puretabix 5.3.0

Pure python implementation Tabix reader.

Pure Tabix

This is a pure-python Tabix index parser. Useful as an alternative to PySAM and PyTabix for rapid read access by position to Tabix indexed block gzipped files such as VCFs and other common bioinfomatics formats.

See https://samtools.github.io/hts-specs/tabix.pdf and https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3042176 for information about Tabix and the detailed file format specification.

from puretabix import TabixIndexedFile

tabix_indexed_file = TabixIndexedFile.from_files(open('somefile.vcf.gz', 'rb'), open('somefile.vcf.gz.tbi', 'rb'))
tabix_indexed_file.fetch("1", 1000, 5000)

Documentation is supported via Python built-in module PyDoc: python3 -m pydoc -b puretabix

VCF

Included in this package is tooling for reading and writing VCF lines.

To read a file:

from puretabix.vcf import read_vcf_lines

with open("source.vcf") as input:
    for vcfline in read_vcf_lines(input):
        if vcfline.is_comment:
            # its a comment or meta-information
            pass
        else:
            # access the parsed information
            if "PASS" not in vcfline._filter:
                print(f"{vcfline.chrom} {vcfline.pos} {vcfline.get_genotype()}")

To write some lines:

from puretabix.vcf import VCFLine

with open("output.vcf") as output:
    output.write(str(VCFLine.as_comment_key_dict("fileformat", "VCFv4.2")))
    output.write("\n")
    output.write(
        str(
            VCFLine.as_comment_raw(
                "\t".join(
                    (
                        "CHROM",
                        "POS",
                        "ID",
                        "REF",
                        "ALT",
                        "QUAL",
                        "FILTER",
                        "INFO",
                        "FORMAT",
                        "SAMPLE",
                    )
                )
            )
        )
    )
    output.write("\n")
    output.write(
        str(
            VCFLine.as_data(
                "chr1",
                123,
                ("rs123",),
                "A",
                ("C",),
                ".",
                ("PASS",),
                {},
                ({"GT": "1/0"},),
            )
        )
    )
    output.write("\n")

VCF with index

If there is a tabix index for a block gzipped VCF file, that index can be used for fast random access

import puretabix

with open("input.vcf.gz", "rb") as vcf:
    with open("input.vcf.gz.tbi", "rb") as vcf_tbi:
        indexed = puretabix.TabixIndexedVCFFile.from_files(vcf, vcf_tbi)
        vcfline = tuple(indexed.fetch_vcf_lines("chr1", 1108138))
        assert vcfline.chrom == "chr1"
        assert vcfline.pos == 1108138
        print(f"gt = {vcfline.get_genotype()}")

development

TL;DR: pip install -e '.[dev]' && pre-commit install

pip install -e '.[dev]'  # Install using pip including development extras
pre-commit install  # Enable pre-commit hooks
pre-commit run --all-files  # Run pre-commit hooks without committing
# Note pre-commit is configured to use:
# - seed-isort-config to better categorise third party imports
# - isort to sort imports
# - black to format code
pip-compile  # Freeze dependencies
pytest  # Run tests
coverage run --source=puretabix -m pytest && coverage report -m  # Run tests, print coverage
mypy .  # Type checking
pipdeptree  # Print dependencies
scalene --outfile tests/perf_test.txt --profile-all --cpu-sampling-rate 0.0001 tests/perf_test.py  # performance measurements

Global git ignores per https://help.github.com/en/github/using-git/ignoring-files#configuring-ignored-files-for-all-repositories-on-your-computer

For release to PyPI see https://packaging.python.org/tutorials/packaging-projects/

git checkout master
git pull
git add setup.py CHANGES.txt
git commit -m"prepare for x.x.x"
git push
git tag x.x.x
git push origin x.x.x
python3 setup.py sdist && python3 -m twine upload dist/*

acknowledgements

Inspired by @yangmqglobe code in https://github.com/cggh/scikit-allel/pull/297