search

phbradley / conga

Clonotype Neighbor Graph Analysis
MIT License
79 stars 18 forks source link

Just iMHC Scoring Method #27

Closed kanefos closed 2 years ago

kanefos commented 2 years ago

Hi authors,

Love CONGA, super interesting innovative idea.

I work with a lot of bulk repertoire seq and would love to start trying to incorporate some of the biology you touched on when you identified the iMHC population.

Briefly, if I simply take the functions and files in imhc_scoring.py and feed these info each other will I acquire a way to score TCRs by their iMHC? Apologies if this is a slightly naive question but if you could lay out how I would go about doing this I would very grateful

Thanks,

phbradley commented 2 years ago

Yes, that is definitely do-able. How many TCRs would you want to score, roughly? ANd would these be beta chain sequences? Just curious if we'll need to do any optimization...

kanefos commented 2 years ago

Yes I have realised I might need to do some optimisation myself - this could be up to 100k unique TCRs per sample, alpha and beta. Now you've helpfully let me know the imhc_scoring.py is the key script I might take it away and let you know if I can accelerate things a little - thanks!