Paper List
-
Simulation and inference methods for non-Markovian stochastic biochemical reaction networks
This paper addresses the computational bottleneck of simulating and performing Bayesian inference for non-Markovian biochemical systems with history-d...
-
Translating Measures onto Mechanisms: The Cognitive Relevance of Higher-Order Information
This review addresses the core challenge of translating abstract higher-order information theory metrics (e.g., synergy, redundancy) into defensible, ...
-
Emergent Bayesian Behaviour and Optimal Cue Combination in LLMs
This paper addresses the critical gap in understanding whether LLMs spontaneously develop human-like Bayesian strategies for processing uncertain info...
-
Vessel Network Topology in Molecular Communication: Insights from Experiments and Theory
This work addresses the critical lack of experimentally validated channel models for molecular communication within complex vessel networks, which is ...
-
Modulation of DNA rheology by a transcription factor that forms aging microgels
This work addresses the fundamental question of how the transcription factor NANOG, essential for embryonic stem cell pluripotency, physically regulat...
-
Imperfect molecular detection renormalizes apparent kinetic rates in stochastic gene regulatory networks
This paper addresses the core challenge of distinguishing genuine stochastic dynamics of gene regulatory networks from artifacts introduced by imperfe...
-
Approximate Bayesian Inference on Mechanisms of Network Growth and Evolution
This paper addresses the core challenge of inferring the relative contributions of multiple, simultaneous generative mechanisms in network formation w...
-
An AI Implementation Science Study to Improve Trustworthy Data in a Large Healthcare System
This paper addresses the critical gap between theoretical AI research and real-world clinical implementation by providing a practical framework for as...
DeepFRI Demystified: Interpretability vs. Accuracy in AI Protein Function Prediction
Yale University | Microsoft
The 30-Second View
IN SHORT: This study addresses the critical gap between high predictive accuracy and biological interpretability in DeepFRI, revealing that the model often prioritizes structural motifs over functional residues, complicating reliable identification of drug targets.
Innovation (TL;DR)
- Methodology Comprehensive benchmarking of three post-hoc explainability methods (GradCAM, Excitation Backpropagation, PGExplainer) on DeepFRI with quantitative sparsity analysis.
- Methodology Development of a modified DeepFool adversarial testing framework for protein sequences, measuring mutation thresholds required for misclassification.
- Biology Revealed that DeepFRI prioritizes amino acids controlling protein structure over function in >50% of tested proteins, highlighting a fundamental accuracy-interpretability trade-off.
Key conclusions
- DeepFRI required 206 mutations (62.4% of 330 residues) in the lac repressor for misclassification, demonstrating extreme robustness but potentially missing subtle functional alterations.
- Explainability methods showed significant granularity differences: PGExplainer was 3× sparser than GradCAM and 17× sparser than Excitation Backpropagation across 124 binding proteins.
- All three methods converged on biochemically critical P-loop residues (0-20) in ARF6 GTPase, validating DeepFRI's focus on conserved functional motifs in straightforward domains.
Abstract: Machine learning technologies for protein function prediction are black box models. Despite their potential to identify key drug targets with high accuracy and accelerate therapy development, the adoption of these methods depends on verifying their findings. This study evaluates DeepFRI, a leading Graph Convolutional Network (GCN)-based tool, using advanced explainability techniques—GradCAM, Excitation Backpropagation, and PGExplainer—and adversarial robustness tests. Our findings reveal that the model’s predictions often prioritize conserved motifs over truly deterministic residues, complicating the identification of functional sites. Quantitative analyses show that explainability methods differ significantly in granularity, with GradCAM providing broad relevance and PGExplainer pinpointing specific active sites. These results highlight trade-offs between accuracy and interpretability, suggesting areas for improvement in DeepFRI’s architecture to enhance its trustworthiness in drug discovery and regulatory settings.