Direct sequencing of mRNA for RNA-seq

[clostridium40]

I was told today by our collaborators that they have found that directly sequencing the mRNA gives them better results then converting to cDNA when they do RNA-seq. Has anyone else found this to be true? Every paper that I have read on RNA-seq converts the mRNA to cDNA, was just curious if anyone else had tried directly sequencing the mRNA.

[pmiguel]

Just a point of clairification: unless your collaborators have early access to some leading edge tech or, more likely, have access to one of the handful of Helicos sequencers, my guess is that they are not "directly sequencing mRNA". Even the proof of principle RNAseq paper from the Broad was about directly converting RNA to cDNA in the flowcell of and Illumina did not "directly" sequence that RNA, but rather the cDNA created in the flowcell.

To "directly" sequence RNA you would need an enzyme that could use RNA as a template (probably reverse transcriptase). In principle this is possible (some of the earliest sequencing templates were RNA).

It is possible that your collaborators have removed one of the intervening layers of conversion/amplification and are calling that "direct" sequencing of the RNA. Here are the layers as I would define them:

(A)Cellular RNA in situ
--break cells isolate RNA
(B)Total RNA
--ribo-deplete or mRNA purify
(C)mRNA
--cDNA conversion
(D)cDNA
--library construction: end polish, adapter ligate
(E)amplicon library (zero amp)
--enrichment PCR
(F)final library
--emulsion PCR/bridge PCR
(G)templated beads/clusters

By this scheme all commonly used next gen sequencers use layer "G" templates. The exceptions being 3rd gen PAC bio and Helicos instruments. Helicos does apparently have a direct RNA sequencing methodology. A Helicos instrument would operate at layer "B".

How many Helicos instruments are in the field? A dozen?

Anyway, I just wanted to add some context...

--
Phillip

[john_king]

Direct sequencing of RNA is available as service from Helicos.