MicroHapulator 0.5

Added

• New --base-qual parameter for mhpl8r type to set the minimum required base quality when iterating over reads in a pileup (#83).
• New mhpl8r balance subcommand for calculating and visualizing interlocus balance (#85).
• Users can now supply marker definitions, frequences, and reference sequences as TSV/FASTA files instead of MicroHapDB references (#93).
• Configuration file examples in microhapulator/data/configs/ (#105).
• New mhpl8r filter subcommand with support for marker-specific thresholds (#113, #114).
• New mhpl8r convert subcommand for converting genotype calls into a format compatible with probgen tools such as LRMix Studio and EuroForMix (#115).

Changed

• Updated mybinder demo (see #69, #110, #113).
• Simulated Illumina sequencing now uses 1 thread by default, which paradoxically leads to better performance (#71).
• Moved panel definition code moved out of the core code and into dedicated notebooks (#74).
• Replaced MissingBAMIndexError with BAM auto-indexing code (#78).
• Improved read names and choice of interleaved or paired output for mhpl8r seq (#80).
• Updated filtering of haplotype calls / typing results
  • Replaced --threshold argument with --static and --dynamic, disabled both by default (#82, #83).
  • Split mhpl8r type subcommand into type and filter, with --static and --dynamic arguments only relevant to the latter (#113).
• Changed the default pysam pileup max_depth parameter, overriding 8000 with 1e6 and exposing as a CLI parameter (#87, #113).
• Removed dependency on MicroHapDB for marker definitions, frequencies, and sequences (#93).
• Refactored CLI and Python API, adding new microhapulator.api module to serve as main entry point (#98, c98bf6c).
• Replaced the "ObservedProfile" terminology with the more appropriate "TypingResult" (#99).
• Documentation now uses Sphinx to render markdown as HTML (c98bf6c, #101, #102, #105, #106, #110).
• Updated JSON schema for simulated profiles and typing results (#109).

Fixed

• Corrected a bug with Fastq headers in mhpl8r seq module (#71).
• Corrected a bug resulting from attempting to do set operations on None (#75).
• Corrected a bug with RMP implementation (#86).
• Set minimum versions for runtime dependencies (#97).
• Corrected the --dynamic filter to operate on total haplotype counts rather than average counts (#114).

MicroHapulator 0.4.1

Fix minor metadata typo

MicroHapulator 0.4

Added

• New contain module for calculating the proportion of alleles from one sample present in another sample (see #41).
• New --dry-run mode for sim and mixture modules (see #41).
• New prob module for calculating likelihood ratio tests based on the random match probability (see #43).
• New seq module focused entirely on sequencing samples where profiles have already been simulated (see #45).
• New mix module for merging simulated profiles into a simulated mixture sample profile (see #45).
• New unite module for "mating" two profiles to create a simulated "offspring" profile (see #47).
• New diff modfule for showing the differences between two profiles (see #58, #60).

Changed

• Huge refactoring effort to accommodate for recent changes to MicroHapDB's Python API (see #66).
• The sim module no longer performs simulated sequencing (now handled by new seq module) and instead focuses entirely on haplotype simulation (see #45).
• The type module now dynamically selects either an automatic threshold or a fixed threshold based on effective coverage (see #51, #61).
• Moved simulation scripts to notebook directory, reimplemented as a Snakemake workflow (see #50, #57, #59).

Fixed

• Corrected a bug in the USA panel code ensuring that all loci have allele frequency data for all relevant populations (see #56).
• Corrected a problem with the contrib API that prevented it from working directly on Profile objects (see #70).

Removed

• Dropped the mixture module, whose functionality is now covered by the more granular sim, mix, and seq modules (see #45).
• Dropped the LocusContext class in favor of MicroHapDB's TargetAmplicon class designed for a similar purpose (see #67).

MicroHapulator 0.3

Added

• New interactive demo using Jupyter notebook and mybinder.org (see #32).
• Scripts for simulating 8k individuals with demographics roughly matching those of the U.S.A. (see #13).

Changed

• Updated dist and genotype modules to better support equality and distance comparisons between simulated and observed genotypes (see #34).
• Simulated and observed genotype data now use a single unified JSON representation, enforced by JSON schema validation (see #38).

MicroHapulator 0.2

Added

• New getrefr module for installing human reference genome (see #30).
• Improved error handling related to BAM index files (see #31).

Changed

• Replaced pip/PyPI installation instructions with bioconda installation instructions (see 1b61c59).

Fixed

• Corrected usage statement for mhpl8r contrib (see #31).

MicroHapulator 0.1.3

• mark failing test
• exclude __init__.py from test invocation

MicroHapulator 0.1.2

Updating file manifest, troubleshooting bioconda build.

MicroHapulator 0.1.1

Minor fix to setup.py for distribution purposes.

MicroHapulator 0.1

Initial release of MicroHapulator!

Command-line entry point: mhpl8r
Command-line operations:

• contrib: estimate number of contributors in a sample
• dist: compute naive Hamming distance between two sample genotypes
• mixture: simulate a multiple-contributor sample
• refr: retrieve reference genome cutouts for a specified microhap panel
• sim: simulate a single-contributor sample
• type: infer genotype from mapped reads