通过功能、证据和验证评估智能体生物信息学

中文摘要 AI 辅助

大型语言模型智能体越来越多地规划、执行和解释生物学分析，但流畅的响应、成功的工具调用和基准性能本身并不能确立科学可信度。现有综述主要按应用、架构和智能体能力对生物智能体进行组织，但未共同将智能体生成工作流的可问责性操作化。我们通过将可检查的工作流轨迹而非仅架构或最终输出作为主要分析单元来解决这一缺口，提出功能-证据-验证（FEV）框架，该框架将已展示的工作流操作、对行动和主张的可追溯支持以及用例特定验证分离开来。使用FEV，我们梳理了109个智能体或类智能体系统以及28个基准或评估资源，这些资源代表了基因组学、单细胞与空间组学、蛋白质科学、药物发现、计算病理学及通用生物信息学自动化领域的128篇独特出版物。跨领域来看，规划和工具介导执行的进展快于可重放性、来源追溯、稳健科学评估、外部验证和前瞻性实证测试。因此，我们主张智能体生物信息学应仅通过工作流正确性而非仅最终答案正确性进行评估。FEV为比较系统以及设计透明、可审计和科学可问责的生物信息学工作流提供了实用基础。

英文摘要

Large language model agents increasingly plan, execute, and interpret biological analyses, yet fluent responses, successful tool calls, and benchmark performance alone do not establish scientific credibility. Existing reviews primarily organize biological agents by application, architecture, and agentic capability, but do not jointly operationalize the accountability of agent-generated workflows. We address this gap by treating the inspectable workflow trajectory, rather than architecture or final output alone, as the primary unit of analysis. We introduce the Function--Evidence--Validation (FEV) framework, which separates demonstrated workflow operations, traceable support for actions and claims, and use-case-specific validation. Using FEV, we map 109 agentic or agent-adjacent systems and 28 benchmark or evaluation resources, representing 128 unique publications across genomics, single-cell and spatial omics, protein science, drug discovery, computational pathology, and general bioinformatics automation. Across domains, planning and tool-mediated execution have advanced more rapidly than replayability, provenance, robust scientific assessment, external validation, and prospective empirical testing. We therefore argue that agentic bioinformatics should be assessed through workflow correctness rather than final-answer correctness alone. FEV provides a practical basis for comparing systems and designing transparent, auditable, and scientifically accountable bioinformatics workflows.

发表机构

• The University of Alabama at Birmingham（阿拉巴马大学伯明翰分校）

机构由 AI 辅助整理，请以论文原文为准。