plot like spottedpy

[worker000000]

How to use calculate_module_scores function in Hotspot library?

Question

Hi Hotspot team,

I'm working with spatial transcriptomics data and noticed there's a calculate_module_scores function in the Hotspot library source code. However, I couldn't find clear documentation or examples on how to use this function with gene sets.

What I'm trying to achieve

I want to calculate module/pathway scores for specific gene sets (like EMT, hypoxia, immune signatures) in my spatial transcriptomics data, and then identify hotspots based on these module scores.

Current understanding

I know the basic Hotspot workflow:

import hotspot
hs = hotspot.Hotspot(adata)
hs.create_knn_graph(weighted_graph=False, n_neighbors=30)
hs_results = hs.compute_autocorrelations(jobs=1)

But I'm interested in using calculate_module_scores for gene set analysis.

Specific questions

1. Function signature: What are the parameters for calculate_module_scores?

2. Input format: How should gene sets be provided?

   # Option 1: Dictionary format?
   gene_sets = {
       'EMT': ['VIM', 'CDH2', 'SNAI1', 'SNAI2', 'TWIST1'],
       'Hypoxia': ['HIF1A', 'VEGFA', 'PDK1', 'LDHA'],
       'Immune': ['CD68', 'CD163', 'CSF1R', 'PTPRC']
   }
   
   # Option 2: DataFrame format?
   # Option 3: GMT file format?

3. Integration with Hotspot analysis:

   • Does calculate_module_scores return scores that can be used with compute_autocorrelations()?
   • Do I need to add the scores back to the AnnData object?

4. Usage example: Could you provide a minimal working example?

   import hotspot
   
   # How to use calculate_module_scores?
   hs = hotspot.Hotspot(adata)
   module_scores = hs.calculate_module_scores(???)  # What goes here?
   
   # Then how to proceed with hotspot analysis?

5. Relationship to other functions: How does this relate to other Hotspot functions like compute_modules() or compute_autocorrelations()?

6 can hotspot plot like spottedpy