Barcode extraction and clone counting

[crist156]

I’m finding the CellTag system very easy to use (thank you!), but have run into hopefully a few small but solvable problems.

Pertinent information – I am using the CellTag V2 library and performed bulk sequencing on my cells to confirm that the barcode complexity of my cells matches that of the starting library. Having done this, I’m not concerned that my issues below are related to reduced barcode number after transduction. I performed single cell RNAseq on three samples – a day 0, day 7 and day 30 timepoint and am currently working on extracting/analyzing the CellTags. My cells had been transplanted in mice, so in order to reisolate them, I sorted for the GFP-expressing cells (thus also still retaining our CellTags).

Issue 1 - When I execute the barcode extraction for my Day 30 sample, I lose ~75% of the cells sequenced. This is in contrast to my Day 0 and Day 7 samples, where the presence of CellTags does filter out some cells, but nowhere near the majority. I'm confused why there would be so much loss in this specific sample when the number of cells sequenced are similar between Day 0 and Day 30, as are the file sizes (rough proxy for the reads) when I've executed the mapping steps just prior to extraction. I'm also not as worried about experimental loss of barcode because these cells were sorted off of the GFP transgene where the CellTag exists, so they should in theory all have at least one CellTag. Any thoughts on why the extraction might be faulty? Or how I might be able to check/troubleshoot this?

Issue 2 - I seem to have some sequence being counted in almost all of the cells. Any idea where this is coming from and how to remove it? I'm guessing this is an error on my end (mapping error? whitelist error?) since it would be incredibly unlikely that one single barcode exists in all of the cells. Have you ever seen something like this before?

Issue 3 - The number of clonal families that I find in my samples is quite low, but I also notice that these families can't exist as single cells (always have 2 or more cells). I think it's possible that there do exist single clones within my population of cells and at present I'm filtering these out. Is there any way to edit the code to allow for this possibility?

Thank you so much for your time and help! Happy to clarify any of the above. Any and all suggestions are incredibly welcome!

Thanks, Sarah