Understanding gene usage

[benjytan88]

Hi MiXCR developers,
I wanted to ask what is the meaning of multiple TRBV genes detected in a sample? For example, what is the meaning of TRBV3-1, TRBV14, TRBV3-2, TRBV4-2 in sample K169_N1t1 in the table below?

[PoslavskySV]

Hi, could you please elaborate how did you produce this table? It does not look like output from MiXCR.

[benjytan88]

Hi @PoslavskySV, that was a table which summarize gene usage in all my samples. Here's the output from MiXCR.

As you can see, there are several rows which have more than 1 V gene called. I would like to understand what are these and how to interpret these.

[PoslavskySV]

Thanks! Could you please paste exact commands you used with MiXCR? The names of the columns look quite different from the standard MiXCR output.

[benjytan88]

Command as below. This data is from bulk RNA-seq and MiXCR v4.3.2 was used.

mixcr analyze rnaseq-full-length --species hsa ./K548-M20_R1_001.fastq.gz ./K548-M20_R2_001.fastq.gz ./K548-M20

[PoslavskySV]

Did you rename columns manually? The column "V.name" is probably "AllVHitsWithScore" from MiXCR output. This column contain all V hits with corresponding alignment scores. E.g. the value in cell:

TRBV12-3*00(240), TRBV12-4*00(240) 

means that the read was aligned to two V genes: TRBV12-3 and TRBV12-4 with equal alignment score (240). You can use "bestVHit" or "bestVGene" or "bestVFamily" to have only one (best) hit in the output (see export docs).

[benjytan88]

I see. OK, I will try and see.
I didn't rename them manually, but the table was obtained by loading the MiXCR results into R using the immunarch package.
Maybe there's some renaming done from there...?

[PoslavskySV]

Yes, this is 100% due to immunarch. You need to check how it converts columns.