search

BiomedicalMachineLearning / stLearn

A novel machine learning pipeline to analyse spatial transcriptomics data
Other
176 stars 23 forks source link

Single cell transcriptomic data for cell-cell communication #280

Closed liangbolyu closed 5 months ago

liangbolyu commented 5 months ago

Hi, I read the paper and the tutorial of your method. But it seems that your method can only fix the cell-cell communication on spots level. Could you please tell me how to use your method to do the cell-cell communication on single cell level data?

BiomedicalMachineLearning commented 5 months ago

You can use between spot mode CCI to analyse single-cell resolution interaction. One of our figures show that. If there are too many cells, to make the computation faster, you can apply the gridding option: https://stlearn.readthedocs.io/en/latest/tutorials/Xenium_CCI.html

liangbolyu commented 5 months ago

Hi, have you shown "between spot mode CCI to analyse single-cell resolution interaction" in your tutorial? Or where can I find it?

BiomedicalMachineLearning commented 5 months ago

in the link above, you can see:

Running the analysis

st.tl.cci.run(grid, lrs, min_spots = 20, #Filter out any LR pairs with no scores for less than min_spots distance=None, # None defaults to spot+immediate neighbours; distance=0 for within-spot mode n_pairs=1000, # Number of random pairs to generate; low as example, recommend ~10,000 n_cpus=None, # Number of CPUs for parallel. If None, detects & use all available. )