DeepCDS: Ab initio coding sequence prediction in prokaryotic short nucleotide sequences
- 1University of Copenhagen
Abstract
Accurate coding sequence prediction in short prokaryotic metagenomic sequences, such as Illumina reads, remains challenging due to sequence fragmentation, unknown sequence origins, and sequencing errors. Here we introduce DeepCDS, a deep learning-based ab initio coding sequence predictor trained on short prokaryotic sequences with and without simulated Illumina-like sequencing errors. DeepCDS integrates ESM-2 protein language model embeddings with nucleotide-level information to predict complete and fragmented coding sequence regions. Benchmarking on 215 phylogenetically diverse prokaryotic organisms demonstrates that DeepCDS consistently outperforms current state-of-the-art methods in coding sequence detection across sequence lengths from 75bp to 1000bp, including improved identification of both canonical and non-canonical start codons as well as stop codons, and greater robustness to different sequencing error profiles. DeepCDS also remains operational on sequences shorter than existing tools support. We additionally evaluate DeepCDS on a simulated metagenomic dataset from the Critical Assessment of Metagenome Interpretation initiative, where it likewise achieves the strongest performance among tested tools. These findings demonstrate that protein language models capture distinct signals relevant for nucleotide-level coding sequence detection, especially at very short lengths. Ultimately, DeepCDS may help uncover the functional potential of the vast microbial diversity that remains genomically uncharacterized.
Related
- Intrafilament nucleotide exchange in a prokaryotic actin homolog
- Characterizing the ecological niche of insertion sequences within prokaryotic genomes
- Systematic mapping of the prokaryotic noncoding sequence space with a genomic language model
- TIGRdisco: a tool for the discovery of TIGR arrays and TIGR-associated Tas proteins from nucleotide sequences
- A linear-transformer hybrid for single-nucleotide polymorphism-based genotype-to-phenotype prediction in grapevine.
BibTeX
@misc{nielsen2026deepcds,
title = {{DeepCDS: Ab initio coding sequence prediction in prokaryotic short nucleotide sequences}},
author = {L. S. Nielsen and H. Nielsen and O. Winther},
howpublished = {bioRxiv},
year = {2026},
month = jun,
doi = {10.64898/2026.06.17.732633},
url = {https://www.biorxiv.org/content/10.64898/2026.06.17.732633v2},
note = {CC BY-ND},
}
RIS
TY - UNPB
TI - DeepCDS: Ab initio coding sequence prediction in prokaryotic short nucleotide sequences
AU - L. S. Nielsen
AU - H. Nielsen
AU - O. Winther
PY - 2026
DA - 2026/06/17
PB - bioRxiv
DO - 10.64898/2026.06.17.732633
UR - https://www.biorxiv.org/content/10.64898/2026.06.17.732633v2
AB - Accurate coding sequence prediction in short prokaryotic metagenomic sequences, such as Illumina reads, remains challenging due to sequence fragmentation, unknown sequence origins, and sequencing errors. Here we introduce DeepCDS, a deep learning-based ab initio coding sequence predictor trained on short prokaryotic sequences with and without simulated Illumina-like sequencing errors. DeepCDS integrates ESM-2 protein language model embeddings with nucleotide-level information to predict complete and fragmented coding sequence regions. Benchmarking on 215 phylogenetically diverse prokaryotic organisms demonstrates that DeepCDS consistently outperforms current state-of-the-art methods in coding sequence detection across sequence lengths from 75bp to 1000bp, including improved identification of both canonical and non-canonical start codons as well as stop codons, and greater robustness to different sequencing error profiles. DeepCDS also remains operational on sequences shorter than existing tools support. We additionally evaluate DeepCDS on a simulated metagenomic dataset from the Critical Assessment of Metagenome Interpretation initiative, where it likewise achieves the strongest performance among tested tools. These findings demonstrate that protein language models capture distinct signals relevant for nucleotide-level coding sequence detection, especially at very short lengths. Ultimately, DeepCDS may help uncover the functional potential of the vast microbial diversity that remains genomically uncharacterized.
ER -