cs.LGMar 1, 2026

Graph Attention Based Prioritization of Disease Responsible Genes from Multimodal Alzheimer's Network

Authors: Binon Teji, Subhajit Bandyopadhyay, Swarup Roy

Organizations: Network Reconstruction & Analysis Lab and Department of Computer Applications, Sikkim University, Sikkim, India, 737102 · Department of Computer Science & Engineering, Tezpur University, Assam, India, 784028

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Prioritizing disease-associated genes is central to understanding the molecular mechanisms of complex disorders such as Alzheimer's disease (AD). Traditional network-based approaches rely on static centrality measures and often fail to capture cross-modal biological heterogeneity. We propose NETRA (Node Evaluation through Transformer-based Representation and Attention), a multimodal graph transformer framework that replaces heuristic centrality metrics with attention-driven relevance scoring. Using AD as a case study, gene regulatory networks are independently constructed from microarray, single-cell RNA-seq, and single-nucleus RNA-seq data. Random-walk sequences derived from these networks are used to train a BERT-based model for learning global gene embeddings, while modality-specific gene expression profiles are compressed using variational autoencoders. These representations are integrated with auxiliary biological networks, including protein-protein interactions, Gene Ontology semantic similarity, and diffusion-based gene similarity, into a unified multimodal graph. A graph transformer assigns NETRA scores that quantify gene relevance in a disease-specific and context-aware manner. Gene set enrichment analysis shows that NETRA achieves a normalized enrichment score of about 3.9 for the Alzheimer's disease pathway, substantially outperforming classical centrality measures and diffusion models. Top-ranked genes enrich multiple neurodegenerative pathways, recover a known late-onset AD susceptibility locus at chr12q13, and reveal conserved cross-disease gene modules. The framework preserves biologically realistic heavy-tailed network topology and is readily extensible to other complex disorders.

Figures & tables

Fig. 1: Overview of the proposed multimodal graph–transformer framework for gene prioritization. (A) Multi-omics gene expression data from microarray, single-cell RNA-seq (scRNA-seq), and single-nucleus RNA-seq (snRNA-seq). (B) Each modality is encoded using a variational autoencoder (VAE) to obtain latent representations Zma\mathcal{Z}_{ma} , Zsc\mathcal{Z}_{sc} , and Zsn\mathcal{Z}_{sn} , which are fused into a unified gene embedding ZF\mathcal{Z}_{F} . (C) Multiple inferred networks are pooled and transformed into node sequences using random-walk sampling, which are then tokenized, positionally encoded, and encoded using a BERT-based Transformer to learn rich contextual representations. (D) The fused embeddings are integrated with the ensemble graph A\mathcal{A} and graph positional encodings to produce node features h\mathbf{h} . (E) A multi-layer graph transformer applies neighborhood attention using graph-specific query, key, and value projections for link prediction and attention aggregation. The aggregated gene-attention scores are used for gene ranking, enrichment analysis (e.g., GSEA, pathway and disease databases), and final reporting and visualization.
Fig. 2: Training dynamics of our learning model. (left) training loss convergence and (right) AUROC progression across epochs, demonstrating stable optimization and robust generalization.
Fig. 3: UMAP visualization of gene embeddings with Leiden clustering and attention-prioritized (top-15) genes. Each point represents a gene embedding projected into two dimensions. Colors indicate Leiden clusters (labels 0–13). Attention-prioritized genes are highlighted in red and annotated with gene symbols using arrowed labels. The spatial distribution demonstrates clear cluster separation while showing the dispersion of prioritized genes across multiple functional communities.
Fig. 4: Visualization of an attention-weighted subgraph induced by the top-10 prioritized genes. Prioritized genes are shown in gold, neighboring context genes in gray, with node sizes proportional to NETRA attention scores. Blue edges represent gene-gene interactions.
Fig. 5: Structural comparison between the input ensemble network and the generated network. (a) Comparison of global graph characteristics between the input ensemble network and the generated network on a logarithmic scale, including maximum degree, triangle count, global clustering coefficient, and global efficiency. (b) Log–log degree distribution of the input ensemble network and the generated network, demonstrating the preservation of heavy-tailed degree behavior in the generated network.
Fig. 6: GSEA enrichment profiles and NES comparison for the Alzheimer’s disease pathway (KEGG: hsa05010). Panels (a)–(e) show the running enrichment score (green curve), gene rank positions (black bars), and score distribution (gray area) for ranking methods: (a) attention-based score NETRA, (b) degree, (c) betweenness, (d) eigenvector, and (e) PageRank centrality. Panel (f) compares the corresponding normalized enrichment scores (NES) across all five methods. The attention-based log2 score NETRA achieves the highest enrichment (NES ≈\approx 3.9, FDR ≈1.54×10−28\approx 1.54\times 10^{-28} ), demonstrating superior biological signal integration compared to topology-only centrality measures.
Fig. 7: KEGG pathway enrichment analysis based on attention-derived gene ranking. (a) Dot plot summarizing Gene Set Enrichment Analysis (GSEA) results, where dot size represents the gene ratio and color intensity indicates the adjusted p-value. Significant enrichment is observed across multiple neurodegenerative pathways, along with metabolic and synaptic signaling processes, supporting the biological relevance of the proposed attention-based gene prioritization. (b) Ridge plot illustrating the distribution of attention-ranked genes across enriched KEGG pathways. Each ridge represents the density of gene rankings within a pathway, with peak positions indicating the concentration of pathway genes along the ranked list. Consistent early enrichment of neurodegenerative pathways highlights robust prioritization of disease-relevant genes.
Fig. 8: Multi-level functional and genomic validation of attention-prioritized genes. (a)KEGG pathway enrichment analysis of the top 40 attention-ranked genes, showing significant overrepresentation of neurodegeneration-related pathways alongside apoptotic and iron metabolism processes. (b)WikiPathways enrichment analysis of the top 40 genes, highlighting Alzheimer’s disease–associated signaling pathways and shared molecular mechanisms relevant to neurodegeneration. (c)DISGENET disease enrichment analysis for the top 40 genes, revealing convergence across multiple neurodegenerative and neurological disorders, indicating shared disease mechanisms. (d) Genomic localization of attention-prioritized genes on chromosome 12, highlighting a cluster of four genes mapping to the 12q13 cytoband, a locus previously implicated in Alzheimer’s disease through genome-wide association studies [ 22 , 23 ]
Fig. 9: UpSet plot showing shared genes across neurodegenerative pathways. Bar heights indicate the number of genes in each intersection. The largest intersections reveal core neurodegeneration genes prioritized by attention scores.
Fig. 10: Jaccard similarity between neurodegenerative pathway gene sets. Higher values indicate greater overlap in core enrichment genes.
Figure 11
Figure 12
Figure 13
More in this topic

Genomics

Genomics uses computational and experimental methods to study genomes, genetic variation, gene regulation, and their links to traits and disease.

See all 125 papers

Explore similar work

CardsList
Jun 2, 2026

Multi-Modal Graph Neural Network with Transformer-Guided Adaptive Diffusion for Preclinical Alzheimer Classification

The graphical representation of the brain offers critical insights into diagnosing and prognosing neurodegenerative disease via relationships between regions of interest (ROIs). Despite recent emergence of various Graph Neural Networks (GNNs) to effectively capture the relational information, there remain inherent limitations in interpreting the brain networks. Specifically, convolutional approaches ineffectively aggregate information from distant neighborhoods, while attention-based methods exhibit deficiencies in capturing node-centric information, particularly in retaining critical characteristics from pivotal nodes. These shortcomings reveal challenges for identifying disease-specific variation from diverse features from different modalities. In this regard, we propose an integrated framework guiding diffusion process at each node by a downstream transformer where both short- and long-range properties of graphs are aggregated via diffusion-kernel and multi-head attention respectively. We demonstrate the superiority of our model by improving performance of pre-clinical Alzheimer's disease (AD) classification with various modalities. Also, our model adeptly identifies key ROIs that are closely associated with the preclinical stages of AD, marking a significant potential for early diagnosis and prevision of the disease.

Jaeyoon Sim, Minjae Lee, Guorong Wu +1

Pohang University of Science and Technology, Pohang, South Korea · University of North Carolina at Chapel Hill, Chapel Hill, USA

Graph Neural NetworksAlzheimer's Disease ClassificationMultimodal Graph LearningGraph TransformerGraph Diffusion
AskPDF
Feb 17, 2026

MRC-GAT: A Meta-Relational Copula-Based Graph Attention Network for Interpretable Multimodal Alzheimer's Disease Diagnosis

Alzheimer's disease (AD) is a progressive neurodegenerative condition necessitating early and precise diagnosis to provide prompt clinical management. Given the paramount importance of early diagnosis, recent studies have increasingly focused on computer-aided diagnostic models to enhance precision and reliability. However, most graph-based approaches still rely on fixed structural designs, which restrict their flexibility and limit generalization across heterogeneous patient data. To overcome these limitations, the Meta-Relational Copula-Based Graph Attention Network (MRC-GAT) is proposed as an efficient multimodal model for AD classification tasks. The proposed architecture, copula-based similarity alignment, relational attention, and node fusion are integrated as the core components of episodic meta-learning, such that the multimodal features, including risk factors (RF), Cognitive test scores, and MRI attributes, are first aligned via a copula-based transformation in a common statistical space and then combined by a multi-relational attention mechanism. According to evaluations performed on the TADPOLE and NACC datasets, the MRC-GAT model achieved accuracies of 96.87% and 92.31%, respectively, demonstrating state-of-the-art performance compared to existing diagnostic models. Finally, the proposed model confirms the robustness and applicability of the proposed method by providing interpretability at various stages of disease diagnosis.

Fatemeh Khalvandi, Saadat Izadi, Abdolah Chalechale

Computer Engineering Department, Razi University, Kermanshah, Iran

Alzheimer's Disease ClassificationGraph Attention NetworksMultimodal Graph LearningMultimodal Model InterpretabilityCross-Modal Alignment
AskPDF
Feb 21, 2026

GLaDiGAtor: Language-Model-Augmented Multi-Relation Graph Learning for Predicting Disease-Gene Associations

Understanding disease-gene associations is essential for unravelling disease mechanisms and advancing diagnostics and therapeutics. Traditional approaches based on manual curation and literature review are labour-intensive and not scalable, prompting the use of machine learning on large biomedical data. In particular, graph neural networks (GNNs) have shown promise for modelling complex biological relationships. To address limitations in existing models, we propose GLaDiGAtor (Graph Learning-bAsed DIsease-Gene AssociaTiOn pRediction), a novel GNN framework with an encoder-decoder architecture for disease-gene association prediction. GLaDiGAtor constructs a heterogeneous biological graph integrating gene-gene, disease-disease, and gene-disease interactions from curated databases, and enriches each node with contextual features from well-known language models (ProtT5 for protein sequences and BioBERT for disease text). In evaluations, our model achieves superior predictive accuracy and generalisation, outperforming 14 existing methods. Literature-supported case studies confirm the biological relevance of high-confidence novel predictions, highlighting GLaDiGAtor's potential to discover candidate disease genes. These results underscore the power of graph convolutional networks in biomedical informatics and may ultimately facilitate drug discovery by revealing new gene-disease links. The source code and processed datasets are publicly available at https://github.com/HUBioDataLab/GLaDiGAtor.

Osman Onur Kuzucu, Tunca Doğan

Biological Data Science Lab, Dept. of Computer Engineering, Hacettepe University, 06800, Ankara, Turkey. · Dept. of Bioinformatics, Graduate School of Health Sciences, Hacettepe University, 06800, Ankara, Turkey. · Dept. of Health Informatics, Institute of Informatics, Hacettepe University, 06800, Ankara, Turkey.

Heterogeneous GNNsBiomedical KGs
AskPDF