Fig. 6: SpatialTopic is scalable to large-scale images and can be run on a regular laptop within minutes. | Nature Communications

Fig. 6: SpatialTopic is scalable to large-scale images and can be run on a regular laptop within minutes.

From: Scalable topic modelling decodes spatial tissue architecture for large-scale multiplexed imaging analysis

Fig. 6: SpatialTopic is scalable to large-scale images and can be run on a regular laptop within minutes.The alternative text for this image may have been generated using AI.

A Runtime of SpatialTopic (region radius r = 60) and Seurat-v5 (K = 30) on simulated datasets with increasing cell numbers on a single image. n = 5 replicates. The Error bar represents the standard error of the mean. B Runtime of SpatialTopic, Seurat-v5, BankSY, UTAG, Spatial-LDA, and CytoCommunity on large-scale Nanostring NSCLC and CODEX mouse spleen datasets. All methods were benchmarked on a standard MacBook Air (M2, 2022) unless exceeding the memory limitation. Source data are provided as a Source Data file.

Back to article page