Skip to content

Graph Neural Network (GNN)

Definition
AI-generated

A graph neural network (GNN) learns representations on graph-structured data by passing and aggregating messages between nodes (and sometimes edges); layers stack neighborhood aggregation so each node embedding reflects local topology and features.

Topics
Machine learning concepts
Related
Deep Learning Gene Regulatory Network Inference Neural Network

Why it matters in GWAS

GNNs encode protein–protein or gene regulatory networks, pathway topology, single-cell kNN graphs, and 3D chromatin contact graphs for tasks such as gene prioritization at loci, drug target prediction, or denoising spatial assays. Graph construction choices (which edges, which cells) strongly affect biological interpretation and can couple to batch or sampling artifacts.

Example usage

"We ranked genes in the credible set with a GNN over STRING interactions and GTEx coexpression edges."

Deep Learning Gene Regulatory Network Inference Neural Network Pathway Analysis Single-Cell Sequencing UMAP
Related-term network

Click any node to open that term page.

Node order. The current term is first. Next come every page linked from Related terms, in the same top-to-bottom order as that list. Then new terms appear in two-hop layers: for each of those direct neighbors (in that same order), their related terms are appended if not already listed. A third pass does the same for two-hop nodes. The walk stops when a fixed maximum number of nodes is reached.

References

  • Zhou J, et al. (2020). Graph neural networks: A review of methods and applications. AI Open.
← Graph genome Genome-based Restricted Maximum Likelihood (GREML) →

Last updated (UTC · Git history)