DiscDiff: Latent Diffusion Model for DNA Sequence Generation

TL;DR

This paper presents DiscDiff, a latent diffusion model for generating realistic DNA sequences, enhanced by the Absorb-Escape algorithm, and introduces a new multi-species DNA dataset to advance genetic sequence modeling.

Contribution

The paper introduces DiscDiff, a novel latent diffusion framework for DNA generation, and Absorb-Escape, a post-training correction method, along with a comprehensive multi-species DNA dataset.

Findings

DiscDiff outperforms existing diffusion models in DNA sequence generation.

Absorb-Escape improves sequence realism by correcting conversion errors.

The new dataset includes 160,000 sequences from 15 species.

Abstract

This paper introduces a novel framework for DNA sequence generation, comprising two key components: DiscDiff, a Latent Diffusion Model (LDM) tailored for generating discrete DNA sequences, and Absorb-Escape, a post-training algorithm designed to refine these sequences. Absorb-Escape enhances the realism of the generated sequences by correcting `round errors' inherent in the conversion process between latent and input spaces. Our approach not only sets new standards in DNA sequence generation but also demonstrates superior performance over existing diffusion models, in generating both short and long DNA sequences. Additionally, we introduce EPD-GenDNA, the first comprehensive, multi-species dataset for DNA generation, encompassing 160,000 unique sequences from 15 species. We hope this study will advance the generative modelling of DNA, with potential implications for gene therapy and protein production.