GAHIB: graph attention VAE with a hyperbolic information bottleneck for biologically structured single-cell representations

Current single-cell RNA-sequencing (scRNA-seq) variational autoencoders (VAEs) usually emphasise local cell graph structure, hyperbolic latent geometry, or bottleneck compression separately, yet these biases are rarely evaluated together in one evidence-gated representation model. We present GAHIB (...

Ausführliche Beschreibung

Bibliografische Detailangaben

Hauptverfasser:	Zeyu Fu, Jiawei Fu, Xiaoxia Wang, Yiyao Liu, Tianfei Ran
Format:	Artigo
Sprache:	Inglês
Veröffentlicht:	Frontiers Media S.A. 2026-06-01
Schriftenreihe:	Frontiers in Genetics
Schlagworte:	graph attention network hyperbolic geometry information bottleneck representation learning single-cell RNA-seq variational autoencoder
Online-Zugang:	https://www.frontiersin.org/articles/10.3389/fgene.2026.1863100/full
Tags:	Tag hinzufügen Keine Tags, Fügen Sie das erste Tag hinzu!