CodonTransformer Dataset

A comprehensive compilation of 1,001,197 DNA and protein sequence pairs, sourced from 164 organisms across Eukaryotes, Bacteria, and Archaea. This dataset provides a rich resource for various computational biology and bioinformatics applications such as studying gene sequences, codon usage, and protein expression across diverse species.

Dataset Contents

• 1,001,197 DNA-protein sequence pairs
• Sequences from 164 organisms, including:
  • Eukaryotes: Homo sapiens, Arabidopsis thaliana, Caenorhabditis elegans, Danio rerio, Drosophila melanogaster, Mus musculus, Saccharomyces cerevisiae, Chlamydomonas reinhardtii, Nicotiana tabacum
  • Bacteria: Various Enterobacteriaceae species including Escherichia coli
  • Archaea: Thermococcus barophilus, Sulfolobus solfataricus
  • Chloroplast genomes: Chlamydomonas reinhardtii, Nicotiana tabacum

Data Collection and Preprocessing

• Source: NCBI resources
• Original Format: Gene or CDS (Coding Sequence)
• Protein Sequences: Translated using NCBI Codon Tables
• Quality Control:
  • DNA sequences divisible by three in length
  • Start with a start codon
  • End with a single stop codon

Dataset Structure

Each entry contains:

• DNA sequence
• Corresponding protein sequence
• Gene and organism information

Uses and Applications

This dataset is valuable for various research areas and applications, including:

• Comparative genomics
• Codon usage analysis
• Protein expression optimization
• Synthetic biology and genetic engineering
• Machine learning models in bioinformatics

It has been used to train the CodonTransformer model for codon optimization tasks.

Authors

Adibvafa Fallahpour^1,2*, Vincent Gureghian³*, Guillaume J. Filion²‡, Ariel B. Lindner³‡, Amir Pandi³‡

¹ Vector Institute for Artificial Intelligence, Toronto ON, Canada
² University of Toronto Scarborough; Department of Biological Science; Scarborough ON, Canada
³ Université Paris Cité, INSERM U1284, Center for Research and Interdisciplinarity, F-75006 Paris, France
* These authors contributed equally to this work.
‡ To whom correspondence should be addressed:
guillaume.filion@utoronto.ca, ariel.lindner@inserm.fr, amir.pandi@cri-paris.org

Additional Resources

• Project Website
  https://adibvafa.github.io/CodonTransformer/

• GitHub Repository
  https://github.com/Adibvafa/CodonTransformer

• Google Colab Demo
  https://adibvafa.github.io/CodonTransformer/GoogleColab

• PyPI Package
  https://pypi.org/project/CodonTransformer/

• Paper
  https://www.nature.com/articles/s41467-025-58588-7

Citation

@article{Fallahpour_Gureghian_Filion_Lindner_Pandi_2025,
  title={CodonTransformer: a multispecies codon optimizer using context-aware neural networks},
  volume={16},
  ISSN={2041-1723},
  url={https://www.nature.com/articles/s41467-025-58588-7},
  DOI={10.1038/s41467-025-58588-7},
  number={1},
  journal={Nature Communications},
  author={Fallahpour, Adibvafa and Gureghian, Vincent and Filion, Guillaume J. and Lindner, Ariel B. and Pandi, Amir},
  year={2025},
  month=apr,
  pages={3205},
  language={en}
}