Deep Generative Models for Discrete Genotype Simulation

TL;DR

This paper evaluates deep generative models like VAEs, Diffusion Models, and GANs for simulating discrete genotype data, demonstrating their ability to capture genetic patterns and genotype-phenotype associations.

Contribution

It introduces adaptations of these models specifically for discrete genotype data and provides a comprehensive comparison and practical guidelines for future research.

Findings

Models effectively capture genetic patterns

Preserve genotype-phenotype associations

Offer practical guidelines for genotype simulation

Abstract

Deep generative models open new avenues for simulating realistic genomic data while preserving privacy and addressing data accessibility constraints. While previous studies have primarily focused on generating gene expression or haplotype data, this study explores generating genotype data in both unconditioned and phenotype-conditioned settings, which is inherently more challenging due to the discrete nature of genotype data. In this work, we developed and evaluated commonly used generative models, including Variational Autoencoders (VAEs), Diffusion Models, and Generative Adversarial Networks (GANs), and proposed adaptation tailored to discrete genotype data. We conducted extensive experiments on large-scale datasets, including all chromosomes from cow and multiple chromosomes from human. Model performance was assessed using a well-established set of metrics drawn from both deep learning and quantitative genetics literature. Our results show that these models can effectively capture genetic patterns and preserve genotype-phenotype association. Our findings provide a comprehensive comparison of these models and offer practical guidelines for future research in genotype simulation. We have made our code publicly available at https://github.com/SihanXXX/DiscreteGenoGen.