Hello,

I am building a gene regulatory network of some single cell data using a pipeline called IReNA v2. The output of the pipeline is a dataframe of TF-gene binding interactions, given a tag of either 'positive' or 'negative' for each interaction. I am given a dataframe per cell type and can filter/data wrangle them so that I can get interactions between cell types as something like:

Cell Type A interacts with Cell Type B with 45 positive interactions and 50 negative interactions. Cell Type B interacts with Cell Type A with 21 positive interactions and 11 negative interactions. Cell Type A interacts with Cell Type C with 33 positive interactions and 7 negative interactions. etc., etc.

I want quantify the regulatory significance/tendency between each pair of cell types to get a global descriptor of the interactions between cell types like what is shown here in Figure 4F (right panel).

In the methods section of their paper, the authors write: "To study the global regulatory relationships among retinal cell types in each development stage, we calculated the regulatory significance and regulatory tendency between each pair of cell types using a previously described approach.... For each pair of cell types, we counted the regulations from cell type A to cell type B and calculated the significance of positive or negative regulations using the hypergeometric test respectively."

The previously described approach is as follows:

In my case, modules would be cell types. For each pairwise comparison, what are my M, Q, m and q? I haven't had to use non-basic statistics in a while...