Equi-mRNA: Protein Translation Equivariant Encoding for mRNA Language Models

2025-08-20

AI Summary

Synonymous codon usage encodes intrinsic symmetries influencing translational efficiency and gene expression, yet existing mRNA language models lack structured modeling of these symmetries. Method: We propose the first codon-level equivariant neural network, modeling synonymous codon sets as cyclic subgroups of SO(2), and integrate group-theoretic priors, equivariant attention, symmetry-aware pooling, and auxiliary equivariant loss to enable biologically interpretable representation learning. Results: Our model achieves up to 10% absolute improvement in accuracy on expression-level prediction, mRNA stability estimation, and riboswitch identification. Generated sequences exhibit ~4× higher fidelity (measured by k-mer distribution similarity) and 28% improved functional retention. This work establishes the first modeling paradigm for mRNA sequence analysis with explicit symmetry-inductive bias, advancing mRNA therapeutics and synthetic biology.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

The growing importance of mRNA therapeutics and synthetic biology highlights the need for models that capture the latent structure of synonymous codon (different triplets encoding the same amino acid) usage, which subtly modulates translation efficiency and gene expression. While recent efforts incorporate codon-level inductive biases through auxiliary objectives, they often fall short of explicitly modeling the structured relationships that arise from the genetic code's inherent symmetries. We introduce Equi-mRNA, the first codon-level equivariant mRNA language model that explicitly encodes synonymous codon symmetries as cyclic subgroups of 2D Special Orthogonal matrix (SO(2)). By combining group-theoretic priors with an auxiliary equivariance loss and symmetry-aware pooling, Equi-mRNA learns biologically grounded representations that outperform vanilla baselines across multiple axes. On downstream property-prediction tasks including expression, stability, and riboswitch switching Equi-mRNA delivers up to approximately 10% improvements in accuracy. In sequence generation, it produces mRNA constructs that are up to approximately 4x more realistic under Frechet BioDistance metrics and approximately 28% better preserve functional properties compared to vanilla baseline. Interpretability analyses further reveal that learned codon-rotation distributions recapitulate known GC-content biases and tRNA abundance patterns, offering novel insights into codon usage. Equi-mRNA establishes a new biologically principled paradigm for mRNA modeling, with significant implications for the design of next-generation therapeutics.

Problem

• Modeling synonymous codon usage for mRNA translation efficiency
• Encoding genetic code symmetries with equivariant language models
• Improving mRNA property prediction and sequence generation accuracy

Innovation

• Equivariant encoding for codon symmetries
• Group-theoretic priors with auxiliary loss
• Symmetry-aware pooling for mRNA representations