Understanding the Natural Language of DNA using Encoder-Decoder Foundation Models with Byte-level Precision

TL;DR

This paper introduces ENBED, a byte-level encoder-decoder Transformer model for DNA sequences, enabling precise analysis and generation of genomic data with improved accuracy over existing models.

Contribution

The paper develops ENBED, a novel byte-level Transformer model with efficient attention, capable of sequence-to-sequence genomic analysis and mutation generation, surpassing prior models.

Findings

ENBED outperforms state-of-the-art models in genomic tasks.

Byte-level analysis preserves detailed sequence information.

The model effectively generates and validates viral mutations.

Abstract

This paper presents the Ensemble Nucleotide Byte-level Encoder-Decoder (ENBED) foundation model, analyzing DNA sequences at byte-level precision with an encoder-decoder Transformer architecture. ENBED uses a sub-quadratic implementation of attention to develop an efficient model capable of sequence-to-sequence transformations, generalizing previous genomic models with encoder-only or decoder-only architectures. We use Masked Language Modeling to pre-train the foundation model using reference genome sequences and apply it in the following downstream tasks: (1) identification of enhancers, promotors and splice sites, (2) recognition of sequences containing base call mismatches and insertion/deletion errors, an advantage over tokenization schemes involving multiple base pairs, which lose the ability to analyze with byte-level precision, (3) identification of biological function annotations of genomic sequences, and (4) generating mutations of the Influenza virus using the encoder-decoder architecture and validating them against real-world observations. In each of these tasks, we demonstrate significant improvement as compared to the existing state-of-the-art results.