When TruSeq RNA 3.0 coming?

[jbrandim]

Version 3 is in the beta testing phase

I hear that it will be directional

Will it be a revised and repackaged Script Seq V2?

[pmiguel]

I heard nothing. But I hope we get 96 or 384 dual index combinations.
Since the TruSeq DNA HT kit has been released with 96 dual index combinations, it seems likely these could be used with the RNA kit. Just dilute them 30x, right?

--
Phillip

[jbrandim]

Tired of indexs

Until they find a way of reliably plexing samples anything above 6 indexes is superfulous. Would anyone in their right mind plex 24 RNA or DNA?

[pmiguel]

Not sure what you mean by "reliably plexing samples". Expand a little?

We do plex 24 now sometimes. Would like to be able to do more -- lots more. Mainly so we could do a single MiSeq run to get a read on what the correct amount of a pool should be loaded to get a given cluster density on all the lanes of a HiSeq run.

--
Phillip

[jbrandim]

Need for normalization

We do not have a MiSeq so for our multiplexing we have to rely on several non perfect measurements to pool libraries. Currently even our best practices still give us some libraries with 2-3 fold more reads than others. Most of the core facilities in my area won't even go above a 4-plex.

[pmiguel]

Ah, that. Yes, we have the same issue, but I thought it must not be a general issue. Doesn't seem to be discussed much here. 2-3x for the most outlying samples of a pool sounds good to me. Well, at least for large pools.

But I don't completely get why, for example, splitting 12 samples up among 3 lanes gets you a better outcome than putting all 12 samples in all 3. Is it that, in addition to the 2-3 fold differences you sometimes get major outliers -- like >5x high for one sample and that leaves you with very few reads to be split among all the rest?

--
Phillip