Credits
PLINK 2.0 beta was developed by Christopher Chang and Benjamin Demaille, with support from Human Longevity, Inc. in 2016-17, and substantial input from Stanford's Department of Biomedical Data Science.
- The .pgen file format is primarily an adaptation of ideas from SNPack (Francisco Sambo, Barbara Di Camillo, Gianna Toffolo, Claudio Cobelli) to more general contexts.
- The KING-robust kinship estimation method was developed by Ani Manichaikul et al. at the University of Virginia.
- The extended chrX Hardy-Weinberg equilibrium test was developed by Jan Graffelman (Universitat Politecnica de Catalunya) and Bruce Weir (University of Washington).
- Christoph Lippert (Human Longevity), Manuel Rivas (Stanford), and Yosuke Tanigawa (Stanford) contributed significantly to development and testing of the Python and R file I/O libraries.
- Firth logistic regression was added due to discussion with Manuel Rivas. The implementation is ported from Georg Heinze's logistf R package.
- The Zstandard compression format and libraries used by PLINK 2.0 were developed by Yann Collet and Przemysław Skibiński, with support from Meta Platforms, Inc.
- The fast PCA approximation is ported from Kevin Galinsky's contribution to EIGENSOFT 6.
- The array-popcount improvement is based on Faster Population Counts Using AVX2 Instructions by Wojciech Muła, Nathan Kurz, and Daniel Lemire, along with Kim Walisch's libpopcnt implementation.
- The bgzip-decompression improvement is based on libdeflate (Eric Biggers), and htslib 1.8 integration work by James Bonfield.
- --extract/--exclude for the bundled vcf_subset utility was implemented by Carsten Erickson.
- Thanks to numerous other testers at Human Longevity, Stanford, and elsewhere for bug reports and helpful suggestions.
File format reference >>
|