Map TCR sequences from bulk mRNA seq

Question

[chi.delta]

Dear all,

I was provided with some sequencing data (FASTQ unmapped files) from an external collaborator's lab where they sorted T cells from PBMCs, isolated whole RNA followed by polyA selection and sequencing ( Ultra-Low Input RNA-Seq, Illumina HiSeq 2x150bp, single index)

Is there a straight forward way (preferentially R based) which I could use in order to extract the TCR sequences out of this pool of data?

The idea would be in the end to compare the clonotypic expansions across different conditions..

Thanks a lot for your time!

Answer 1

[theHumanBorch]

Hey Chi.delta,

I would recommend MiXCR to extract the TCR sequences and then the use of the awesome Immunarch R package - they have a specific vignette for MiXCR ouputs. If MiXCR doesn't end up working for you Immunarch also supports a number of different workflows.

[chi.delta]

Great, so MiXCR can use unmapped files to extract TCR sequences? Or would you recommend mapping first to the human genome and then proceed with the TCR? Is this possible or make any difference?

Thanks :)

[theHumanBorch]

Yes - MiXCR will align against a reference genome and it supports single or paired-end fasta/fastq.

[chi.delta]

Hey again,

I am a bit puzzled on what reference to use in this case. Is it possible to have a single fasta file with VDJ sequenes for both alpha and beta chains? I cant seem to find such a reference..

Thanks!

[theHumanBorch]

There is both mouse and human references built into MiXCR, you'll just need to indicate species. It will align for both chains as well.

mixcr align --species <species> [options] input_file1 [input_file2] output_file.vdjca

Check out more at the their documentation

[chi.delta]

OK great thanks! Mapping has worked fine. I was just curious about the fact that the top most abundant clones that popped up were B cell clones. However the sequenced sample was sorted CD8 T cells. Is this all false positives?