3.2.0

3.2.0

• limit samtools sort to using 6 threads at 500M ram each.
• tie_break will detect missing bam file and produce empty results.
• tweak tie_break params to reduce spurious low depth hits. (May mildly impact flu results).
• remove old unused dynamic comp mapping.
• When running dynamic comp mapping (with tie break). Can provide fixed ani groups via metadata file

3.1.1

What's Changed

• Fix empty df schema by @mcolpus in #131
• Release/3.1.1 by @mcolpus in #132

Full Changelog: 3.1.0...3.1.1

3.1.0

What's Changed

• fix: produce sensible output with empty bam by @mcolpus in #127
• Mmm 1217 extra outputs by @mcolpus in #123
• ci: move cargo test into docker build by @mcolpus in #128
• Release/3.1.0 by @mcolpus in #129
• bump: version from 3.1.0 (3.1.0) to 3.1.0 by @mcolpus in #130

Full Changelog: 3.0.2...3.1.0

3.0.2

3.0.2

• Correct pipeline names
• Correct container prefixes

3.0.1

3.0.1

• Correct file extension for build system

3.0.0

3.0.0

• add tie_break approach for determining read depth and coverage from bam file
• python script extract_reads for selecting reads for reference with tie_break
• group targets by ANI for use in tie_break
• rename output channels to be more consistent/descriptive
• code to produce new manifest from RefSeq in reproducible way

2.2.3

2.2.3

• Same container image name on both tenancies

2.2.2

2.2.2

• Rename flu manifest (doesn't affect myco)
• Improve H7 reference for flu (doesn't affect myco)
• Parameterise container prefix

2.2.1

2.2.1

• Remove duplicate mappings (only affects "intermediary file" output, not used in UI)
• Code to generate flu manifest (and example of output)

2.2.0

2.2.0

Feat

• Breaking: Make reference name a parameter so reads mapping to references other than M. tuberculosis can be output if needed

Chores

• Split the minimap2 and the analysis steps into separate python scripts
• manifest_builder params relabelled to remove gtdb as can use any db
• support supplying multiple dbs to manifest building

Tests

• Tests for reference name as a parameter, and for flu specifically