arXiv q-bio.MN 分子生物学预印本· Judith Bernett, Anton Spannagl, Joel {\AA}s, Markus List, David B. Blumenthal·· 22 小时前AI 评分48
蛋白质相互作用数据集偏倚审计:区分生物信号与学习捷径的系统方法
Are You Learning Biological Signal or Shortcuts? Auditing and Mitigating Bias in Protein-Protein Interaction Datasets
AI 导读
该预印本研究指出PPI数据库受技术与研究偏差影响,机器学习模型可能利用分布捷径而非生物信号进行预测。文章系统性刻画了HIPPIE、IntAct、STRING及两个PDB衍生数据集中的已有和未知偏倚,发现随机数据分割引入强拓扑捷径;移除训练-测试蛋白重叠后仍残留源自自相互作用、分类学身份及功能相关性的可乘偏倚;从高度重复非相互作用中采样负样本会放大功能相关性偏倚。
正文
Abstract:Protein-protein interaction (PPI) databases do not faithfully reflect biological realities. Instead, they are influenced by study and technical biases that distort certain protein and interaction attributes. Machine learning models can exploit these as learning shortcuts if the negative dataset is not constructed with care. So far, the shortcuts introduced during PPI dataset construction have only been examined in isolation. Here, we systematically characterize both reported and, to our knowledge, previously unreported biases in PPI datasets that lead machine learning models to learn shortcuts instead of biological signal. We analyze HIPPIE, IntAct, and STRING, dedicated PPI databases, as well as two datasets derived from 3D-structural information in the Protein Data Bank (PDB). We show that random data splitting introduces strong topological shortcuts. When train-test protein overlap is removed, the resulting datasets still retain usable shortcuts stemming from self-interactions, taxonomic identity, and functional relatedness, whose prevalence interestingly depends on the data source. We further show that sampling negatives from a set of highly replicated non-interactions, an intuitively appealing choice, can amplify the shortcut stemming from functional relatedness. To detect and mitigate these biases, we provide an open Nextflow pipeline that combines similarity-aware, data-loss-minimizing dataset splitting with bias-minimizing negative sampling, both formulated as integer linear programs. Its key concept of quantifying biases to minimize them through optimization-based negative sampling can, in principle, be extended to any machine learning problem where the pool of negative candidates is much larger than the positives and is thus of interest also beyond PPI prediction.
| Subjects: | Molecular Networks (q-bio.MN); Machine Learning (cs.LG) |
| Cite as: | arXiv:2609.10193 [q-bio.MN] |
| (or arXiv:2609.10193v2 [q-bio.MN] for this version) | |
| https://doi.org/10.48550/arXiv.2609.10193 arXiv-issued DOI via DataCite |
Submission history
From: Judith Bernett [view email]
[v1]
Wed, 9 Sep 2026 14:02:18 UTC (816 KB)
[v2]
Wed, 18 Nov 2026 12:43:58 UTC (1,441 KB)
来源:arXiv q-bio.MN 分子生物学预印本 · arxiv.org