Commit Graph

10 Commits (d5ac822e976f442fae74119e1aad320b8b7fbe71)

Author SHA1 Message Date
ebanks 318035c147 Fixing up the output system of the Unified Genotyper. Deprecating the -all_bases and -genotype arguments. Adding instead the --output_mode (EMIT_VARIANTS_ONLY, EMIT_ALL_CONFIDENT_SITES, EMIT_ALL_SITES) and --genotyping_mode (DISCOVERY, GENOTYPE_GIVEN_ALLELES) arguments. UG now does the correct thing when passed alleles (bound to the 'alleles' rod) to use for genotyping; added several integration tests to cover this case. This commit will break the batched calls merging script, but Chris knows this and is ready for it...
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@5288 348d0f76-0448-11de-a6fe-93d51630548a
2011-02-22 06:07:18 +00:00
chartl 2235245af0 PrivatePermutations generalized to compute transition counts and average probabilities (and thus was renamed). Changes in some pipelines to reflect the change. Bugfix in the batch merging pipeline (it would halt because the allele VCF for genotyping batches could become off-spec).
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4894 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-22 15:16:15 +00:00
chartl 7bc2049031 Updates and bug fixes to private mutations qscript and pipeline libraries. Hand filter strings are now not busted (boo to having to escape quotes); convenience method added to VariantCalling to propagate standard trait data to a given GATK command line -- should be made more scala-esque in the future.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4824 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-13 04:55:13 +00:00
chartl cf75caf653 java changes:
VariantEvalWalker's logger is made public, so that variant eval modules can access it through the parent object.
 DesignFileGenerator comment lists how best to bind things to it, and the feature accessor is better refined to grab the genome loc. (old change)

scala changes:

convenience addAll( List[CommandLineFunction] ) added to QScript class (and thus removed from the fCPV2)
useful command line functions added to a new library package for command line functions (these are fast simple VCF command lines)
bug fixed in ProjectManagement for the class where there's only one batch to be batch-merged (not really part of the use-case, but an edge-condition that came up during pipeline testing)
first draft of a private mutations pipeline which will be elaborated in future



git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4823 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-12 05:10:45 +00:00
chartl f8dd59c1d1 Tightening of the batch merging pipeline. Optimized to run on hour queue, so please: if you run this, crush 'hour' with it. Testing is forthcoming, but it merged 700 samples overnight.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4805 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-08 14:36:23 +00:00
chartl 02de9a9764 With multi-sample genotyping must come scatter+gather. Also Khalid informed me of the .group(size) method, so removing my useless (but pretty) code.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4797 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-06 20:12:23 +00:00
chartl f4c43f013f Due to the overhead for reading VCF files (>32g for 700 5MB VCF files), batched merging has to generate likelihoods in batches.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4796 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-06 18:23:54 +00:00
chartl 0944184832 Major refactoring of library and full calling pipeline (v2) structure.
Arguments to the full calling qscript (and indeed, any qscript that wants them) are now specified via the PipelineArgumentCollection

Libraries require a Pipeline object for instantiation -- eliminating their previous dependence on yaml files

Functions added to PipelineUtils to build out the proper Pipeline object from the PipelineArgumentCollection, which now contains 
additional arguments to specify pipeline properties (name, ref, bams, dbsnp, interval list); which are mutually exclusive with
the yaml file.

Pipeline length reduced to a mere 62 lines.




git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4790 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-05 02:33:54 +00:00
chartl 670ae814b3 Get rid of files from the grep string
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4773 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-02 18:39:59 +00:00
chartl 220fb0c44a Added a pipeline for merging batches. For now takes a file containing a list of VCFs, and a file containing a list of bams. Does not do anything smart (e.g. if you leave out some .bams or add some extra ones, you will not be warned). Heavy lifting done in (the beginnings of) a library for managing multi-batch or multi-project tasks.
git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4771 348d0f76-0448-11de-a6fe-93d51630548a
2010-12-02 07:31:59 +00:00