Skip to content

LUMC/array-as-vcf

Repository files navigation

Array As VCF

array-as-vcf is a small library and tool to convert common SNP array formats to VCF format.

There are four currently supported array formats:

  • Affymetrix (TSV export)
  • Cytoscan HD Array (TSV export)
  • Lumi 317k array (TSV export)
  • Lumi 370k array (TSV export)
  • Multi-sample OpenArray (TSV export)

Binary formats are not (yet) supported.

Requirements

  • Python 3.6
  • requests

CLI usage

The array-as-vcf tool will convert array files to VCF format. It will auto-detect the type of array file, and throw an error if it can't determine it.

The generated VCF file is printed to stdout.

A sample name to be used in the VCF file must be supplied.

The REF and ALT alleles will be queried from Ensembl if no lookup-table is supplied. This requires a working internet connection, and can be quite slow due the amount of HTTP requests that are necessary.

When supplied with lookup-table, no requests are made for the rsIDs which exist within the lookup table. The lookup table is a JSON file, containing a single large object of shape:

{
  "rs0": "{ref_allele}:{alt_alleles}:{ref_is_minor_allele}"
}

E.g.

{
  "rs1000003": "A:G:F"
}

If you have never run array-as-vcf before , you can run array-as-vcf sans lookup table and dump the generated internal lookup table to a file for next iterations.

Usage: array-as-vcf [OPTIONS]

Options:
  -p, --path PATH              Path to array file  [required]
  -b, --build [GRCh37|GRCh38]
  -s, --sample-name TEXT       Name of sample in VCF file
  -c, --chr-prefix TEXT        Optional prefix to chromosome names
  -l, --lookup-table PATH      Optional path to existing lookup table for
                               rsIDs.
  -d, --dump PATH              Optional path to write generated lookup table
  --help                       Show this message and exit.