Codon Transformer

Overview

Codon optimisation of the heavy and light chains allows you to increase the efficiency of mRNA and overcome protein limitations. It is essential to enhancing protein expression efficiency by exploiting the genetic code's redundancy. (https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2024.1371596/full

Inputs

• Path to FASTA file with protein sequence to optimise

• Organism to optimise for - E coli, homo sapien, or mus musculus

Outputs

• FASTA file with optimised DNA

• Summary report containing key statistics like GC content, sequence complexity, and DNA length

• FASTA file containing codon optimised sequence to download

Description

For EMLy to achieve this, it sends the input parameters to a remote CodonTransformer FastAPI service which returns a JSON payload containing the optimised sequence.

CodonTransformer is used to generate DNA sequences that replicate the natural codon distribution profiles, with minimum cisregulatory element. It captures intrinsic sequence patterns by harnessing the capacity of transformer architecture. It works across multiple species, learning codon usage patters across organisms to design host specific DNA sequences. CodonTransformer is a comprehensive Python package that streamlines the entire workflow, including data set processing, model traning, and sequence evaluation. (https://www.nature.com/articles/s41467-025-58588-7)