"GGGGGGGG"I7 index read in my scRNA-seq library

Question

[Sasaki]

Hello~ When I was analyzing scRNA-seq data generated from my customized platform. I always found "GGGGGGGG" I7 index read in the header. I designed the I7 index read as my cell barcode, so they should be different 8bp sequences. However, when I was checking my data, for example, if I get ~100M reads, there would be ~60M reads with GGGGGGGG as the i7 index read. This is pretty confusing and frustrating because those reads prevented me from getting enough sequencing depth for every single cell. Does anyone have a solution or explanation for this? Thanks a lot for the help!

Answer 1

[GenoMax]

Poly-G represent no signal in 2-color chemistry, so if the remaining 40M reads actually have good sequence then it is likely that this run is partially failing on i7 index. One thing to try would be to under load the sample. If none of the reads have usable i7 sequence then it would be bad news. You will need to check your adapters/library prep.

[Sasaki]

Thank you so much for your help. But what does "under load the sample" mean?

[GenoMax]

If you end up having to re-run this sample (because you are not able to get usable sequence) then reducing the loading concentration that is loaded on the new flow cell would be something that may need to be considered (assuming your libraries have been made properly).