Skip to content

Latest commit

 

History

History
174 lines (126 loc) · 4.84 KB

CHANGELOG.md

File metadata and controls

174 lines (126 loc) · 4.84 KB

CHANGELOG

v0.2.5: 2020-03-03

Added:

  • GP-2295 Downsampled input BAM as argument
  • GP-2302 Filtered read totals in JSON output

v0.2.4: 2020-01-24

Added:

  • GP-2252 Additional metadata fields for merged BAM files

v0.2.3: 2020-01-14

Added:

  • GP-2234 Export picard2 duplication metrics histograms
  • GR-968 Aggregate and export picard duplication metrics calculated from merged libraries.
  • GR-986 Run a fast subset of metrics for RNASeqQC
  • New command-line script write_fast_metrics.py
  • Shared functionality refactored into base and validator classes
  • Updated tests and documentation

v0.2.2: 2019-12-03

Add support for picard MarkDuplicates metrics input that is missing ROI histogram.

Added:

  • GR-908 Support missing ROI histogram

v0.2.1: 2019-11-26

Add support for picard 2.21.2 MarkDuplicates metrics histogram input.

Added:

  • GR-883 Support picard 2.21.2 input

v0.2.0: 2019-09-19

Increment minor version as we deploy bam-qc-metrics for production testing

Fixed:

  • GP-2126 VERSION file installed to unexpected location
  • Move VERSION to /etc/versions/bam_qc_metrics/VERSION and find correctly from __init__.py

v0.1.8: 2019-09-12

Fixed:

  • Bugfixes GP-2115
  • Correctly update ur_stats dictionaries
  • Run samtools in subprocess instead of pysam.stats; enables error handling

v0.1.7: 2019-09-10

Complete set of metrics and JSON output schema for first production release.

Enhancements/bugfixes may be needed before production release, but metrics should not change.

Fixed:

  • GP-2055 Run without error if --target not given

Added:

  • GP-2048 Downsampling to exact number of reads
  • Command-line options --all-reads and --sample control sample level
  • Default number of reads to sample is approximately 1.1 million
  • Downsampling has no effect if fewer than 1.1 million reads present
  • Number of reads actually sampled is approximate, due to limitation of samtools
  • GP-2057 Logging and profiling
  • Command-line options for --log and --profile
  • --verbose and --debug options set the log level
  • Option to set random seed, with updated tests
  • GP-2045 Implementation of additional bedtools metrics
  • GP-2093 Bedtools coverage depth metrics
  • Additional metrics: Total coverage, coverage per target, target size, coverage histogram

Changed:

  • GP-2095 Improved tmpdir handling; log and cleanup on fatal error
  • GP-2099 Update diff script for changes to JSON format

v0.1.6: 2019-08-16

Fixed:

  • Bugfix: setup.py now copies the VERSION file to the installation directory

v0.1.5: 2019-08-16

Fixed:

  • GP-2035 More efficient finding unmapped reads
  • GP-2036 Failure to skip BED file headers
  • GP-2037 Count mismatches using MD tag, not CIGAR string
  • GP-2042 Correct reads per start point calculation

Added:

  • GP-2038 Report software versions in JSON
  • Package version is read from the VERSION file
  • Specify workflow version using the '-w' option for run_bam_qc.py
  • Convenience script update_test_data_version.py to update package version in test data
  • GP-2045 Placeholders for additional bedtools metrics

Changed:

  • GP-2040 refactoring
  • Separate classes for "fast" metrics (found before downsampling) and "slow" (after)

v0.1.4: 2019-07-26

Fixed:

  • Bug in downsampling; samtools view expects downsample rate as decimal, not integer

v0.1.3: 2019-07-26

Fixed:

  • Use correct read length if all data is 'unknown read'
  • Correctly handle empty read length histogram
  • Correctly handle missing FFQ/LFQ in samtools stats

Added:

  • 'Reads per start point' metric
  • Sanity checks on output variables in Python tests

Changed:

  • Run samtools stats before downsampling and after quality filtering (if any)
  • Do downsampling using samtools view and a random seed instead of iterating over the reads
  • Rename trim_quality parameter as skip_below_mapq
  • Replace obsolete distutils with setuptools in setup.py

v0.1.2 : 2019-07-23

Fixed:

  • Count unmapped reads before, not after mapping quality filter is applied
  • If read is aligned to reverse strand, also reverse order of CIGAR tuple
  • Fixups for reading mark_duplicates text file

Added:

  • Metric specifications document metrics.md

Changed:

  • Initialize all instance variables at start of __init__() in bam_qc.py

v0.1.1 : 2019-07-12

Fixed:

  • Do not crash on empty CIGAR string

Added:

  • Add install_requires in setup.py, allows pip to install prerequisites

Changed:

  • Move diff.py script to bin/ and rename as diff_bam_qc_output.py

v0.1.0 : 2019-07-11

Initial release for testing

Added: