Geometric Mean of Relative Overlap Frequencies

[rogershijin]

I have a question regarding this page in the docs: https://vdjtools-doc.readthedocs.io/en/master/overlap.html

[mikessh]

You're absolutely right, k goes over indices of clonotypes that belong to both i and j, so j is used implicitly. I'll replace it with your notation as it is more intuitive.