AlphaImpute2 is a phasing and imputation algorithm for massive livestock populations. The method uses a approximate version of multi-locus iterative peeling for pedigree based imputation, and a novel imputation algorithm that uses the Positional Burrows Wheeler Transform for population imputation. AlphaImpute2 has been successfully used to perform imputation in populations of hundreds of thousands of individuals. AlphaImpute2 was developed by Andrew Whalen, and is currently being supported by Andrew Whalen and Steve Thorn.
The AlphaImpute2.zip file contains a python3 wheel file, a manual, and an example dataset.
AlphaImpute2 is part of a suite of software that our group has developed. It is fully and freely available for all use under the MIT License.
Whalen, A, J.M. Hickey, (2020), AlphaImpute2: Fast and accurate pedigree and population based imputation for hundreds of thousands of individuals in livestock populations, bioRxiv.https://doi.org/10.1101/2020.09.16.299677