graph neural network

A Unique Graph-Theoretic Truth Table for Cross-Gene Branch Identity

This paper introduces a graph-theoretic framework to uniquely represent cross-gene branch identities in phylogenomic data, ensuring consistent and auditable comparative analyses.

Graph in Graph (GiG): A novel graph AI framework for integrating and interpreting whole-person healthmedical and omics data

This paper introduces GiG, a graph AI framework that integrates and interprets whole-person medical and omics data to prioritize biomarkers and improve disease diagnosis.

DeorphaNN: Virtual screening of GPCR peptide agonists using AlphaFold-predicted active-state complexes and deep learning embeddings

DeorphaNN identifies peptide agonists for GPCRs using AlphaFold structural predictions and deep learning embeddings, successfully uncovering agonists for two orphan receptors across species.

SpatialJEPA: JEPA-inspired graph-context distillation for spatially aware multiomics integration

SpatialJEPA transfers spatial context from spatial multiomics data to dissociated non-spatial data through graph-context distillation, improving alignment and recovery of spatially organized programs in mouse brain multiomics.

Panomap: Unbiased Nanopore Signal Mapping with Pangenome Variation Graphs

Panomap maps raw nanopore signals to pangenome variation graphs, improving mapping accuracy and reducing reference bias compared to linear-reference methods.

Extended t-cores for the de novo identification of transposable elements and other inexact repeats from short read RNAseq data

This paper introduces a de novo method to identify transposable elements and inexact repeats from RNA-seq data without a reference genome, showing effectiveness in both model and non-model species.

FloREN: Decoding Immune Regulatory Networks through Interpretable Graph Transformer Patient Representations.

FloREN improves patient-level sample stratification and biomarker discovery by modeling single-cell data as an interpretable graph transformer network for immune regulatory analysis.

Learnable Graph Network Model (LGNM): A Physics Constrained Graph Neural Network with Quantum Hamiltonian Learning

The paper introduces LGNM, a physics-constrained graph neural network that improves protein flexibility prediction by learning per-residue flexibility coefficients, achieving higher accuracy than mENM on test proteins.

Spatial Topology Reveals Biologically Distinct Recurrent Motifs in Colorectal Cancer

The paper introduces STORM, a method to identify recurrent spatial motifs in colorectal cancer without cell-type labels, finding a fibroblast barrier linked to immune exclusion and prognosis.

Hidden sampling biases inflate performance in gene regulatory network inference

The paper shows that sampling biases in negative regulatory interactions inflate performance in gene regulatory network inference and proposes a degree-aware sampling approach to address this issue.

ORIGAMI: Orientation-Aware Graph Neural Network for Assessing Multimeric Interfaces of Protein Complex Structures

ORIGAMI improves assessment of multimeric protein complex structures by incorporating orientation-aware geometric features into graph neural networks, achieving better performance than existing methods.

Cross-Propagative Graph Learning Reveals Spatial Tissue Domains in Multi-Modal Spatial Transcriptomics

This paper introduces st-Xprop, a graph learning method for spatial domain detection in multi-modal spatial transcriptomics, improving clustering accuracy and biological relevance.

SPARC Reveals Distinct Osteoclastogenesis-Associated Malignant States in Metastatic Osteosarcoma

SPARC improves trajectory inference by jointly modeling state progression and transitions between conditions, revealing osteoclastogenesis-associated malignant states in osteosarcoma metastasis.

Gene Program Negotiation Defines Cellular Identity in Single-Cell Transcriptomes

This paper introduces Gene Program Negotiation (GPN), a framework to analyze regulatory interactions in single-cell transcriptomes, revealing new regulatory states and their association with cellular plasticity.

Graph-based modeling of multiparametric MRI deciphers molecular states of high-grade glioma invasion with prognostic implications

This paper uses graph-based modeling of multiparametric MRI to identify molecular states in high-grade glioma and shows their prognostic significance.

Graph neural network modeling of receptor interaction kinetics from single-molecule imaging data

The paper develops Deep-FISIK, a graph neural network, to predict receptor interaction kinetics from single-molecule imaging data without explicit tracking.

Biological Network Organization, Not Generic Graph Topology, Drives Graph-Based Gene Essentiality Prediction

This study evaluates whether graph-based gene essentiality prediction relies on biological network structure or generic graph topology, finding that biological connectivity is critical for performance.

Connectome-scale self-supervised representation learning reveals neuronal organization beyond canonical labels

This paper develops a self-supervised framework for learning structure-connectivity representations in dense connectomes, revealing neuronal organization beyond canonical labels.

Synteny-aware microbial pangenome graphs reveal blueprints of genomic variation

The paper introduces synteny-aware pangenome graphs to visualize genomic variation in marine bacteria, revealing structured continua of variability across genomes.

Zero-Shot Metabolite Prediction from Gene Expression via Physics-Informed Graph Neural Networks

GAZE predicts metabolite concentrations from gene expression using a physics-informed graph neural network, achieving better performance than existing methods in zero-shot scenarios.

SpaGRD deciphers signaling architectures in spatial transcriptomics using graph reaction-diffusion systems

SpaGRD models cell-cell communication using graph reaction-diffusion systems, revealing spatially resolved signaling patterns in spatial transcriptomics data.

IRCAS: a novel end-to-end approach to identify, rectify and classify comprehensive alternative splicing events in a transcriptome without genome reference

IRCAS improves reference-free alternative splicing detection in non-model organisms by integrating identification, rectification, and classification modules with specialized models.

DUET: a graph-based workflow for TCR-epitope prioritization and tumor-reactive T-cell identification

DUET improves TCR-epitope prioritization and tumor-reactive T-cell identification using a graph-based workflow, outperforming existing methods on benchmark datasets.

A Chromatin-Structure-Guided Framework for Predictive and Interpretable Regulatory Genomics

CHROME improves regulatory genomics predictions by integrating physically validated chromatin interactions into graph models, enhancing accuracy and interpretability for ChIP-seq, eQTL, and variant pathogenicity.

ACCURATE PREDICTION OF ASPARAGINE DEAMIDATION IN BIOLOGICS USING ADVANCED MACHINE LEARNING MODELS

This study develops machine learning models to predict asparagine deamidation in proteins, showing improved accuracy over existing methods and reducing development costs.

Kente: A Graph-based Pangenomic Approach for Horizontal Gene Transfer Detection in Microbiomes

Kente detects horizontal gene transfer in microbiomes using a pangenome graph-based approach, achieving higher precision and reducing false positives in simulated and real human gut metagenomes.

PepMCP: A Graph-Based Membrane Contact Probability Predictor for Membrane-Lytic Antimicrobial Peptides

PepMCP predicts membrane contact probability for short peptides using graph-based modeling, achieving high correlation with experimental data.

Zero-shot design of drug-binding proteins via neural selection-expansion

This paper introduces a method using two neural networks to design drug-binding proteins with high affinity, surpassing current state-of-the-art methods.

Structural Connectivity Selectively Constrains Intrinsic BOLD Timescales through Graph-Smooth Neural Activity

The paper uses graph signal processing to analyze how structural brain connectivity influences intrinsic BOLD timescales, showing structure-coupled signals strongly correlate with connectivity strength.

Deciphering shared and divergent tissue architectures from cross-species spatial transcriptomics

This paper introduces STACAME, a framework for analyzing cross-species spatial transcriptomics data to identify shared and divergent tissue architectures across species.