Release 0.3.2 (pending full-pipeline validation)

[qclayssen]

Merges release/0.3.0 into main. Covers bolt 0.3.0 → 0.3.1 → 0.3.2.

0.3.2

Fixes (sash #52)

• PCGR_MUTATION_HOTSPOT=. treated as truthy in retention check — was preventing tiered filtering from running for any sample with >450k PASS variants (#32)
• Graceful PCGR skip when select_pcgr_variants cannot reach MAX_SOMATIC_VARIANTS — logs warning, continues without cancer report, non-PCGR outputs still publish (#32)
• Disable --estimate_msi/--estimate_tmb in chunked PCGR annotation runs — estimates on partial VCFs are not meaningful
• Pin jlumbroso/free-disk-space CI action to v1.3.0
• Handle duplicate PCGR/CPSR TSV keys instead of asserting
• Check SAGE VCF headers against sage_vcf_fp, not input_fp
• Disable PCGR estimates for hypermutated tiered-selection path

Tests added: TestRunSomaticCommandArgs (estimate flags), TestEntrySkipsPcgrOnOverflow, TestSelectPcgrVariantsRaisesOnUnresolvableOverflow, disable_estimates forwarding in TestRunSomaticChunkArgMapping, duplicate-key regression coverage, bcftools-guarded lossless regression for merge_vcf_files, comprehensive unit tests for bolt/common/pcgr.py and bolt/util.py

0.3.1

Fixes

• Drop --estimate_signatures from PCGR somatic invocation — signature analysis is downstream in gpgr/sigrap; keep --estimate_msi/--estimate_tmb (#31, sash #57)
• merge_vcf_files wrong output filename — Path.with_suffix() stripped .pass component
• VCF writers not closed in transfer_annotations_somatic/transfer_annotations_germline — BGZip output could be truncated
• split_vcf writing uncompressed .vcf chunks — now .vcf.gz with tabix index
• PCGR_ACTIONABILITY_TIER VCF header description mismatched stored short-form values
• pkg_resources missing in bolt:0.3.0-multiqc — add setuptools <81
• Bump r-gpgr to 2.3.1

0.3.0

• PCGR/CPSR 2.2.5, new reference data 20250314
• Hypermutation handling — chunked annotation + tiered selection for >450k PASS (#9)
• oncoanalyser v2 / eSVee SV caller (#6)
• gpgr 2.2.0 (#14)
• DRAGEN HRD file optional (#17)
• PCGR/CPSR argument handling improvements (#3)

Prerequisites

• bolt #35 — Release 0.3.0 + 0.3.1 validated (merge first)
• bolt #36 — single-chunk bcftools merge guard

[Copilot]

Pull request overview

Copilot reviewed 17 out of 17 changed files in this pull request and generated no new comments.

Suppressed comments (2)

tests/test_pcgr.py:303

• TestMergingPcgrFiles shells out to bcftools via util.execute_command(...), so it will fail in environments where bcftools is not installed. This contradicts the stated goal that the unit test suite should run binary-free, and also makes local runs brittle. Guard this class (or the individual tests) with @unittest.skipUnless(shutil.which('bcftools'), ...) so it cleanly skips when bcftools is unavailable.

class TestMergingPcgrFiles(unittest.TestCase):

TESTING.md:30

• This section says the only binary-dependent exception is TestMergeVcfFiles, but tests/test_pcgr.py::TestMergingPcgrFiles also depends on bcftools (it calls bcftools view/index). Update the wording so the documentation matches the actual test suite behavior (and the skip-guarding used for these integration tests).

Tests in `tests/` are pure Python + in-memory `cyvcf2` logic and must run
without any bioinformatics binary installed (no `bcftools`, `pcgr`, `cpsr`,
`vcfanno`, `snpEff`, `gpgr`, VEP). Functions that shell out to those tools are
either mocked/patched in tests or are not unit-tested (see below). The one
exception is `TestMergeVcfFiles`, an integration test that exercises the real
`merge_vcf_files` → `bcftools merge` path; it is guarded with
`@unittest.skipUnless(shutil.which('bcftools'), ...)`, so it runs in the conda
CI env and skips cleanly (never fails) where `bcftools` is absent.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.