BCR/TCR analysis using target capture sequencing data

Question

[J.F.Jiang]

Hi all,

MIXCR is widely used to analyze the TCR/BCR sequencing data, which usually use the multiplex amplicon sequencing method.

We have design the 120bp probe to capture all the genes across IGH/IGL/IGK and TCR locus.

Since the latest mixcr v4 used the preset to analyze the data, which command should I use to analyze these data to get the types and percentage of the most enriched alleles ?

Or should I use igblast to get the results?

I am completely unfamiliar with this aspect, it will be of great help if someone could share some experiences.

Best, Junfeng

[Jesse]

I still haven't tried MIXCR so I can't really answer your question itself, but, IgBLAST could give you a rough estimate of what you're looking for (see -outfmt 19 in the command-line version and then the v_call/d_call/j_call columns from that). You have to be careful about definitions of things when trying to figure out abundance with methods like these, though, as in, what does a single sequence and set of allele calls represent (amplicon/transcript/cell/...?) I've found that's a tricky question with bulk repertoire sequencing, especially if you're not using UMIs. (Much clearer with single cell!) You also have to trust the automated calls, which can get iffy with the more mutated sequences (especially for D; that'll be guesswork for a good chunk of your more-mutated sequences).

And just to make sure: when you say capture all the genes for the loci, do you mean you're sequencing amplicons from RNA transcripts of the recombined sequences, or doing genomic sequencing of the relevant loci pre-recombination, or sequencing the recombined (genomic) receptor sequences? Or something else?

[J.F.Jiang]

Thanks, Jesse

We designed the probes based on the genomic sequences of V/D/J genes within the TCR/BCR loci.

I've tried both IgBlast and MixCR, which gave similar alleles and fraction for the V/D/J genes.