Collect real-world datasets for benchmarking.

[lzj1769]

Collect as many real-world spatial transcriptomics datasets from different technologies (e.g., Visium, Slide-seq, MERFISH, seqFISH, STARmap) as possible

This paper and website can be helpful:
https://www.nature.com/articles/s41592-023-01773-7
https://gene.ai.tencent.com/SpatialOmics/
https://pysodb.readthedocs.io/en/latest/api/api.html#Import-and-Initialization

For benchmarking, we might also need to care about the data quality.