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

arXiv:2311.02333v2 Announce Type: replace
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 …

abstract architecture arxiv attention cs.lg decoder dna encoder encoder-decoder ensemble foundation foundation model genomic implementation language natural natural language paper precision q-bio.gn transformer transformer architecture type understanding