arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.13791cs.LGcs.AI
收藏

MechPert: 机制共识作为未见扰动预测的归纳偏置

MechPert: Mechanistic Consensus as an Inductive Bias for Unseen Perturbation Prediction

Marc Boubnovski Martell, Josefa Lia Stoisser, Lawrence Phillips, Aditya Misra, Robert Kitchen, Jesper Ferkinghoff-Borg, Jialin Yu, Philip Torr, Kaspar Märten

AI总结:

MechPert通过机制共识引导LLM生成定向调控假说,提升未见扰动预测的准确性与实验设计效果。

摘要:

预测未见的基因扰动的转录反应对于理解基因调控和优先考虑大规模扰动实验至关重要。现有方法要么依赖于静态的、可能不完整的知识图谱,要么提示语言模型寻找功能相似的基因,检索由科学文本中的对称共现塑造的关联,而非定向调控逻辑。我们引入了MechPert,一个轻量级框架,鼓励LLM代理生成定向调控假说,而非仅依赖功能相似性。多个代理独立提出候选调控因子并关联置信度评分;这些通过共识机制聚合,过滤虚假关联,产生加权邻域用于下游预测。我们在四个人类细胞系的Perturb-seq基准上评估了MechPert。在低数据量情况下(N=50观察到的扰动),MechPert将皮尔逊相关性提高了10.5%以上,超过基于相似性的基线。在实验设计中,MechPert选择的锚基因在特征明确的细胞系中,比标准网络中心性启发式方法高出46%。

英文摘要:

Predicting transcriptional responses to unseen genetic perturbations is essential for understanding gene regulation and prioritizing large-scale perturbation experiments. Existing approaches either rely on static, potentially incomplete knowledge graphs, or prompt language models for functionally similar genes, retrieving associations shaped by symmetric co-occurrence in scientific text rather than directed regulatory logic. We introduce MechPert, a lightweight framework that encourages LLM agents to generate directed regulatory hypotheses rather than relying solely on functional similarity. Multiple agents independently propose candidate regulators with associated confidence scores; these are aggregated through a consensus mechanism that filters spurious associations, producing weighted neighborhoods for downstream prediction. We evaluate MechPert on Perturb-seq benchmarks across four human cell lines. For perturbation prediction in low-data regimes ($N=50$ observed perturbations), MechPert improves Pearson correlation by up to 10.5\% over similarity-based baselines. For experimental design, MechPert-selected anchor genes outperform standard network centrality heuristics by up to 46\% in well-characterized cell lines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑