wisp 0.1.0

What is wisp?

wisp is a command line tool for building population count masks: sparse, population-level summaries of how many individuals have sufficient depth and mapping quality to call genotypes at sites across the genome. These masks can be produced from BAM/CRAM alignments or an all-sites VCF.

chrom	start	end	GBR	YRI
chr20	9999999	10001042	7	4
chr20	10001042	10001881	10	6
chr20	10500000	10501200	3	0

Each row records how many samples in each population cleared the depth threshold over that interval. Intervals where all counts are zero are omitted.

Population count masks let you correctly compute the denominators of π, d_xy, Watterson's θ, and Tajima's D when working from a variants-only VCF — callable sites are counted per population rather than collapsed into a single cohort-wide pass/fail.

wisp is designed for use with pixy, but works equally well on its own. Source code is on GitHub.

The tool produces the same wisp.bed.gz output from two input modes:

• BAM/CRAM alignments, using mosdepth to quantize each sample and bedtools multiinter to combine samples. In this mode, an optional variants-only VCF can estimate omitted thresholds and exclude non-SNP variant spans from the final sparse mask.

• A prefiltered all-sites VCF, using per-sample FORMAT/DP values directly.

The output is bgzip-compressed and tabix-indexed. Large cohorts and large target regions stay compact because absent intervals are interpreted as zero passing samples in every population.

Documentation

• About
  • Why population count masks?
  • Input modes
  • Sparse output
• Installation
  • Requirements
  • Conda environment
  • Verify the CLI
  • Build these docs locally
• Arguments
  • Commands
  • Core arguments
  • Input-specific arguments
  • Shared optional arguments
  • BAM/CRAM mode options
  • Examples
• Input files
  • Sample table for BAM/CRAM mode
  • Popfile for all-sites VCF mode
  • Sample and population labels
  • All-sites VCF
  • Variants-only VCF for BAM/CRAM mode
  • Mask BED
• Examples
  • BAM/CRAM smoke run
  • Keep intermediate files
  • BAM/CRAM with a variants-only VCF
  • All-sites VCF run
  • Query the output
  • Overwrite existing outputs
• Output files
  • Final outputs
  • BED columns
  • Metadata
  • Sparse interpretation
  • Intermediate files

Reference

• Development
  • Set up a development environment
  • Run checks
  • Build documentation
  • Project layout
  • Implementation overview
• API Reference
  • Configuration
  • Workflow
  • Samples
  • Output summaries
• Changelog
  • 0.1.0