Rethinking Genomic Modeling Through Optical Character Recognition

TL;DR

OpticalDNA introduces a vision-based OCR-style approach to genomic modeling, enabling more efficient and detailed understanding of DNA sequences by rendering them into visual layouts and training a specialized vision-language model.

Contribution

This work pioneers a visual OCR-inspired framework for genomic modeling, moving beyond traditional sequential token methods to improve efficiency and information retention.

Findings

Outperforms recent baselines across diverse genomic benchmarks.

Achieves nearly 20x fewer effective tokens on sequences up to 450k bases.

Uses only 256k trainable parameters to surpass models with up to 985x more parameters.

Abstract

Recent genomic foundation models largely adopt large language model architectures that treat DNA as a one-dimensional token sequence. However, exhaustive sequential reading is structurally misaligned with sparse and discontinuous genomic semantics, leading to wasted computation on low-information background and preventing understanding-driven compression for long contexts. Here, we present OpticalDNA, a vision-based framework that reframes genomic modeling as Optical Character Recognition (OCR)-style document understanding. OpticalDNA renders DNA into structured visual layouts and trains an OCR-capable vision--language model with a \emph{visual DNA encoder} and a \emph{document decoder}, where the encoder produces compact, reconstructible visual tokens for high-fidelity compression. Building on this representation, OpticalDNA defines prompt-conditioned objectives over core genomic primitives-reading, region grounding, subsequence retrieval, and masked span completion-thereby learning layout-aware DNA representations that retain fine-grained genomic information under a reduced effective token budget. Across diverse genomic benchmarks, OpticalDNA consistently outperforms recent baselines; on sequences up to 450k bases, it achieves the best overall performance with nearly $20\times$ fewer effective tokens, and surpasses models with up to $985\times$ more activated parameters while tuning only 256k \emph{trainable} parameters.