Cell Lineage Analysis with SoptSC

[boyiguo1]

As one of the proposed analyses [#16], I tried out the SoptSC. However, the analysis is extremely slow at the prepping step, i.e. calculating the similarity matrix even for a toy example that consists of 1000 cells and 10 genes. The documentation of the package is very hard to understand either. Hence, I will be de-prioritize this analysis, or using other methods to provide the similarity matrix, e.g SC3.

[boyiguo1]

Deprioritize this analysis. Close it now

[lcolladotor]

If you want to, you could ask in the meantime to the authors if that's the right input. Maybe their method works best on counts vs logcounts as input (where there are less entries in the gene x spot matrix; the logcounts matrix converts 0s into non-zero values).

[boyiguo1]

This is a good idea! I'll remember to do this in the future. The context I was in was I saw the package is not maintained in the past couple of years and the documentation is somewhat poor. So I didn't spend time following up.

But I think you are right. I could be more active in contacting the authors, which I never thought of that much!