Gregor Sturm

Breaking mudata is not your fault. It was caused by an anndata release and should be fixed by now. Just rerun the tests :)

@MKanetscheider, two more considerations 1. Can we somehow verify that the alignments are really using the IMGT reference? Could we at least check something like length? If they were using...

> I think I actually do already check for length, because I count differences via hamming-distance, which raises a ValueError if germline and sequence alignments have different lengths I think...

> I think this sounds rather amazing...as you are already well aware, this whole function is rather ugly and not that user-friendly at the moment...I would really love to see...

Closing in favor of #573. Thanks for all your input @MKanetscheider! The optimized version in #573 wouldn't have been possible without it!

I can see your point... the parameters in the paramsheet would be method-specific though, so I would find it a bit confusing if they were overriden by global parameters. Really...

Hi @zktuong, actually, the J gene is currently not considered at all. But actually, afaik @felixpetschko is working on having a `same_j_gene` feature as part of the optimized clonotype calling...

@zktuong, could d_call/c_call also be relevant? Mostly asking because I'm debating the interface with @felixpetschko in https://github.com/scverse/scirpy/pull/470#issuecomment-2289156164 We could have either ``` same_v_gene: bool = True, same_j_gene: bool = True...

As discussed with @zktuong, it would be nice to refer to the dandelion preprocessing workflow (which addresses some issues with the cellranger output) from this package and/or scirpy. In the...

Hi, unfortunately, there is no straightforward answer. I am not aware of any benchmark that shows that at a given cutoff, receptors still have a X% chance of binding to...