on in numerous places in ReferenceAlignment builds. In SomaticVariation and SomaticValidation, I didn't see any eval statements preceeding the calls to ->variants_directory so I think it is safe to relax this check. 10 April 2014, 15:48:57 UTC
c684828 Merge branch 'clip_overlap' 10 April 2014, 13:12:12 UTC
f539988 implement shortcut method for RefineReads 10 April 2014, 13:10:14 UTC
2b586fc Updated `gmt` tab completion. 10 April 2014, 12:19:06 UTC
8d4d335 Updated `genome` tab completion. 10 April 2014, 12:18:25 UTC
4cfd38e Updated class browser cache. 10 April 2014, 12:16:47 UTC
e315550 Somatic variation adaptor for annotation subsystem 09 April 2014, 21:37:53 UTC
3749246 reduce usage metrics rate 09 April 2014, 21:29:52 UTC
e526512 combine logic 09 April 2014, 21:27:39 UTC
3853107 add logic to reduce the amount of logging We have been logging 100 records per second (8.6M per day). To reduce this we will now only log 'gmt' and 'genome' commands on blades. 09 April 2014, 21:27:38 UTC
c05b290 memoize import instead of running on require 09 April 2014, 21:27:38 UTC
681ac65 extract should_record_usage 09 April 2014, 21:27:37 UTC
c481639 extract method genome_path to avoid uninitialized value warning 09 April 2014, 21:27:36 UTC
9f68ff8 Add accessor for DV2 detailed vcf results 09 April 2014, 21:13:16 UTC
d558061 croak if child fails to acquire lock 09 April 2014, 20:45:50 UTC
312853d Allow imported data if it comes from a bam file 09 April 2014, 18:59:52 UTC
f4c429d Use MixIn class to access DV2 results. 09 April 2014, 17:54:04 UTC
8afe9a5 Make accessors for get_detailed_{snvs,indels}_vcf 09 April 2014, 17:54:04 UTC
b4f5d52 use a default creation_time (epoch) To avoid uninitialized value warnings use a default value for creation_time (epoch). Otherwise assumptions about creation_time being defined, e.g. trying to sort allocations by creation_time, may produce uninitialized value warnings. Test that creation_time is set to ensure this didn't break the creation_time accessor. CHANGELOG: Genome::Disk::Allocation's creation_time now returns a default value, the epoch, instead of undef. 09 April 2014, 17:29:12 UTC
a67926e remove ability to try to lock the same resource twice I searched for "wait_on_self" and only found it set in its own test. With wait_on_self defaulting to false if something tried to lock the same resource twice it would unlock its current lock and then acquire a new lock. This should almost certainly be a fatal error because it probably means something else forgot to unlock its lock or if this is a means of "sharing" a lock it means something else could acquire the lock out from under this process. CHANGELOG: It is no longer possible to lock the same resource twice (in the same process). 09 April 2014, 17:27:36 UTC
6887b3a Make Annotation/Readcount command and SR 09 April 2014, 15:45:00 UTC
8f01f66 Whitespace 09 April 2014, 15:44:10 UTC
765c7fd Add ability to *not* remove filtered calls 09 April 2014, 15:44:10 UTC
56b4b87 Updated `gmt` tab completion. 09 April 2014, 12:18:55 UTC
bb3fbd1 Updated `genome` tab completion. 09 April 2014, 12:18:10 UTC
d215b5a making sample-names a required input 08 April 2014, 21:25:43 UTC
c635076 updated graphite submodule to fe06c16 08 April 2014, 20:18:48 UTC
1667b04 update doc to indicate that IDs or Names are now accepted 08 April 2014, 18:18:26 UTC
c5bf0d1 make SomVal AnP pairing importer support sample names as well as ids 08 April 2014, 18:16:24 UTC
7c822d2 handle return value of _validate_results correctly 08 April 2014, 18:02:27 UTC
c695545 link merge result to first refinement step only 08 April 2014, 18:02:00 UTC
72ce0b7 Don't try to shortcut to vcf in build 08 April 2014, 17:40:03 UTC
9e68bcf replace all usages of merged_alignment_bam_path with bam_file 08 April 2014, 17:40:02 UTC
6ba655e use the bam_file property instead of merged_alignment_bam_path so it also works for clip overlap and gatk results 08 April 2014, 17:40:01 UTC
3d92b62 only hook up the last step to the output connector 08 April 2014, 17:40:00 UTC
79349f5 used packaged version 08 April 2014, 17:39:59 UTC
4a0a434 update submodules to include bamUtil 08 April 2014, 17:39:58 UTC
6ab854d test for clip overlap, then gatk 08 April 2014, 17:39:57 UTC
631d645 index bam after clipping since downstream tools might expect index 08 April 2014, 17:39:57 UTC
cd557bb rename input/output_bam to input/output_file the tool supports both bam and sam add a file_format parameter to distinguish the two cases 08 April 2014, 17:39:56 UTC
96414e9 use the bam_path provided by Genome::InstrumentData::AlignedBamResult instead of custom output_bam_path 08 April 2014, 17:39:55 UTC
1f26f6b handle wiring up of two refinement operations to each other 08 April 2014, 17:39:53 UTC
f8bdae4 refactor the refiner input handling to work with multiple refiners 08 April 2014, 17:39:52 UTC
1e711f8 return the shortcut result when shortcutting, not 1 08 April 2014, 17:39:52 UTC
bdb1d1f bugfixes in the test The gatk_results need to be uniquified between the different subtests. This is achieved by adding the know_sites build to the retrieval method. For the gatk and clipoverlap combined test the retrieval of the clip overlap result should use the gatk results since that is the previous step (not the merge). 08 April 2014, 17:39:50 UTC
aa7c76b subtest renamed for clarity 08 April 2014, 17:39:49 UTC
8bb3b5a Added a bunch of code to begin considering multiple refiners. Refiner operation names and refiner properties on the input_connector are now unique by refiner name (multiple instances of the same refiner will not work right now). Some code cleanup is needed. Also we still need to make the "wire" methods work for multiple refine operations. 08 April 2014, 17:39:48 UTC
f0a4cab new test for gatk and clip overlap refinement (doesn't pass yet) 08 April 2014, 17:39:47 UTC
8fbe7fd RefineAlignments should really be the RefineReads base class 08 April 2014, 17:39:46 UTC
a4f269f refactor for clarity 08 April 2014, 17:39:45 UTC
84cb130 Added a test for running clip overlap after merge instead of gatk. Passes. Had to add a hack -- optional known_sites in clip_overlap. This should be pulled into a base class for refine reads and made optional (but required for gatk) 08 April 2014, 17:39:45 UTC
8b24b29 removed breakpoint 08 April 2014, 17:39:44 UTC
7ab8a00 removed unnecessary id specification when defining objects 08 April 2014, 17:39:43 UTC
81cefaf Use instrument data objects instead of hardcoded ids 08 April 2014, 17:39:42 UTC
back to top