c5325b03be1) Removed hard-coded strings. Please let's use the fields defined in VCFConstants. 2) General code cleanup.
ebanks
2010-07-23 01:49:47 +0000
e9d243babbMore improvements to exception handling during multithreaded runs based on a bug reported by Ryan.
hanna
2010-07-22 22:13:01 +0000
83798225acRepackaged datasource-specific command-line tools into their own package. Added a tag renamer tool.
hanna
2010-07-22 19:50:34 +0000
98caedb5f0Forgot to update VCF4 unit test.
delangel
2010-07-22 16:25:51 +0000
485023ba8ethis.intersect(that) method added to GenomeLoc (returns intersection of two intervals or dies if the locations do not overlap)
asivache
2010-07-22 16:00:30 +0000
473ec91633a) Bug fix in VCFHeader parsing - Info fields were not being parsed properly, with the result that the Count field was not being properly displayed in records (e.g. if Count=0 for a particular field, the INFO tag was still being displayed as ...;Field=x;... instead of ...;Field;... b) Bug fixes and update to how we represent indels and other complex events in a VariantContext object. Convention is now that all events are left aligned, with the first variant context location marking the common base before an event occurs. However, alleles in a VC don't have the common base in all VC's. Two new functions are now part of VariantContextUtils: CreateVariantContextWithPaddedAlleles and CreateVariantContextWithTrimmedAlleles. Both take a VC as an input and create a VC as an output. Main flow is that a VCF reader would create a VC with trimmed alleles, all walkers would ideally work with these trimmed alleles, and then the VCF writer would pad back the alleles before writing. However, there are special cases where we need to pad alleles like for example when merging/combining VC's.
delangel
2010-07-22 02:36:45 +0000
b696c3ea98No more traversal reduce results.
chartl
2010-07-21 18:34:54 +0000
365b42390dSupport for generating (very basic) wiggle files for use with IGV (see UCSC for wiggle spec); and a walker to take in a variant track and create a transition transversion rate track for the whole genome (due to the wiggle spec, this has to be done by chromosome). It's interesting to see the effect of genes!
chartl
2010-07-21 18:04:30 +0000
3596b1529fmodule for listing out samples for data processing and firehose reporting
corin
2010-07-21 15:05:41 +0000
1209b165bfNow accepts command line args and prints paths to vcf, bams and beds
corin
2010-07-21 15:04:58 +0000
f7957bc7f2Fixed memory leak in VariantEval
depristo
2010-07-21 12:35:46 +0000
cebddb5bd3removing said file
aaron
2010-07-21 03:29:48 +0000
25d26ab6eea little experiment for better integration with tribble; using external definition to pull other repositories into ours seamlessly. This will be deleted in a few minutes.
aaron
2010-07-21 03:27:36 +0000
1cba81c16fupdates to tribble with fixes for some bugs I've found in some new indexing code.
aaron
2010-07-20 22:08:04 +0000
ff6748d1cdoops - missed one
ebanks
2010-07-20 18:55:19 +0000
c6ad26e04f1) When quals/GQs are really integers (x.00), strip off the floating points. 2) Keep track of whether vcf records are unfiltered vs. pass filters in the variant context so we can regenerate the records on output. 3) No more "ID" hard-coded all over the code to set the VariantContext ID. Use a static variable instead.
ebanks
2010-07-20 18:01:45 +0000
0db7fab1a9Fixing genotype filtering for VF and adding integration tests
ebanks
2010-07-20 07:30:21 +0000
2a6c2d3098re-enable test; I was moving the input file in prep for my last commit around on Eric, so he rightfully removed the test
aaron
2010-07-20 07:14:59 +0000
0108517b98updating the Tribble track loading code to use the new shared locks, updated lots of new tests, add infrastructure for the TreeInterval, and removed the old locking class.
aaron
2010-07-20 07:08:10 +0000
af6b5f000eupdating the Tribble library; added writing of indexes to the index interface for working with the tree index.
aaron
2010-07-20 07:02:08 +0000
f7429808641. Refactoring of GenoypeWriters so that parallelization now works again with VCF4.0. We now have just a single reference to the old VCF classes, and that one will be purged soon. 2. Moved Jared's VCFTool code into archive so that everything would compile. 3. Added the vcf reference base (needed for indels) as an attribute to the VariantContext from the reader. 4. TribbleRMDTrackBuilderUnitTest was complaining that a validation file didn'r exist, so I commented it out.
ebanks
2010-07-20 06:16:45 +0000
70b07206a2CombineVariants tests for Guillermo and Eric to explore the correctness of the in/out reader, writer behavior of the system.
depristo
2010-07-19 22:41:48 +0000
c47a5ff5abOfficial parallel CountCovariates, passes all integration tests. Now poster-child example of parallelism in GATK (Matt H). Apparent general performance improvements throughout too.
depristo
2010-07-19 22:13:18 +0000
0b56003d1aRemove stray commented out line
rpoplin
2010-07-19 19:14:39 +0000
8e31c01680Solid processing in base quality recalibrator now has several options for how to handle no calls in the color space. --ignore_nocall_colorspace is removed and replace by --solid_nocall_strategy. Fixed some of the @Deprecated tags in BaseUtils. LocusWalkers now filter out FailsVendorQualityCheck reads. HLA caller integration test bam file had bad vendor reads so its integration test changed.
rpoplin
2010-07-19 19:10:29 +0000
35ce367898adding the annotations for findbugs as dependencies in the GATK. They have to be in the default config so that we can annotate code without running findbugs.
aaron
2010-07-19 16:34:57 +0000
b990a22bacA very nice way of automatically plotting the results of a VariantEval run. All of the hard work is actually in the common R repository, gsacommons.R, including methods for creating a Venn diagram. It also provides a mechanism for the output of a VariantEval run to be loaded into a single list object.
kiran
2010-07-19 12:38:26 +0000
250ab70fedupdate the Tribble library too.
aaron
2010-07-19 05:00:37 +0000
f4cfb0f990The first step in integrating Jim's tree based index scheme: - changed to a better method for getting headers from Codecs - some removal of old commented out code in the GATKAgrumentCollection - changes for the rename of FeatureReader to FeatureSource - removed the old Beagle ROD - cleaned up some of the code in SampleUtils
aaron
2010-07-19 04:49:27 +0000
40a963541dUniquify the registered MXBean by adding an instanceNumber=... tag to the ObjectName. In the Queue-enabled future, we might want to come up with GUIDs (or at least semi-unique IDs) so that we could use JMX to track runtime attributes for multiple jobs running simultaneously.
hanna
2010-07-19 00:58:54 +0000
5a1a3fc79aFix bad VariantContext creation in unit test
ebanks
2010-07-18 20:21:01 +0000
7c42e6994fFindBugs fixes throughout the code base
depristo
2010-07-18 16:29:59 +0000
693672a461Refactoring the VCF writer code; now no longer uses VCFRecord or any of its related classes, instead writing directly to the writer. Integration tests pass, but some are actually broken and will be fixed this week.
ebanks
2010-07-18 13:19:56 +0000
379584f1bfRe-enable (most of) these tests. Guillermo will re-enable the other one when the VCF->VC conversion is done for indels
ebanks
2010-07-18 03:24:28 +0000
982947d328update to deal with partial indels (I/D with no bases) in the HM records
ebanks
2010-07-18 02:56:37 +0000
414ec6f20aRemoving version argument constructors that shouldn't be used. Temporary allow -- with global variant to indicate this should be removed -- header records without description fields. Real error checking in the headers.
depristo
2010-07-17 22:30:08 +0000
d40299840cindenting clean up
depristo
2010-07-17 22:28:28 +0000
9207c58b8fA fix for the integration test I broke on Friday on my way out the door -- some workflows using AlignmentContext were working with it in a way I didn't expect and wound up treating extended pileups as base pileups. I'll work to make sure the AlignmentContext interface is crystal clear.
hanna
2010-07-17 22:22:44 +0000
ea117957b9Add CountFunctionalVariants to local release (for firehose)
chartl
2010-07-17 01:36:16 +0000
55b756f1ccFirst step in major cleanup/redo of VCF functionality. Specifically, now: a) VCF track name can work again with 3.3 or 4.0 VCF's when specifying -B name,VCF,file. Code will read header and parse automatically the version. b) Old VCF codec is deprecated. Reader goes now direct from parsing VCF lines into producing VariantContext objects, with no intermediate VCF records. If anyone can't resist the urge to still input files using the old method, a new VCF3Codec is in place with the old code, but it will be eventually deleted. c) VCF headers and VCF info fields no longer keep track of the version. They are parsed into an internal representation and will be output only in VCF4.0 format. d) As a consequence, the existing GATK bug where files are produced with VCF4 body but VCF3.3 headers is solved. e) Several VCF 4.0 writer bugs are now solved. f) Integration test MD5's are changed, mostly because of corrected VCF4.0 headers and because validation data mostly uses now VCF4.0. g) Several VCF files in the ValidationData/ directory have been converted to VCF 4.0 format. I kept the old versions, and the new versions have a .vcf4 extension.
delangel
2010-07-16 22:49:16 +0000
75bea4881aModified SampleFilter to allow for multiple samples to be given. AminoAcidTransition now turns on when you give VariantEval the right commands.
chartl
2010-07-16 21:27:32 +0000
b0fc42906eBetter DOT support and updated recalibration pipeline
depristo
2010-07-16 20:54:51 +0000
36ac73cf9acomment out broken test until it can be fixed.
aaron
2010-07-16 20:04:40 +0000
96034aee0eCleanup for Steve Hershman's issue. In the midst of doing this, I discovered that the semantics for which reads are in an extended event pileup are not clear at this point. Eric and I have planned a future clarification for this and the two of us will discuss who will implement this clarification and when it'll happen.
hanna
2010-07-16 18:57:58 +0000
6aedede7f3Added Type.MNP to allowed variant context types; this does not break the tests (yet)
asivache
2010-07-16 15:50:25 +0000
2bdb011865Improvements to 1KG processing pipeline
depristo
2010-07-16 15:33:47 +0000
1dd8a28a5dAdded new query: isMNP(feature); returns true if dbsnp feature is multi-nucleotide polymorfism (e.g. a di-nuc TA ->CC)
asivache
2010-07-16 15:32:10 +0000
917469ef43This script produces information for a firehose job-finished email
corin
2010-07-16 14:44:52 +0000
ec94cfdf05remove unit test for VCF writer, it's not applicable now that we produce only VCF4. Guillermo, it's up to you if you want to adapt this or remove it.
aaron
2010-07-16 14:33:25 +0000
b29eda83bbParallelized CountCovarites! percent_ref_called_var now a standard genotype concordance module (for validation!). Really much smarter merging of headers for combineVariants. VCF codecs now actually look at the file version and blow up if they are the wrong versions. setHeaderVersion() in VCFHeaderLine.
depristo
2010-07-16 14:10:18 +0000
f293eb7de1Fix for Kim: for some ungodly reason, I was initializing the bins that were maintaining counts to 1 instead of 0.
ebanks
2010-07-16 03:40:29 +0000
e7e58d7129The SAM spec has now officially reserved my new tags for original cigar and original alignment start... except that OS has been named OP ('original POS')
ebanks
2010-07-16 00:09:36 +0000
81eef0d993DOT visualization with Queue. More sophisticated recalibation queue script with scatter/gather
depristo
2010-07-15 22:32:48 +0000
ab84ed8c68Fix for Mark: get rid of old program tags whose IDs clash with the recalibrator/realigner tag (including if the id has a .1 at the end, etc.). Keeping them around is dangerous because we don't know which one refers to the latest run of the tool on the bam.
ebanks
2010-07-15 19:13:50 +0000
dfddf8fd75- Bring the PaperGenotyper up to code. - Remove some old debugging cruft regarding handling of threaded engine exceptions.
hanna
2010-07-14 22:31:21 +0000
f65cba6b9aAdding support for shared file locking via a new class for file locking, FSLockWithShared. This will eventually take over for FSLock, the current file locking class - I'll work with Aaron to merge the tribble code that uses FSLock right now.
bthomas
2010-07-14 20:45:51 +0000
a8caa20378Previously the hierarchical microscheduler defensively coded around and reported exceptions of the walker itself, but didn't do a great job of catching framework exceptions. This became extremely unfortunate in the case where walkers caused exceptions that manifested themselves in the framework, such as when the walker opens more files than file handles are available.
hanna
2010-07-14 20:34:43 +0000
bf384f48e1Reverting previous change because it won't always work. More investigation needed.
ebanks
2010-07-14 19:13:17 +0000
e4bfb06888Check header type instead of rod type, since rod type will now be VC and not VCF
ebanks
2010-07-14 19:10:09 +0000
0226412b11Add GQ to list of genotype attributes for reg exp
ebanks
2010-07-14 19:01:11 +0000
78a4d8ec3dRemoving more references to VCFRecord
ebanks
2010-07-14 16:34:15 +0000
af23762778Removing more references to VCFRecord
ebanks
2010-07-14 11:54:23 +0000
a4f8d70d8doops, forgot to update this integration test
ebanks
2010-07-14 11:38:33 +0000
460283f6d2No more manually converting VariantContexts to VCFRecords. You should be utilizing VCs and not VCFRecords.
ebanks
2010-07-14 05:21:28 +0000
6b5c88d4d6The GATK no longer writes vcf3.3; welcome to the world of vcf4.0. Needed to fix a few output bugs to get this to work, but it's looking great. Much more still to come. Guillermo: hopefully this doesn't break your local build too badly.
ebanks
2010-07-14 04:56:58 +0000
530a320f28Intermediate commit of scatter/gather recalibation pipeline
depristo
2010-07-13 22:46:08 +0000
19a5830186Restore "type" annotation (but not genomechange or cDNA change, which are already encoded in the VCF)
chartl
2010-07-13 17:33:15 +0000
9d2a485532Update to AminoAcidTransition eval module
chartl
2010-07-13 17:12:03 +0000
9cc1a411b2Altering the formatting of the annotation to work better with VariantEval's AminoAcidTransition
chartl
2010-07-13 16:31:14 +0000
3db7fbb5e9Fix for added EOF in csv file
rpoplin
2010-07-13 16:09:48 +0000
9a05e8143dMove to 4.0 and away from VCFRecord.
ebanks
2010-07-13 15:54:54 +0000
6442dabf94Deleting/archiving as instructed
ebanks
2010-07-13 15:23:50 +0000
7e7da75d27Moving over to 4.0 and away from VCFRecord
ebanks
2010-07-13 14:07:10 +0000
d896d03554Moving VF to vcf 4.0. Still need to fix genotype filters.
ebanks
2010-07-13 11:39:51 +0000
76b3b39720Technically, Mark broke this with his commit earlier. But since I had an outstanding broken test, I lose and have to fix this one too...
ebanks
2010-07-13 03:58:38 +0000
2e445262f2Promotion to . for variable numbers of arguments
depristo
2010-07-12 22:53:53 +0000
297f15a60cProtect ProduceBeagleInputWalker against evil users who feed to it VCF's with indels, no variation sites or other interesting markers: Write to Beagle input only in biallelic SNP sites since that's the only thing Beagle can do.
delangel
2010-07-12 20:54:42 +0000
52c534a8f2Updating to VCF 4.0
ebanks
2010-07-12 20:18:30 +0000
5992b79159a) Simplify normalization code in ProduceBeagleInputWalker, as to always normalize, and use MathUtils.normalizeFromLog10 to do this. b) Several improvements to BeagleOutputToVCFWalker: 1. If a Hapmap input track is provided (e.g. -B comp,VCF,file), Hapmap sites will be annotated with Hapmap Allele count and allele frequency (key ACH, AFH). 2. If probability of correct genotype is lower than ncthr (optional argument provided by user, default = 0.0), walker will keep original calls instead of using Beagle calls. 3. Instead of annotating just whether Beagle had modified a site, annotate instead HOW MANY genotypes in a site were actually changed by Beagle.
delangel
2010-07-12 19:54:58 +0000
46c39f2d53Quick python scripts for going from genotype VCFs to site-only VCFs, and one to fix BC vcf files (which had "het" genotypes at non-variant sites)
chartl
2010-07-12 19:13:32 +0000
e50627a49e1. Updated tests and added integration test for liftover code. 2. Updated liftover code (and scripts) to emit vcf 4.0 and no longer depend on VCFRecord. 3. Beagle walker now also emits vcf 4.0.
ebanks
2010-07-12 17:58:18 +0000
221e01fb27deleting/archiving as instructed
ebanks
2010-07-12 16:59:45 +0000
8086ab1f75Pulled sample/header merging routines out of CombineVariants and into util classes. Added more generalized methods for retrieving samples. Updated the Beagle walkers to use these methods.
ebanks
2010-07-12 16:51:54 +0000
e75b3e13bdupdating unit test for previous fix
ebanks
2010-07-12 03:23:53 +0000
0427f3554bBug fix: valid fields were being stripped off the FORMAT for samples because String.match was used instead of String.equals. Also, please use VCFConstants from now on instead of hard-coding e.g. missing values into the code.
ebanks
2010-07-12 03:06:51 +0000
fb717fe128First pass needed to remove old VCF code: moving all VCF-related constants into a single unified class
ebanks
2010-07-11 07:19:16 +0000
6b960bd9c5Fix for Steve: genotype filters still want to see the values from the VC
ebanks
2010-07-11 04:30:15 +0000
1d399aa2f3Added a temporary gatkLoggingLevel field to the soon to be obsolete GatkFunction while finishing up the delayed generic gatk walker utility.
kshakir
2010-07-11 03:27:32 +0000
ebanks
hanna
delangel
asivache
chartl
corin
depristo
aaron
rpoplin
kiran
bthomas
kshakir