Exploring the Latent Information in Spatial Transcriptomics Data via Multi-View Graph Convolutional Network Based on Implicit Contrastive Learning.

This result underscores an important trend: methods that incorporate multi‐view learning (such as STMIGCL and Spatial‐MGCN) or contrastive learning (such as STMIGCL and GraphST) tend to produce smoother domain boundaries in domain identification tasks, significantly reducing the occurrence of discrete points.

also in 513 other papers