HAPNEST: An efficient tool for generating large-scale genetics datasets from limited training data

Sophie Wharrie, Zhiyu Yang, Vishnu Raj, Remo Monti, Rahul Gupta, Ying Wang, Alicia Martin, Luke J O'Connor, Samuel Kaski, Pekka Marttinen, Pier Palamara, Christoph Lippert, Andrea Ganna

Keywords: synthetic data, ML for healthcare, computational genetics, approximate Bayesian computation

TL;DR: HAPNEST is a new, efficient software tool for generating and evaluating large synthetic datasets for human genetics applications

Abstract: In this extended abstract we present a new highly efficient software tool called HAPNEST that enables machine learning practitioners to easily generate and evaluate large synthetic datasets for human genetics applications. HAPNEST enables the generation of diverse synthetic datasets from small, publicly accessible reference datasets. We demonstrate the suitability of HAPNEST-generated data for supervised tasks such as genetic risk scoring.