Genetic prediction of quantitative traits: a machine learner's guide focused on height
It addresses the problem of phenotype prediction for geneticists and machine learning practitioners, but is incremental as it synthesizes existing knowledge rather than introducing novel methods.
The paper provides a guide for the machine learning community on predicting complex traits from genetics, using height as an example, by reviewing current state-of-the-art models and addressing data subtleties without presenting new experimental results.
Machine learning and deep learning have been celebrating many successes in the application to biological problems, especially in the domain of protein folding. Another equally complex and important question has received relatively little attention by the machine learning community, namely the one of prediction of complex traits from genetics. Tackling this problem requires in-depth knowledge of the related genetics literature and awareness of various subtleties associated with genetic data. In this guide, we provide an overview for the machine learning community on current state of the art models and associated subtleties which need to be taken into consideration when developing new models for phenotype prediction. We use height as an example of a continuous-valued phenotype and provide an introduction to benchmark datasets, confounders, feature selection, and common metrics.