ProtoCycle: Reflective Tool-Augmented Planning for Text-Guided Protein Design
Yutang Ge, Guojiang Zhao, Sihang Li, Zheng Cheng, Zifeng Zhao + 5 more
TLDR
ProtoCycle is an LLM-driven agentic framework for text-guided protein design that uses reflective, tool-augmented planning to bridge the plan-execute gap.
Key contributions
- Introduces ProtoCycle, an agentic framework for text-guided protein design using LLMs.
- Couples an LLM planner with a lightweight tool environment for iterative design.
- Employs LLM-driven reflection on tool feedback to revise and improve protein plans.
- Achieves strong language alignment and competitive foldability with limited supervision.
Why it matters
This paper addresses the challenge of designing proteins from natural language, a critical goal in protein engineering. ProtoCycle offers an efficient, data-light approach by leveraging LLMs for planning and reflection, significantly improving sequence quality.
Original Abstract
Designing proteins that satisfy natural language functional requirements is a central goal in protein engineering. A straightforward baseline is to fine-tune generic instruction-tuned LLMs as direct text-to-sequence generators, but this is data- and compute-hungry. With limited supervision, LLMs can produce coherent plans in text yet fail to reliably realize them as sequences. This plan-execute gap motivates ProtoCycle, an agentic framework for protein design that uses LLMs primarily to drive a multi-round, feedback-driven decision cycle. ProtoCycle couples an LLM planner with a lightweight tool environment designed to emulate the iterative workflow of human protein engineering and uses LLM-driven reflection on tool feedback to revise plans. Trained with supervised trajectories and online reinforcement learning, ProtoCycle achieves strong language alignment while maintaining competitive foldability, and ablations show that reflection substantially improves sequence quality.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.