LipoAgent: Coordinating Fine-Tuned LLM Agents for Safer Lipid Design

2026-05-24

AI Summary

This study addresses the longstanding challenge in nucleic acid therapeutics wherein lipid-based delivery systems often struggle to balance high transfection efficiency with biocompatibility due to toxicity concerns. To overcome this, the authors propose LipoAgent—a safety-first, multi-agent large language model framework that innovatively embeds toxicity constraints directly into the prediction pipeline, treating toxicity assessment as a prerequisite for efficiency estimation. The approach integrates domain-specific fine-tuning, conditional prediction objectives, and a lightweight human-in-the-loop multi-agent verification mechanism. Evaluated across multiple base models, LipoAgent improves mRNA transfection efficiency prediction accuracy by an average of 32%, with wet-lab experiments confirming a strong correlation between its virtual screening rankings and actual transfection performance.

Abstract

Lipid nanoparticles (LNPs) are among the most clinically mature platforms for nucleic acid delivery, yet designing lipids that are both effective and biologically safe remains a major bottleneck. In practical screening, toxicity is a decision-level constraint: if a lipid is toxic, its efficiency prediction is clinically irrelevant. We propose LipoAgent, a safety-aware multi-agent LLM framework for lipid discovery. LipoAgent combines domain-specific finetuning with a conditional prediction objective that enforces toxicity as a prerequisite for efficiency prediction, and further improves reliability via multi-agent verification with lightweight human oversight when disagreement persists. Across multiple foundation models, LipoAgent achieves an average 32% relative improvement in mRNA transfection efficiency prediction compared with other reported models for lipid design. Wet-lab validation confirms that virtual screening rankings reliably translate to biological transfection outcomes. The code is publicly available at https://github.com/SAI-Lab-NYU/LipoAgent.git.

Problem

• lipid design
• toxicity
• nucleic acid delivery
• mRNA transfection
• biological safety

Innovation

• safety-aware LLM
• conditional prediction
• multi-agent verification
• lipid nanoparticle design
• toxicity-constrained optimization