E-ABIN: an Explainable module for Anomaly detection in BIological Networks

2025-06-25

AI Summary

Existing gene aberration detection methods for biological networks are largely confined to single-dataset analysis and suffer from limited interpretability and lack of interactive analytical capabilities. To address these limitations, we propose a generalizable, interpretable graph neural network framework that integrates support vector machines, random forests, graph autoencoders (GAE), and graph adversarial attribute networks (GAAN) for end-to-end identification of aberrant modules from gene expression- and methylation-derived networks. The framework enables cross-disease transfer learning and incorporates a user-friendly visualization interface to significantly enhance result interpretability and clinical readability. Validation on bladder cancer and celiac disease datasets demonstrates high predictive accuracy (AUC > 0.92), robust identification of known pathogenic genes (e.g., TP53, HLA-DQA1), and discovery of novel candidate therapeutic targets—thereby bridging high-performance prediction with mechanistic biological insight.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The increasing availability of large-scale omics data calls for robust analytical frameworks capable of handling complex gene expression datasets while offering interpretable results. Recent advances in artificial intelligence have enabled the identification of aberrant molecular patterns distinguishing disease states from healthy controls. Coupled with improvements in model interpretability, these tools now support the identification of genes potentially driving disease phenotypes. However, current approaches to gene anomaly detection often remain limited to single datasets and lack accessible graphical interfaces. Here, we introduce E-ABIN, a general-purpose, explainable framework for Anomaly detection in Biological Networks. E-ABIN combines classical machine learning and graph-based deep learning techniques within a unified, user-friendly platform, enabling the detection and interpretation of anomalies from gene expression or methylation-derived networks. By integrating algorithms such as Support Vector Machines, Random Forests, Graph Autoencoders (GAEs), and Graph Adversarial Attributed Networks (GAANs), E-ABIN ensures a high predictive accuracy while maintaining interpretability. We demonstrate the utility of E-ABIN through case studies of bladder cancer and coeliac disease, where it effectively uncovers biologically relevant anomalies and offers insights into disease mechanisms.

Problem

• Detects anomalies in biological networks using explainable AI
• Integrates multiple ML techniques for high predictive accuracy
• Provides user-friendly interface for interpreting disease mechanisms

Innovation

• Combines classical and graph-based deep learning
• Integrates SVM, Random Forests, GAEs, and GAANs
• User-friendly platform for anomaly detection