Fig. 2: STAGATE improves the identification of layer structures in the human dorsolateral prefrontal cortex (DLPFC) tissue. | Nature Communications

Fig. 2: STAGATE improves the identification of layer structures in the human dorsolateral prefrontal cortex (DLPFC) tissue.

From: Deciphering spatial domains from spatially resolved transcriptomics with an adaptive graph attention auto-encoder

Fig. 2

a Ground-truth segmentation of cortical layers and white matter (WM) in the DLPFC section 151676. b Boxplot of clustering accuracy in all 12 sections of the DLPFC dataset in terms of adjusted rand index (ARI) scores for seven methods. In the boxplot, the center line, box limits and whiskers denote the median, upper and lower quartiles, and 1.5× interquartile range, respectively. c Cluster assignments generated by SCANPY, SEDR, SpaGCN, BayesSpace, and STAGATE in the DLPFC section 151676. d UMAP visualizations and PAGA graphs generated by SCANPY, stLearn, SEDR, and STAGATE embeddings respectively in the DLPFC section 151676. As end-to-end clustering approaches, SpaGCN and BayesSpace cannot be visualized using UMAP and PAGA.

Back to article page