Processing fastqs generated by inDrop protocol

Question

[bobbybobbobbo]

Hi all,

I'm trying to re-analyse scRNA-seq data from a recently published manuscript. The methods state that the data was generated using the inDrops protocol, but doesn't mention which version of the protocol was used, although it does cite the 2017 Nat Protocols paper.

There are 3 reads (R1, R2, R3) available for each library on SRA (e.g. see https://www.ncbi.nlm.nih.gov/Traces/study/?acc=SRX18198640), but as far as I'm aware the inDrops protocol generates either 2 reads (for v1 and v2) or 4 reads (for v3). Does anyone have any insight into how these files were generated and what should be done to process them into a counts matrix?

Any help would be greatly appreciated.

[benformatics]

i'm just guessing but there is an 8 bp read and a 16 bp read so my guess is those are the non-transcriptomic part (maybe a UMI + cell barcode? you would have to check the indrop library construction possibilities). the 61-bp is probably the actual mRNA section

[Ram]

You've cross-posted this on bioinformatics SE. Don't do that - a bunch of people are on both forums are it is bad etiquette to waste volunteers' time just because you're in a hurry.