pixy 2.0.0

What is pixy?

pixy is a command-line tool for painlessly computing unbiased estimators of population genetic summary statistics that measure genetic variation within (π, θ_W, Tajima's D) and between (d_xy, F_ST) populations from a VCF.

Many tools for computing these summary statistics from VCFs produce biased estimates in the presence of missing data. This is because they often make the simplifying assumption that if a genotype is missing, it is homozygous reference (0/0) by state. See the pixy paper and the Watterson's θ / Tajima's D follow-up paper for the details.

As of version 2.0, pixy also supports organisms of arbitrary ploidy and VCFs with variable ploidy across contigs (e.g. diploid autosomes plus haploid sex chromosomes in one file — no need to split the VCF), multiallelic sites, and both .tbi and .csi VCF indexes.

Documentation

• About pixy
  • pixy avoids common pitfalls in computing pi and dxy
  • Notable features of pixy
• Installation
  • Via conda
  • From source
• Arguments
  • Required
  • In addition, one of
  • Optional
  • Example
• Companion files
  • Populations file
  • BED file
  • Sites file
• Contributing
  • Prerequisites
  • Installing the development version
  • Project layout
  • Primary development commands
  • Shortcut task commands
  • Working on the documentation
  • Code style
  • Continuous integration
  • Reporting bugs and requesting features
  • Pull request workflow
  • Commit messages
  • Code of conduct
  • License
  • Release process (for maintainers)
• Changelog
  • pixy 2.0.0
    • New features
    • Bug fixes
    • Project / packaging
  • pixy 1.0.0
    • Summary of major changes
    • Detailed changelog
      • Major changes
      • New features
      • Removed features
      • Minor updates
      • Other bugfixes

Guides

• Generating invariant sites VCFs
  • Generating AllSites VCFs using BCFtools (mpileup/call)
  • Generating AllSites VCFs using GATK
• Step by Step Installation and Usage
  • 1. Generate a VCF with Invariant Sites and perform filtering
    • Site-level filtration
    • Optional: Population genetic filters
  • 2. Install a conda distribution
  • 3. Create a New Environment
  • 4. Install pixy
  • 5. Create a populations file
  • 6. Compress and Index your VCF
  • 7. Run pixy
  • 8. Interpret your results
  • 9. Stay up to date
• Usage Examples
  • Basic usage
  • Including the new 2.0 statistics
  • Focusing on a specific genomic interval
  • Using a .BED file to define windows
  • Using a sites file to limit calculations to specific sites
  • Including multiallelic SNPs
  • Switching the F_ST estimator
  • Site-level estimates
• Example data
  • Download
  • How the data were generated
  • Running pixy on the example
    • Expected results
    • Variants of the same data
• Understanding pixy output
  • Output file contents
    • Within-population nucleotide diversity (pi)
    • Between-population nucleotide divergence (dxy)
    • F_ST (fst)
    • Watterson's θ (watterson_theta)
    • Tajima's D (tajima_d)
  • Working with pixy output data
    • Post-hoc aggregating
      • Pi (pi)
      • Dxy (dxy)
      • FST (fst)
      • Watterson's θ (watterson_theta)
      • Tajima's D (tajima_d)
• Plotting pixy output
  • Converting pixy output files to long format
  • Single chromosome view
  • A genome-wide plot of summary statistics

How should I cite pixy?

If you use pixy in your research, please cite the manuscript below, as well the Zenodo DOI of specific version of pixy used for your project.

Manuscript: Korunes, K.L. and Samuk, K. (2021), pixy: Unbiased estimation of nucleotide diversity and divergence in the presence of missing data. Molecular Ecology Resources. Accepted Author Manuscript. https://doi.org/10.1111/1755-0998.13326

And, if using the unbiased estimator of Tajima's D or Watterson's theta:

Bailey, N., Stevison, L., & Samuk, K. (2025). Correcting for bias in estimates of θw and Tajima’s D from missing data in next-generation sequencing. Molecular Ecology Resources, e14104. https://doi.org/10.1111/1755-0998.14104

Zenodo DOI for various versions of pixy: Go to https://zenodo.org/record/4432294 and find the DOI that matches the version used (the current version is shown first).