PAST capably captures local spatial patterns with self-attention mechanism. (A) Quantitative performance evaluation of spatial domain characterization on the osmFISH MSC data set via supervised cross-validation and unsupervised spatial clustering with a specified number of clusters (Ncluster) and with default resolution (Dlouvain). The cross-validation performance was evaluated by the average score of Cohen's kappa value (κ) and mean F1 score (mF1), respectively, in the fivefold experiments. The spatial clustering performance was evaluated by adjusted rand index (ARI) and adjusted mutual information (AMI). (B) Visualization of the results of spatial clustering with a specified number of clusters on MSC. (C) The manual annotation of MSC. (D) UMAP visualization of spots in MSC. The blue, orange, and brown circles denote Layer 2–3 lateral, Layer 2–3 medial, and ventricle, respectively. (E) The neighbor proportion and attention proportion for border spots of domains Layer 2–3 lateral, Layer 2–3 medial, and ventricle, respectively.
