New LSCO method optimizes protein expression using AI language models

Researchers have developed a novel method called Latent-Space Codon Optimization (LSCO) to improve the efficiency of protein expression. This technique recasts the discrete problem of codon selection into a continuous one by mapping sequences into the latent space of a pretrained mRNA language model, allowing for gradient-based optimization. LSCO integrates an expression predictor, a stability regularizer, a naturalness prior, and constrained decoding to ensure protein fidelity. In experiments with antibody expression data, LSCO demonstrated superior performance compared to existing heuristic and deep generative methods. AI

IMPACT This AI-driven approach could accelerate the development of therapeutic proteins and mRNA vaccines by improving expression efficiency.

RANK_REASON The cluster contains an academic paper detailing a new method for protein expression optimization. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• antibody
• arXiv
• Codon Adaptation Index
• Hugging Face
• Latent-Space Codon Optimization
• LSCO
• protein

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →