Fylo›ARCADIA›Graph
Hubs
Association·arcadia

Tree sequence data and GNNs are well-suited for population genetics

Claim that evolutionarily meaningful, graph-structured data such as tree sequences are best handled by GNNs for population genetic inference.

Confidence
80%
active

Evidence Quote

“GNNs can infer everything from demographic history to selection using tree sequence data.”

Relationship

Tree sequence enables Demographic history inference

Arguments

Tree sequencesubject
Demographic history inferenceobject

Connections (5)

GNNs as flexible models for evolutionary inference tasksInferenceChain
Structural similarity often diverges from sequence similarityAssociation
Distinction between explanatory and predictive modeling in evolutionary inferenceInferenceChain
Theoretical and empirical foundations of GNN applicabilityInferenceChain
Graph neural network foundations enable inference in population geneticsInferenceChain

Evidence

“Reference to Whitehouse LS, Ray DD, Schrider DR. (2024)”

(2024). Tree Sequences as a General-Purpose Tool for Population Genetic Inference doi:10.1093/molbev/msae223 ↗

“Reference to Kelleher J, Wong Y, Wohns AW, Fadil C, Albers PK, McVean G. (2019)”

(2019). Inferring whole-genome histories in large population datasets doi:10.1038/s41588-019-0483-y ↗