Permalink
Commits on Nov 3, 2016
  1. Merge pull request #328 from sambrightman/testspecifics

    Make some tests more specific
    committed on GitHub Nov 3, 2016
  2. Merge pull request #323 from mcshane/ad

    replace FORMAT/DPR annotation by FORMAT/AD
    committed on GitHub Nov 3, 2016
  3. Merge pull request #324 from mcshane/AO-RO

    clarify header description of AO and RO
    committed on GitHub Nov 3, 2016
  4. Merge pull request #327 from sambrightman/nh_fasta_iupac

    VCF's should not include IUPAC bases in the REF column.
    committed on GitHub Nov 3, 2016
Commits on Oct 11, 2016
Commits on Oct 3, 2016
  1. Merge pull request #316 from walaj/master

    SeqLib integration (HTSlib-based BamTools switch)
    committed on GitHub Oct 3, 2016
Commits on Oct 1, 2016
  1. updated seqlib

    walaj committed Oct 1, 2016
  2. update seqlib

    walaj committed Oct 1, 2016
Commits on Sep 30, 2016
  1. added back test suite

    walaj committed Sep 30, 2016
  2. added back region test suite

    walaj committed Sep 30, 2016
  3. updated seqlib

    walaj committed Sep 30, 2016
  4. update seqlib repos pointer

    walaj committed Sep 30, 2016
  5. removed debug print statement

    walaj committed Sep 30, 2016
  6. removed initial print statement

    walaj committed Sep 30, 2016
  7. clarify header description of AO and RO

    Closes #142
    mcshane committed Sep 30, 2016
  8. replace FORMAT/DPR annotation by FORMAT/AD

    DPR is mislabeled as a `Number=A` field, but writing out `Number=R` values.
    Also, as @nh13 noticed, the REF allele count was wrong anyway -
    printing the total depth rather than the REF depth.
    
    Replace by FORMAT/AD which is a Number=R field now defined in the spec. Also
    need to bump the VCF version to 4.2 (`Number=R` was introduced in 4.2, but AD in 4.3)
    
    Fixes #267
    mcshane committed Sep 30, 2016
Commits on Sep 29, 2016
  1. merged with upstream

    walaj committed Sep 29, 2016
  2. Merge pull request #319 from sambrightman/uniquesampleids

    Ensure FreeBayes fails on multiple BAMs with conflicting read groups
    committed on GitHub Sep 29, 2016
  3. Merge pull request #318 from sambrightman/regiontests

    Fix region tests and ensure that they are used
    committed on GitHub Sep 29, 2016
  4. lighter seqlib

    walaj committed Sep 29, 2016
Commits on Sep 15, 2016
  1. Add similar check for different sequencing technologies

    Unsure if stricly necessary, but seems like an error to map the same
    read group ID to different PL. Previously the map entry was blindly
    over-written. Add tests for this.
    sambrightman committed Sep 15, 2016
  2. Ensure failure when read group to sample mapping is not clear

    * Check BAM headers individually so that bamtools does not make them
      unique before FreeBayes can check them.
    * Add basic tests for multiple BAM input.
    
    This does not work for standard input, because the stream would be used
    up on the first read. It seems less likely that standard input is used
    in combination with other BAMs (but possible). More sustainable would be
    adding a method to MultiBamReader to provide the raw header text.
    sambrightman committed Sep 15, 2016
Commits on Sep 14, 2016
  1. Fix region tests and ensure that they are used

    * Region tests were not actually being run by 'make test'
    * Region tests were failing
    * Fix variant detection at start of region (caused by commit 840b412)
    * Add tests for IGV-style regions (commit a412676)
    * Refactor tests are little
    * Add -S for old samtools
    
    This is an *attempted* fix for the first-of-target problem; I cannot
    be sure that it is a correct fix. There is another line that
    increments the counter just above, but from what I can tell this one
    should stay as it is.
    sambrightman committed Sep 14, 2016
Commits on Sep 5, 2016
  1. tested multi-bam run

    walaj committed Sep 5, 2016
  2. makefile typo

    walaj committed Sep 5, 2016
  3. move to seqlib

    walaj committed Sep 5, 2016