bap

Testing `bap`

This directory holds the automated test suite (pytest) and the small data fixtures it runs against.

Quick start

Install the package with its test dependencies, then run the fast suite:

pip install -e .[test]

# Fast: unit + CLI smoke tests only (no external tools needed)
pytest -m "not integration"

The fast suite requires no bioinformatics tools and runs in under a second. It is what CI runs on every push (see .github/workflows/ci.yml).

What each test file covers

File	Scope	External tools
`test_helpers.py`	Unit tests for the pure functions in `bap/bapHelp.py` (sequence math, file helpers, tool discovery, the `run_cmd` subprocess wrapper) and `mitoChr`.	none
`test_cli_smoke.py`	In-process `click.testing.CliRunner` checks that every active CLI imports, renders `--help`, and reports `--version`; verifies `bap2 support` lists built-in genomes.	none
`test_integration.py`	End-to-end `bap2 bam` run against a bundled hg19 `.bam`, asserting the final `.bap.bam`, `.barcodeTranslate.tsv`, and `.fragments.tsv.gz` are produced.	samtools, bedtools, R, snakemake

Shared fixtures (paths to the bundled data, small_bam) live in conftest.py.

Running the integration tests

Integration tests are marked with @pytest.mark.integration and skip automatically when samtools, bedtools, R, or snakemake are not on PATH, so the default pytest run stays green on a bare machine.

To run them, install the external tools (plus the R packages the pipeline uses: Rsamtools, GenomicAlignments, GenomicRanges, dplyr, data.table) and then:

# Run everything, including integration tests
pytest

# Run only the integration tests
pytest -m integration

A convenient way to get the external tools is conda/mamba:

mamba install -c bioconda -c conda-forge samtools bedtools snakemake bioconductor-genomicalignments bioconductor-rsamtools r-dplyr r-data.table

Test data

data/ — small indexed .bam files (jaccardPairsForIGV.bam, small_mix.bam, test.small.bam), a peaks bed, a whitelist, and a barcode prior table used by the pipeline tests.
fastq_br/ — small BioRad and Scale-ATAC FASTQs for barcode-parsing tests.
for_frag/ — pre-computed annotated fragment files used to exercise the fragment / adjacent-Tn5 code paths.

Manual smoke tests

The commands below are handy for exercising the full pipeline by hand against the bundled data (run from this tests/ directory). They require the external tools listed above.

Basic run

bap2 bam -i data/jaccardPairsForIGV.bam -bt XB -r hg19 -z -o bap2

Note: the legacy bap (v1) command is deprecated — use bap2.

Species-mix / peaks-file output

bap2 bam -i data/small_mix.bam -bt XB -ji 0.0001 -r hg19-mm10 -z --mapq 0 -bf 10 -o SM

Skip merging when a prior is known

bap2 bam -i data/jaccardPairsForIGV.bam -bt XB -r hg19 -z -o bap2 \
  -pf data/test.small.peaks.bed -bp data/jaccardPairsTest_sep.tsv

BioRad barcode parsing

bap-barcode v2.1 -a fastq_br/biorad_v2_R1.fastq.gz -b fastq_br/biorad_v2_R2.fastq.gz -o test

Scale-ATAC barcode parsing

bap-scale -f fastq_br/scale -s ScaleTest -o ScalePro

This site is open source. Improve this page.

bap

Testing bap