Researchers develop ProMORNA for de novo mRNA design from protein sequences

Researchers have developed ProMORNA, a novel framework for designing therapeutic messenger RNA (mRNA) sequences. This system uses a BART-style encoder-decoder model trained on millions of protein-mRNA pairs and employs multi-objective reinforcement learning to optimize for stability, translation efficiency, and immune safety simultaneously. ProMORNA demonstrated improved performance in silico for predicting half-life and translation efficiency on an unseen target, outperforming existing supervised methods. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Introduces a new ML-driven approach for designing complex biological molecules, potentially accelerating therapeutic development.

RANK_REASON Academic paper detailing a new method for mRNA design using machine learning. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• ProMORNA
• BART
• firefly luciferase

• paper
• other