Cluster mouse single CD4 T cells according to TCR seq data

Question

[chi.delta]

Hey hey,

I have sequenced mouse CD4 T cells (5' 10x protocol, rna and tcr seq) and I was wondering if anyone is aware of a package where I could cluster my CD4 T cells according to their specificity (which could be predicted based on the TCR sequences).

Thanks a lot :)

Answer 1

[theHumanBorch]

There are a number of python-based tools for clustering TCRs based on sequence information

More popular pipelines:

• tcrdist - This approach uses edit distance to generate a network of TCRs. The authors of tcrdist have also come out with an approach that combines Gene expression and TCR information called CoNGA, which is very interesting
• GLIPH - Uses motif-based analysis for clustering
• ClusTCR - By far the fastest, uses a 2-step clustering approach to identify superclusters and then focused clusters.

I am not aware of clustering approaches available in R for specificity. I have implemented a tcrdist approach in my scRepertoire package, but it is still in the testing stage.

Let me know if you have any other questions!

Nick

[chi.delta]

Hey thanks a lot, this looks really helpful. An integration of tcr clustering in scRepertoire will be really game changing! thanks a lot for that, i will keep myself posted for your next update. any prediction for the timeframe? thanks!