zzh
/
gatk-3.8
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Commit Graph

  • d3b1732cca Start of refactoring effort. Make construction of alignment object simpler. hanna 2009-09-30 15:19:31 +0000
  • 70e1aef550 Better integrate the @ArgumentCollection into the command-line argument parser. Walkers can now specify their own @ArgumentCollections. Also cleaned up a bit of the CommandLineProgram template method pattern to minimize duplicate code. hanna 2009-09-29 22:23:19 +0000
  • b1c321f161 Adjusted Genotype concordance to more accurately use the new Genotyping code, fixed the VCF rod, and temp. fix the build by reintroducing Shermans ReadCigarFormatter aaron 2009-09-29 21:28:21 +0000
  • 9b78a789e2 HLA Caller 2.0 Walkers: CalculateBaseLikelihoodsWalker.java walks through reads calculates likelihoods using SSG at each base position CalculateAlleleLikelihoodsWalker.java walks through HLA dictionary and calculates likelihoods for allele pairs given output of CalculateBaseLikelihoodsWalker.java CalculatePhaseLikelihoodsWalker.java walks through reads and calculates likelihoods score for allele pairs given phase information sjia 2009-09-29 20:45:55 +0000
  • 281a77c981 Bugfix. isMismatch() was actually computing isMatch(). chartl 2009-09-29 20:04:59 +0000
  • e28b45688c More NQS Related Walkers to play with chartl 2009-09-29 20:01:04 +0000
  • 9ef80e3c3c One minor addition: to incorporate Pooled calling (and to be as general as possible), we allow the genotype calculation model to use rods if it wants. ebanks 2009-09-29 17:05:59 +0000
  • 19bfe43173 First pass at a unified caller, being checked in now so Mark can give feedback if he chooses and so Matt can debug issues with the ArgumentCollection class. ebanks 2009-09-29 16:48:15 +0000
  • 8bd345ba00 Generalized deletions in pileup ebanks 2009-09-29 15:58:43 +0000
  • 6134f49e3c Convert de novo SNP caller to run using parent1 and parent2 BAM files (by splitting contexts by reader using getMergedReadGroupsByReaders) instead of geli files providing a large speed-up and obviating the need for large whole-genome geli files. andrewk 2009-09-29 06:42:21 +0000
  • 5dab95aa5a Fix getMergedReadGroupsByReaders so that it provides read groups in the same way Picard does so that it works correctly when input read files have no clashes in their read groups and retain their original read group names. andrewk 2009-09-29 06:35:50 +0000
  • 5662a88ee1 Cosmetic change to list sampling functions: the typical usage of n and k were reversed. No change in functionality of the classes has been made and unit tests still pass. andrewk 2009-09-28 18:12:32 +0000
  • 39598f1f0a switching the concordance walker over to the new Variation system aaron 2009-09-28 15:46:36 +0000
  • bce2f0d7cf Now instantiates the list of alternative consenses to evaluate as LinkedHashSet to guarantee iterator traversal order. Old implementation used HashSet and exhibited unstable behavior when two alt consenses turned out to be equally good: depending on the run conditions (including size of the interval set being cleaned??), either one could be seen first as selected as the 'best' one asivache 2009-09-28 06:15:46 +0000
  • 663175e868 Bug fix: when jumping onto next contig (chromosome), the walker was erasing last mismatch interval from the previous chr it was still holding without printing it; now it gets printed. asivache 2009-09-25 22:24:34 +0000
  • 92c6efabb7 moving IndelGenotyper out of playground asivache 2009-09-25 19:44:49 +0000
  • aec61c558b moving IndelGenotyper out from playground asivache 2009-09-25 19:43:53 +0000
  • fe6d810515 Some basic commits that I've been sitting on for a while now: chartl 2009-09-25 19:35:50 +0000
  • f7684d9e1b ImputeAllelesWalker fills missing portions of HLA dictionary based on best allele matches sjia 2009-09-25 18:51:46 +0000
  • 235de38c2e Updates to FindClosestAlleleWalker and CreateHaplotypesWalker sjia 2009-09-25 16:41:58 +0000
  • 130a01a40a delete the integration test temp files when the test is over aaron 2009-09-25 16:34:08 +0000
  • 2b7d39035a switched over the FastaAlternateReferenceWalker to the Variation system aaron 2009-09-25 16:09:43 +0000
  • de9f2b11da Detects unmapped (no bai) bam files and doesn't blow up depristo 2009-09-25 12:56:28 +0000
  • 7ffc1d97ef Cut DeNovoSNPWalker over to the new Variation system, some renaming of methods on the Variation interface, and some corrections on the interface. aaron 2009-09-25 04:35:52 +0000
  • 392152f149 1000x performance improvements to MSG for crisis control depristo 2009-09-24 23:44:33 +0000
  • 44879c81b0 Add in weights. Massive performance improvements. hanna 2009-09-24 23:19:15 +0000
  • 3b79f9eddc Support 'N's and other mismatch characters in the reference. hanna 2009-09-24 21:41:30 +0000
  • 08e8d2183a Indels supported. Variable gap penalties are not yet taken into account. hanna 2009-09-24 21:03:02 +0000
  • d2af26e81f Pooled EM SNP Rod converted over to the Variation interface aaron 2009-09-24 16:33:11 +0000
  • 97105ac001 We need to return a null RODRecordList when the default value is null (as opposed to a list with a single null value), because that's what everyone is expecting. ebanks 2009-09-24 16:23:12 +0000
  • d4b40bc06f Filter for reads with missing read groups so we can safely assume all reads have valid read groups ebanks 2009-09-24 16:10:26 +0000
  • 90de2e0cde Added ability to specify whether you want to use a point estimate or fair coin test calculation; for now you can use either but fair coin test is still experimental as it needs to be parametrized correctly. This job will hopefully be done by the future Bioinformatic Analyst... ebanks 2009-09-24 15:29:50 +0000
  • d262cbd41c changes to add VCF to the rod system, fix VCF output in VariantsToVCF, and some other minor changes aaron 2009-09-24 15:16:11 +0000
  • 8349004414 Generalize the regexp for analysis files ebanks 2009-09-24 03:17:41 +0000
  • 1ee8ba590c Reads cigar files sjia 2009-09-24 03:14:10 +0000
  • 9422156e09 Finds closest allele for each read in bam file sjia 2009-09-24 03:12:20 +0000
  • 5c5151c4e7 Creates ped file from reads sjia 2009-09-24 02:48:29 +0000
  • b0ec7fc144 More comprehensive testing of BWT (mismatches only) module, and lots of bug fixes. hanna 2009-09-23 23:44:59 +0000
  • b446b3f1b6 CreateHaplotypeWalker now gives correct output sjia 2009-09-23 21:13:52 +0000
  • eeb14ec717 a couple of light changes to GenomeLocSortedSet. aaron 2009-09-23 20:38:53 +0000
  • 3916e165fb New walker to output haplotypes for each read (for SNP analysis or imputation, etc) sjia 2009-09-23 20:26:43 +0000
  • 423a3ee894 Added a sequenom rod to empower Carrie to convert 1KG validation SNPs to sequenom format ebanks 2009-09-23 20:22:09 +0000
  • 63f3d45ca4 fixing the build chartl 2009-09-23 20:04:09 +0000
  • 540e1b971f And we fix one boneheaded mistake, which was actually causing the problem; though the last change was still correct. chartl 2009-09-23 19:26:45 +0000
  • 124ca68fa8 And an IMMEDIATE minor fix (want neighborhood quality > base quality to be represented correctly) chartl 2009-09-23 19:21:09 +0000
  • 8cdb78ebee More sophisticated version of the NQSCovariantWalker - modified to be more explicit about how much higher the quality score of a particular base is than the quality score of its neighbors. The granularity of the binning jumps from 32 groups to 860 groups. chartl 2009-09-23 19:18:24 +0000
  • 856bbd0320 Let Picard specify the default compression level. hanna 2009-09-23 19:01:48 +0000
  • f783cb30e0 adding an interface so that the current @Requires with ROD annotations work in walkers like VariantEval aaron 2009-09-23 18:24:05 +0000
  • ebfbe56b43 Make sure compression level always gets pushed into SAMFileWriterFactory. hanna 2009-09-23 18:20:26 +0000
  • fa87dd386d Now uses rodRefSeq in its new reincarnation asivache 2009-09-23 18:19:36 +0000
  • bf7cd66d53 New, simpler rodRefSeq. Fully relies on the ROD system standard mechanisms. Multiple transcripts over a given location will be now returned by the ROD system itself as RodRecordList<rodRefSeq>; and yes, rodRefSeq does represent a single transcript record now and implements Transcript interface asivache 2009-09-23 18:18:25 +0000
  • 8fa4c93f5a Transcript is now simply an interface asivache 2009-09-23 18:13:31 +0000
  • fe36289e44 Noone needs this, probably... Old experimental code. asivache 2009-09-23 18:11:50 +0000
  • 1bd4c0077c Now that ROD system supports overlapping RODs, we do not need rodRefSeq to be too smart and read in all the overlapping records (transcripts) on its own; leave it to the generic ROD mechanism. asivache 2009-09-23 18:11:16 +0000
  • aa66074a0e Compares each read to the HLA dictionary and outputs closest allele, as well as other stats sjia 2009-09-23 16:17:23 +0000
  • 3928751bf5 Fail for missing resource files / scripts. hanna 2009-09-22 21:31:01 +0000
  • eebe2978a1 Add code to fail if any of the classes explicitly listed as a dependency is not found. Will still *not* fail if a package or script or resource file is missing. hanna 2009-09-22 21:22:27 +0000
  • 11c32b588f fixing VariantEvalWalkerIntegrationTest md5 sums, a couple comment changes, and a little bit of cleanup aaron 2009-09-22 20:54:47 +0000
  • b0fa19a0b2 Fixed recal integration test ebanks 2009-09-22 20:22:32 +0000
  • 0748d80baa Added a convenience method in rodDbSNP to deal with Andrey's changes to the rod. Now you can just ask for the first real SNP rod from the list and not have to think about how it works. CountCovariates uses it. ebanks 2009-09-22 20:15:40 +0000
  • 6780476fb5 updated to deal with new dbSNP rod ebanks 2009-09-22 19:46:32 +0000
  • 14477bb48e Unidirectional alignments with mismatches now working. Significant refactoring will be required. hanna 2009-09-22 19:05:10 +0000
  • 22932042ea Combined Scores, bug fixed for printing HLA-C sjia 2009-09-22 18:28:16 +0000
  • 682b765536 bug: need to upper case chars so that == works throughout ebanks 2009-09-22 18:20:43 +0000
  • 575da25fde R script for selecting a variety of baits (using %GC content and normalized coverage) for Nanostring assessment from those used in the Agilent whole exome hybrid selection design. andrewk 2009-09-22 18:10:14 +0000
  • d7d0b270d1 now supports blacklisting lanes (with -BL option will ignore reads from any of the specified lanes) asivache 2009-09-22 16:46:57 +0000
  • 57d31b8e9b Filter that discards reads from specific lanes; and also its friend that helps blacklisting a set of lanes from GATK command line a one-liner. asivache 2009-09-22 16:46:06 +0000
  • db9390811a Package tool doesn't provide an error if classes listed in the package are not found, and classes were moved from playground to core. This step: correct the classes listed in the package. Next step: update the packaging tool to fail if class is not found. hanna 2009-09-22 15:06:45 +0000
  • 83a9eebcc4 fixed a bug I checked in that Eric found, for intervals with no start or stop coordinate. Now I owe Eric a cookie, and Milk Street is so far away. Damn. aaron 2009-09-22 04:34:18 +0000
  • e9eeed8c6f add wait id for 454 ebanks 2009-09-22 01:41:58 +0000
  • 5ce42cbab3 After thinking about this a bit more, it makes sense to pull this functionality out of my walker and into the GenomeLocParser where everyone else can benefit from it... ebanks 2009-09-22 01:32:35 +0000
  • 7bfb5fad27 fixing the dbSNP test. Also removing unnessasary comments from the GenomeLocParser, added some tests, and commented out the performance test aaron 2009-09-21 23:32:24 +0000
  • 39a47491a9 changes to make GenomeLoc string parsing 25% faster aaron 2009-09-21 22:37:47 +0000
  • b1dc6d65e4 interval merging is now blazingly fast ebanks 2009-09-21 21:15:04 +0000
  • 15135788ca OK, let's bite the bullet. Now rodDbSNP objects are 'isSNP()' only when they are annotated as 'exact', not a 'range'. asivache 2009-09-21 19:25:16 +0000
  • 8ad181f46f Note to myself: do 'ant clean' now and then or old versions of the code that suddenly became invalid will stick around. The world is not perfect, and neither is automatic dependency resolution. asivache 2009-09-21 17:40:52 +0000
  • fb09835ef8 Changed to accomodate new ROD system asivache 2009-09-21 17:10:56 +0000
  • d2d1354199 Now uses BrokenRODSimulator class to pass the test. CHANGE the code to use new ROD system directly and MODIFY MD5 in corresponding tests, since a few snps are seen differently now. asivache 2009-09-21 17:03:49 +0000
  • f4d270cba4 These classes now use BrokenRODSimulator class to pass the test. CHANGE the code to use new ROD system directly and MODIFY MD5 in corresponding tests, since a few snps are seen differently now. asivache 2009-09-21 17:03:15 +0000
  • 29adc0ca1c Little class that can be used to simulate the results returned by the old ROD system. This is needed to keep couple of tests from breaking. All the code that uses this class must be changed urgently to accomodate the data as returned by new ROD system, and the corresponding tests (MD5 sums) have to be modified as well since some data as seen through the new ROD system is indeed different. asivache 2009-09-21 16:58:56 +0000
  • a6bd509593 Changing the carpet under your feet!! New incremental update to th eROD system has arrived. asivache 2009-09-21 16:55:22 +0000
  • 4c67a49ccb Removed unused imports asivache 2009-09-21 16:45:22 +0000
  • e7f44ada98 Make unpackList public static so that Doug can use it in the scatter/gather framework. hanna 2009-09-21 15:32:49 +0000
  • 7b627fd622 Check for empty interval lists to merge ebanks 2009-09-21 04:34:26 +0000
  • 7f5778c966 Update gsadevelopers -> gsahelp. hanna 2009-09-20 23:36:54 +0000
  • 3a487dd64e little fixes; also fixed a tyPo aaron 2009-09-18 22:38:51 +0000
  • b6d7d6acc6 fix for the eval tests, and a change to the backedbygenotypes interface, more changes to come aaron 2009-09-18 22:25:16 +0000
  • 4318f75910 tiny cleanup depristo 2009-09-18 21:04:25 +0000
  • 3a341b2f06 Fixes for VariantEval for genotyping mode depristo 2009-09-18 21:01:43 +0000
  • 7b39aa4966 Adding the VCF ROD. Also changed the VCF objects to much more user friendly. aaron 2009-09-18 20:19:34 +0000
  • d9ee515a9b rename some jobnames so that LSF doesn't complain ebanks 2009-09-18 19:02:15 +0000
  • 83e6e5a3e4 Calculates Probability for each allele combination (using likelihood score and allele frequencies only) sjia 2009-09-18 18:46:38 +0000
  • 4c89140f21 quick fix ebanks 2009-09-18 18:17:13 +0000
  • b19fd4d45c Damn unit tests have a null Toolkit()... ebanks 2009-09-18 17:10:49 +0000
  • 90626c843d oops - we don't need reference bases, but we still need reference ebanks 2009-09-18 16:24:45 +0000
  • 2b2df4e1ba - Fix the CleanedReadInjector to deal with -L intervals correctly. - Some walkers don't use the ref base, so speed up traversals by not requiring it ebanks 2009-09-18 16:17:58 +0000
  • 7da9ff2a9e Put back the check that both chip and variant are not null. Also, sanity check that ref is not 'N'. ebanks 2009-09-18 16:03:54 +0000
  • 94618044e8 Starting an update of ROD system. These basic classes will completely replace old ones, but with this update they are not linked to anything, so this checkpoint should be safe. asivache 2009-09-18 15:58:37 +0000
  • 66a4de9a1d Genotype check should be case-insensitive ebanks 2009-09-18 03:23:30 +0000
  • c186a49d55 Time for a reorganization. Repackage generally useful alignment classes lower in the package structure, and create a subpackage for bwa-specific code. Repackage BWA alignment code away from BWT representation. Isolate byte- and word-packing streams in another package that will ultimately be killed off en masse. hanna 2009-09-17 23:28:47 +0000