ISAAC: Auditing Causal Reasoning in Deep Models for Drug-Target Interaction
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909013107539968 |
|---|---|
| author | Tarantino, Barbara Kim, Sun Lu, Yijingxiu Giudici, Paolo |
| author_facet | Tarantino, Barbara Kim, Sun Lu, Yijingxiu Giudici, Paolo |
| contents | Deep learning models for drug--target interaction (DTI) prediction often achieve strong benchmark performance without necessarily relying on mechanistically meaningful molecular features, a limitation that standard accuracy-based evaluation cannot detect. We introduce ISAAC (Intervention-based Structural Auditing Approach for Causal Reasoning), a post-hoc framework that evaluates prior-relative structural sensitivity by probing frozen models through matched mechanistic and spurious input-level interventions, independently of predictive accuracy. Applied to three sequence-based DTI architectures on the Davis benchmark, ISAAC reveals approximately 25\% relative differences in reasoning scores across models with comparable AUROC (within around 3\%), stable across training and intervention seeds and two distinct perturbation operators. These discrepancies, undetectable under conventional accuracy metrics, motivate the use of post-hoc structural auditing as a complement to standard performance evaluation in scientific machine learning for molecular modeling. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_02962 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ISAAC: Auditing Causal Reasoning in Deep Models for Drug-Target Interaction Tarantino, Barbara Kim, Sun Lu, Yijingxiu Giudici, Paolo Machine Learning Computation Deep learning models for drug--target interaction (DTI) prediction often achieve strong benchmark performance without necessarily relying on mechanistically meaningful molecular features, a limitation that standard accuracy-based evaluation cannot detect. We introduce ISAAC (Intervention-based Structural Auditing Approach for Causal Reasoning), a post-hoc framework that evaluates prior-relative structural sensitivity by probing frozen models through matched mechanistic and spurious input-level interventions, independently of predictive accuracy. Applied to three sequence-based DTI architectures on the Davis benchmark, ISAAC reveals approximately 25\% relative differences in reasoning scores across models with comparable AUROC (within around 3\%), stable across training and intervention seeds and two distinct perturbation operators. These discrepancies, undetectable under conventional accuracy metrics, motivate the use of post-hoc structural auditing as a complement to standard performance evaluation in scientific machine learning for molecular modeling. |
| title | ISAAC: Auditing Causal Reasoning in Deep Models for Drug-Target Interaction |
| topic | Machine Learning Computation |
| url | https://arxiv.org/abs/2605.02962 |