Oxford Nanopore MAP: quantitative transcriptomics?

[jchap14]

Anybody have an update on this?

Only thing I have seen is the Graveley lab at UConn using MinION to do RNAseq on RT-PCR amplified targets of interest (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4588896/).

@Azazel Did you try it?

[AllSeq]

Only data I've seen is genome data. The read lengths are pretty impressive.

You won't get a lot of coverage but you can do 48 hour runs and re-use the flowcells to increase coverage.

What do you think the upper limits are? When ONT first talked about the MinION, they said it could generate ~1Gb of data before it burned out, but I'm not sure if that limit is still in place. If it is, then it means you could get roughly 1M reads of 1kb length, 500k reads of 2kb, 100k reads of 10kb, etc. Would love to see some real world RNA results!

[NextGenSeq]

Only data I've seen is genome data. The read lengths are pretty impressive.

You won't get a lot of coverage but you can do 48 hour runs and re-use the flowcells to increase coverage.

[AllSeq]

It looks like each run is generating something on the order of 10s of thousands of reads, so you're not going to learn much in the way of relative expression levels. However, if you're careful with your prep to get truly full length cDNAs, you might get some interesting info on splice isoforms (similar to what PacBio is doing with their ISO-Seq application). If you end up doing it, I hope you come back to share the data - it should be quite interesting!

[Bukowski]

Technically of course you can. Whether you would want to is another matter. I doubt you will get much from a minION EAP to do anything other than (as you point out) sequence lots and lots of rRNA.

Of course you could pA+ or ribodeplete first, but still these machines do not generate a lot of data (yes I am in the EAP) compared to what I would consider to be required for meaningful mammalian transcriptomics.

I think the bigger brother machine is going to be very interesting. It will be nice to not have to reconstruct isoforms from short-read data.

[Azazel]

I have the option to join the Oxford Nanopore MAP. What I want to do is use MinION for (quantitative) transcriptomics, i.e. I want to reverse transcribe RNA to DNA and then sequence on MinION. This is on mouse.

Does this make any sense with respect to the capabilities of MinION?

I can't find anything on number of reads, sequencing depth... the documents I have read so far all seem to assume that I want to do genome (re-)sequencing, where - as I understand - depth is less important than length.

So can I do transcriptomics on MinION? Will I get anything except the most highly expressed genes and ribosomal RNA? Does the sequencing depth somehow depend on runtime? Also how does the speed of sequencing (bases per second) affect things?

I understand that there is not much info on MinION out there yet and the point of the MAP is probably to generate experience on this, but I would be happy if I could get ANY info on performance before I decide to join the MAP. Thanks.

tl;dr: Can I do transcriptome sequencing on MinION?