Geza Ephifania

Geza Ephifania

Authors and Affiliations Ephifania Geza1,2*, Nicola. Mulder2, Emile R. Chimusa3, and Gaston K. Mazandu1,2,3*

1Computational Biology Division, Department of Integrative Biomedical Sciences, University of Cape Town, Health Sciences Campus. Anzio Rd, Observatory, 7925, South Africa,

2African Institute for Mathematical Sciences Melrose Rd, Muizenberg, 7945, Cape Town, South Africa, and

3Division of Human genetics, Department of Pathology, University of Cape Town Health Sciences Campus. Anzio Rd, Observatory, 7925, South Africa

Content Aim

To integrate existing state-of-the-art local ancestry inference methods within a unified framework.


The inference of ancestry at every chromosomal site of individuals that resulted from the mixing of two or more genetically distinct populations (local ancestry inference) have found applications in several biomedical studies.  Such studies include understanding complex diseases, population history and demographics and personalizing medicines. Although several local ancestry inference methods exist, they are available as individual scripts requiring unique inputs and producing unique outputs. As a result, existing methods do not facilitate the local ancestry inference process and its applications. Thus, local ancestry inference is challenging to researchers with limited programming language.


We use python scripts, PLINK and EAGLE softwares to manipulate tool-specific inputs, deconvolve and standardize local ancestry inference results.


We introduce a unified framework for multi-way local ancestry inference, FRANC, integrating nine state-of-the-art local ancestry deconvolution tools. FRANC is adaptable, expandable and a portable tool that manipulates tool-specific inputs, deconvolves ancestry and standardizes tool-specific results.

Conclusions FRANC is a single piece of python portable application integrating nine state-of-the-art local ancestry deconvolution models filling the bioinformatics needs of facilitating the local ancestry estimate process in an admixed individual. It provides an easy-to-use and flexible tool, enabling different admixture scenario

analysis within the same framework and the assessment of local ancestry deconvolution models, helping users to select an appropriate tool based on data and application under consideration.


Share this post

Leave a Reply

Your email address will not be published. Required fields are marked *