Adaptive T cell receptor productive clone missing

Question

[bioinformatics.cancer]

Hi, I am analyzing T Cell Receptor sequencing data received from Adaptive Biotechnology which consists of summarized clonotype data which includes NT and AA sequence among many other variables. Some of the clones don't have an AA sequence but have a NT sequence. I believe these are called "unproductive" clones. I was wondering if anyone knows why some clones don't have an AA sequence (are unproductive) and more over how does Adaptive determine which clones don't have an AA sequence (what is their algorithm for this). Some of these clones that don't have an AA sequence have quite high frequencies, so I am particularly surprised that these don't have an AA sequence. One example of such as NT sequence is:

CAGAACTGGAGGATTCTGGAGTTTATTTCTGTGCCAGCAGCCTTTACAATGAGCAGTTCTTCGGGCCA

Thanks, - Pankaj

Answer 1

[mikhail.shugay]

This is a productive sequence, just load it to IgBlast and see for yourself:

' TRBV14 is productive and in frame with J

I think the situation can probably be explained as follows: the mapping software checks for conserved FGXG motif of Joining segment to ensure that J segment is mapped and as the last G is not in the read so the clonotype is not translated.

I've implemented a parser for ImmunoSEQ data a while ago in VDJtools: executable, docs. It converts it into a more compact format and checks if the rearrangements are truly non-productive, translating CDR3 sequences. I haven't tested it much so if anything goes wrong feel free to write in the issues section of the repository. Note that the option for conversion with VDJtools/Convert is either -S immunoseq or -S immunoseqv3 depending on the version of Immunoseq software.

---

PS. It seems to be that it is a public clonotype: https://www.google.com/search?q=CASSLYNEQFF :)

PPS. Marked as Rasmussen encephalitis-specific clonotype http://www.nature.com/articles/ncomms11153 found in 0% control samples, possibly due to HLA restriction.