How to identify translocation events within cell lines

Question

[shawn.w.foley]

Hello!

We have several cancer cell lines that have been reported as harboring multiple translocations, and I'd like to know the best ways to test them and verify these results. Recent papers have shown that tissue culture cells are less clonal than the field has assumed, so before investing a lot of time in studies we want to confirm that ~10 cell lines are harboring the translocations that have been reported previously (especially since there are publications with contradictory translocations reported).

We'd like to do this analysis informatically. Ideally we'd like a targeted panel for sequencing or even microarray/qPCR (performing FISH is outside of our realm of expertise). I'm imagining something analogous to whole exome-seq, where we're enriching for specific regions of the genome then sequencing.

Alternatively, if whole genome sequencing is considered the best option, what sequencing depth and analysis pipelines would you recommend to identify high confidence translocations in cell lines that may be polyploidy?

Thank you very much for the help!

[ATpoint]

Are you restricted to Illumina sequencing or is Nanopore long read sequencing an option?

[shawn.w.foley]

Nanopore would be a great idea, but we're restricted to Illumina.

Answer 1

[Vitis]

If you're looking for de novo (new and unknown) translocations, I think approaches with longer reads would be better. It also depends on the types of translocations: if most of them are un-balanced, you will be able to use coverage and copy number to find them (in this case, Illumina with sufficient coverage, even as low as 10X would do). The hardest type is the balanced translocations, for which coverage would not help you and you probably need long reads to identify the breakpoints. Even in optimal technical conditions, there still could be translocations happening around repetitive regions, for which all methods would have a hard time identifying.

Here is an example in human fertility studies.

https://www.biorxiv.org/content/biorxiv/early/2018/09/18/419531.full.pdf

[shawn.w.foley]

Thank you for the advice. In this case we're looking to verify reported translocations, so we do have a priori loci to test. Long reads likely won't be an option for our in-house sequencing core, but I'll look into it.

[Vitis]

Then I think a panel of short amplicons covering the breakpoints would be the best option. Amplicons of 200~300bp size with PE150 sequencing could give you both high coverage and multiplex level, and can be easily scaled up.

[shawn.w.foley]

Are you suggesting a pool of primers, or are there panels for common/known breakpoints available? Thank you!

[Vitis]

Yes. Design primer sets surrounding known breakpoints, preferably at least two sets for each breakpoint. Then do multiplex PCR and sequencing. The multiplex PCR may need some tunings to make all amplicons work together, which might be the trickiest part. But it is normal to see up to 50~100 amplicons working in one pool without any problems.