Inferring the heritability of bacterial traits in the era of machine learning

Quantification of heritability is a fundamental aim in genetics, providing answer to the question of how much genetic variation influences variation in a particular trait of interest. The traditional computational approaches for assessing the heritability of a trait have been developed in the field of quantitative genetics. However, modern sequencing methods have provided us with whole genome sequences from large populations, often together with rich phenotypic data, and this increase in data scale has led to the development of several new machine learning based approaches to inferring heritability. In this review, we systematically summarize recent advances in machine learning which can be used to perform heritability inference. We focus on bacterial genomes where heritability plays a key role in understanding phenotypes such as drug resistance and virulence, which are particularly important due to the rising frequency of antimicrobial resistance. Specifically, we present applications of these newer machine learning methods to estimate the heritability of antibiotic resistance phenotypes in several pathogens. This study presents lessons and insights for future research when using machine learning methods in heritability inference.