What is mStruct?
mStruct is a software for detection of population structure in the presence of admixing and mutations from multi-locus genotype data. It is a mixed membership model (also referred to as an admixture model) which incorporates a mutation process on the observed genetic markers. Our model is an extension of the Structure model by Pritchard et al (2000).
A sample result on world human populations.
What is it useful for?
mStruct is useful for detecting population structure from genotype data, checking for presence of admixture, identifying migrants, etc. At the same time, the results from mStruct also contain mutation information about the various populations that can provide various insights about the population, such as genetic diveristy at loci, age of the alleles etc.
The figure above shows the result obtained by running mStruct on the HGDP data. Like Structure, mStruct also produces a population structure map. Moreover, we can also find the average mutation rate for each of ancestral population as well as each modern population.
Citation: S. Shringarpure and E. P. Xing, mStruct: A New Admixture Model for Inference of Population Structure in Light of Both Genetic Admixing and Allele Mutations, Proceedings of the 25th International Conference on Machine Learning (ICML 2008).