PacBio read lengths.. How long?

[bkingham]

Lately we've been running a number of 16-20kb libraries using XL enzyme and C2 chemistry. We're getting 30-35kb reads, and 25kb subreads... Only a few per run, but they definitely improve assembly. The cluster of subreads in the 10-20kb range has gone up considerably also.
I'm curious what other machines are producing?

Thanks

[jbingham]

Wow. That's pretty amazing. This is the first report I've heard of a PacBio system out there getting reads that long at the extreme tail.

[GenoMax]

bkingham: Can you post the run time length (# of movies) you are using for these runs?

[Derekp]

Binkham, we have ran some XL enzyme with XL chemistry and saw some pretty nice data as well. A few reads approaching the 30K length like you are seeing. Lots in the 9-15K range. You are running the XL enzyme with the C2 chemistry. Have you also ran XL/XL and how did that compare?
Looks like this pab bio data is just getting more and more useful

[bkingham]

The best have all been 1x120' movie, hot start with XL enzyme/C2 chemistry/C2 SMRT cell, 16kb library. We haven't tried XL/XL yet... soon.
It is starting to generate some impressive data

[flxlex]

We've tried both C2XL and XLXL, and are not yet getting beyond 30kb, but close (with XLXL). Unless you are talking raw-raw (before filtering for high-quality regions)...

[tacana]

Our results are similar to what flxlex reports. We have tried XLC2 and XLXL using 1x120 movies and got about 20kb-25kb Maximum Mapped Readlength for the XLC2 and 25-30KB for the XLXL (Maximum Mapped Readlength again)

bkingham- are the values you are reporting also Maximum Mapped Readlength?

[jlj]

What are you using for size selection?

[jlj]

Can you share your on plate loading concentrations? We just ran a 20kb library and got very poor ZMW productivity. In my experience from other (shorter) libraries this was because of sub-optimal on plate concentration. We actually have doubled PacBio's recommendations and achieved great results.

[tacana]

At least for us, we just started using Blue Pippin for size selections.
In regards to loading it seems to vary from library to library but for large libraries (>10kb) we have gone as high as 0.06nM on plate concentration

[jlj]

We are also using BluePippin. I did a titration that PacBio recommended in their 20kb protocol using 150pM, 200pM, and 250pM on plate concentration. Our productive ZMWs were very low. We got >85% empty for each cell and 3, 4, 5% singly loaded for the corresponding titrations. I'm currently working with Pacbio to figure out what went wrong, but any suggestions are helpful!

[Dave_Baker]

I'm a bit confused, tacana, you have said you go as high as 0.06 nM (equiv to 60 pM) and jlj is using the recommended 150 pM from the 20kb protocol for on plate concentrations. I've checked the 20kb protocol and it is suggesting loading almost 10-fold higher concentrations, and as you say run a titration from 150pM-250pM, almost as if it were loading as diffusion. I'm very confused.

[tacana]

Sorry for the confusion that I may have cause. You are correct for 20kb fragments that have been selected with Pippin we also load 150pM -250pm. For the data that I posted, which was not Pippin selected, I loaded at 0.06nM (60pM).

[Dave_Baker]

My FAS has just confirmed the reason why they recommend to load a 20kb Pippin size-selected library at such high on-plate concentrations is because the larger fragments do not load as well as the smaller ones. This isn't stated in the protocol which can confuse people. Seems strange that a non-size selected library of the same size has a loading concentration of 10 x lower.

[Dave_Baker]

Thanks. All makes sense now. I posted my last post before i read your reply.