扰动预测器是否学到基因互作？诊断agentic筛选中的外推
原标题：Do Perturbation Predictors Learn Gene Interactions? Diagnosing Extrapolation in Agentic Screening
AI 摘要
该研究针对高通量扰动筛选中的组合泛化问题，提出用「加性分歧」（AD）诊断模型对未见基因组合的预测可靠性。作者在 Norman、Joung–Zhang 和 Adamson 三个 Perturb-seq 数据集上比较了八种预测方法，发现专用神经预测器并不稳定优于简单加性基线，且偏离加性基线的预测误差更大。相比 kNN、马氏距离、集成不确定性等通用信号，AD 在组合分布偏移下更一致，可用于指导 agentic 扰动筛选工作流。
正文节选
NeurIPS 2026 Workshop on Agentic Systems for Molecular Sciences Do Perturbation Predictors Learn Gene Interactions? Diagnosing Extrapolation in Agentic Screening Abstract High-throughput perturbation screens increasingly use machine-learning models to predict the effects of gene combinations that were never measured. A difficult case arises when both genes in a test pair were seen individually during training, but their combination was not. In this setting, standard OOD methods often fail becaus