Single Cell RNA-seq Troubleshooting

[tmm343]

Hello All,

We are trying to perform single cell RNA-seq in our facility. We would like to do this in a high-throughput manner on our robotic system and are working off of this published protocol:

http://www.nature.com/nbt/journal/v33/n5/full/nbt.3192.html#/methods

We wanted to and have done the protocol in hand first to work out any issues before conducting the experiment on the robot. We are using a Universal RNA Standard diluted down to a single cell concentration (10pg/ul) for testing.

We have not been able to get past the cDNA stage as we are seeing a strange BioA trace when comparing to that in the troubleshooting table of the Picelli Paper: http://www.nature.com/nprot/journal/....2014.006.html.

Attached are our BioA traces as well as TapeStation results of the same sample that also yield inconsistent results.

We have played with concentration as well as PCR cycles and still see this "double-hump" in the traces.

Any advice or thoughts would be greatly appreciated!

[plyguy69]

try changing RT enzyme

We are currently going through something similar but the nonspecific peak is shorter. We are trying to compare different RT enzymes to fix the problem.

[seq198]

We are currently going through something similar but the nonspecific peak is shorter. We are trying to compare different RT enzymes to fix the problem.

Have you been able to find a solution? We are seeing also high levels of smaller peaks.

[seqgirl123]

Have you been able to resolve this issue? I would try each of the following: 1) use a different RT, 2) clean up the cDNA with SPRI beads (you can size select the cDNA) and proceed into library prep, and 3) take it all the way to the end (make libraries) and see if the library profiles have outlier peaks, even here you can size select purify them out prior to sequencing. It looks to me you have some dimer formation during cDNA synthesis. I would actually try #3 first, make libraries of what you have as the library making process has so many clean ups, the issue could resolve itself.

[dwong]

Not sure if you have a solution yet but I have seen this double hump phenomenon before. Usually what causes this is gDNA contamination or overamplification of your libraries. Try using a solution based DNAse treatment on your samples (on column DNAse treatments seldom result in gDNA free preps). Also bear in mind if you are sorting your cells in PBS and are doing DNAse treament, the salts in the PBS can interfere with the gDNA digestion. There also seems to be some primer dimer in your libraries. As seqgirl123 suggested, I would do a double SPRI bead cleanup. That would clean up your libraries.

[tmm343]

We have not found a solution and are picking these experiments and troubleshooting back up very soon! Will keep all updated.

[plyguy69]

We are still troubleshooting too

Hey guys,

We are trying to switch out our RT enzyme. Just tried the Maxima evolved RT from Thermo and the results for cDNA look promising. Need to carry these libraries through sequencing though to see what was generated. I'll keep you updated.

Thanks for feedback!