graph neural network

ChemOrigins: A community curated database for origins of life chemistry

ChemOrigins is a community-curated database that organizes experimentally supported prebiotic reactions into a knowledge graph for research and education.

Graph embeddings for identifying symmetries in connectomes

The paper introduces a graph embedding algorithm to identify symmetries in neural connectomes, demonstrating its ability to detect rotational, translational, and toroidal symmetries in Drosophila and synthetic grid cell connectomes.

SpatialRNA: a Python package for easy application of Graph Neural Network models on single-molecule spatial transcriptomicsdataset

The paper introduces SpatialRNA, a Python package enabling the application of Graph Neural Networks to spatial transcriptomics data for domain detection.

PRIorI: a graph-based mining of structural arrangements in protein-protein interfaces

PRIorI identifies conserved structural motifs in protein-protein interfaces using graph isomorphism, demonstrating its utility in analyzing the SARS-CoV-2 Spike-ACE2 interface.

Spatially distinct chromatin compaction states predict neoadjuvant chemotherapy resistance in Triple Negative Breast Cancer

The paper uses FLIM to measure chromatin compaction in TNBC tissues and finds that spatial chromatin states predict resistance to neoadjuvant chemotherapy.

Deep Learning Reveals Endogenous Sterols as Allosteric Modulators of the GPCR-Gα Interface

The paper identifies endogenous sterols as intracellular allosteric modulators of the GPCR-Gα interface using a deep learning toolkit and experimental validation.

predPPI-GReMLIN: prediction of protein-protein interactions through mining of conserved bipartite graphs

predPPI-GReMLIN predicts protein-protein interactions by analyzing conserved bipartite graphs, achieving high accuracy across multiple datasets.

SpatialProp: tissue perturbation modeling with spatially resolved single-cell transcriptomics

SpatialProp predicts the effects of multi-gene perturbations on whole tissue sections using spatially resolved single-cell transcriptomics and graph neural networks.

Novae: a graph-based foundation model for spatial transcriptomics data

Novae is a graph-based foundation model for spatial transcriptomics that improves domain inference and batch correction across multiple tissues and technologies.

DRIVE-KG: Enhancing variant-phenotype association discovery in understudied complex diseases using heterogeneous knowledge graphs

DRIVE-KG integrates multi-omics data into a knowledge graph to discover SNP-phenotype associations and predict endometriosis status, uncovering new variant-phenotype links and improving classification performance.

VaxjoGNN: A Graph Neural Network for Ontology-Grounded Vaccine Adjuvant Recommendation

This paper introduces VaxjoGNN, a graph neural network for recommending vaccine adjuvants based on biomedical ontologies, achieving significant improvements over random baseline.

GRNITE: Gene Regulatory Network Inference with Text Embeddings

GRNITE improves gene regulatory network inference by combining text embeddings and graph neural networks, enhancing AUROC and recall across methods.

Prior knowledge informs graph neural networks to improve phenotype prediction from proteomics

The paper develops a graph neural network framework that integrates biological knowledge to predict phenotypes from proteomics data, outperforming linear models and other architectures.

PRSNet-2: End-to-end genotype-to-phenotype prediction via hierarchical graph neural networks

PRSNet-2 predicts complex phenotypes from genotypic data using a hierarchical graph neural network, outperforming baselines and reducing overfitting.

PathWeigh II: Graph Based Belief Propagation for Pathway Activity Analysis

PathWeigh II improves pathway activity analysis by using graph decomposition and belief propagation to handle feedback loops and integrate gene-level evidence.

Cis-TERN: a graph-based method to integrate single-cell chromatin accessibility with gene expression data to infer the pseudotime ordering of cells

Cis-TERN integrates scRNA-seq and scATAC-seq data to infer pseudotime ordering without prior biological knowledge, aligning with known lineages and tracing cancer cells to progenitors.

Semantic-Aware Graph Embedding Approach Uncovers LC-61, a Potent Anti-Leishmania infantum Compound

The paper introduces a semantic-aware graph neural network framework for drug discovery, identifying LC-61 as a potent antileishmanial compound with nanomolar activity.

Characterizing structural and kinetic ensembles of intrinsically disordered proteins using writhe

This paper uses writhe to analyze conformational ensembles and dynamics of intrinsically disordered proteins, showing it improves Markov state models and enables equivariant neural network sampling.

CIRPIN: Learning Circular Permutation-Invariant Representations to Uncover Putative Protein Homologs

CIRPIN identifies protein homologs related by circular permutation using a permutation-invariant graph neural network, uncovering thousands of novel pairs and revealing natural PDZ domain permutations.

GAT-HiC: Efficient Reconstruction of 3D Chromosome Structure via Residual Graph Attention Neural Networks

GAT-HiC predicts 3D chromosome structure from Hi-C data using graph attention networks, showing generalization across different datasets and outperforming existing methods.

Multimodal AI Decodes Extreme Environment Functional Dark Matter Beyond Homology

The paper develops ACCESS, a multimodal graph neural network to annotate functional dark matter proteins from extreme environments beyond homology-based methods.

UniGEN-DDI: Computing drug-drug interactions using a unified graph embedding network

UniGEN-DDI computes drug-drug interactions using a unified graph embedding network, identifying 12 unknown interactions validated in DrugBank 6.0.

AstraBIND: Graph Attention Network for Predicting Ligand Binding Sites

AstraBIND predicts ligand binding sites using a graph attention network, achieving high accuracy across multiple ligand classes while maintaining efficiency.

PyrMol: A Knowledge-Structured Pyramid Graph Framework forGeneralizable Molecular Property Prediction

PyrMol improves molecular property prediction by integrating expert knowledge into a hierarchical graph framework, outperforming existing methods on benchmark datasets.

Graph neural networks for integrated information and major complex estimation

This paper uses graph neural networks to estimate integrated information and major complex in IIT 3.0, showing comparable performance on extrapolated systems.

MetaboliteChat: A Unified Multimodal Large Language Model for Interactive Metabolite Analysis and Functional Insights

MetaboliteChat is a multimodal LLM that integrates molecular graphs and images with natural language to provide interactive metabolite analysis and functional insights.

ECLIPSE: Exploration of Complex Ligand-Protein Interactions through Learning from Systems-level Heterogeneous Biomedical Knowledge Graphs

ECLIPSE predicts compound-protein interactions using heterogeneous biomedical knowledge graphs and graph neural networks, achieving strong performance on benchmark tasks.

Modeling protein-small molecule conformational ensembles with PLACER

An atom-level graph neural network that denoises corrupted structures generates small-molecule and protein-ligand conformational ensembles, and its active-site pre-organization metric enriched for more active designed retroaldolases.

MolUNet++: Adaptive-grained Explicit Substructure and Interaction Aware Molecular Representation Learning

MolUNet++ improves molecular representation learning by explicitly analyzing substructures and interactions, outperforming traditional GNN models in prediction tasks and offering interpretable insights.

Graph attention with structural features improves the generalizability of identifying functional sequences at a protein interface

The paper improves prediction of functional protein interface sequences by integrating structural and sequence features using a graph attention model, achieving better generalization across diverse variants.