PacBio ZMW Loading Productivity 2

[HeidiJTP]

I recently got back my first data set from a PacBio run and was wondering what are 'typical' values for the different productivity levels? Here's what I received:

4.8% ZMW Loading for Productivity 0
38.6% ZMW Loading for Productivity 1
56.6% ZMW Loading for Productivity 2

The productivity 2 value seemed pretty high to me - these will mostly be unusable reads unless one of the polymerases drops off pretty early in the run, correct?
Could this indicate anything about my sample prep?

Thanks for any insight

[rhall]

these will mostly be unusable reads unless one of the polymerases drops off pretty early in the run, correct?

You will recover very little usable sequence from ZMWs with Productivity 2, if any.

It may be a little overloaded, 56.6% is quite high, but 38.6% is actually pretty good given the statistics of loading. Better metrics to look at are the number of bases, read length and read quality, preferably after a simple filter. This will give you an indication if the overloading is having a large effect on the Productivity 1 ZMWs too.

[HeidiJTP]

Thanks for the feedback rhall.

I think the post-filter read # aligns pretty well with 38.6% productivity 1 (~75k pre-filter reads, ~29k post-filter). Readlength mean is 3,701 bp and quality is 0.849.

[rhall]

Those results look fairly typical. Although there is some room for optimization.

[tacana]

rhall- you mentioned that there is room for optimization. What would you consider an good loading distribution of ZMWs?

[rhall]

50-55% Probuctivity 1 would be excellent for a short (2kb) insert library. The distribution between 0 and 2 may depend on the instrument software version, probably best talk to someone familiar with the particular instrument.

[bkingham]

The # of productive ZMWs is influenced by the size of the library. We regularly get 50-55% with 2-6kb libraries, for 16kb libraries we can't get above 33%.... Lower than I'd like, but the reads are very high-quality. The recent upgrade tweaked our system so that we see considerably more productivity 0, considerably less productivity 2, same productivity 1.

[rhall]

A good place to start if you want to look at optimizing loading, or are just interested in the many metrics that are recorded during sequencing:
https://github.com/PacificBiosciences/stsPlots

[HeidiJTP]

Thanks again for the information rhall, I appreciate it. By the way, this was for a 2kb library.

[raibiotech]

hi,

One 0f my experiments is giving me loading values like these:
P0=16%, P1=12% and P2=72%(1Kb library)
P0=12%, P1=18% and P2=70%(3Kb library)

Does anyone encountered anytime these kind of Loading values?
What could be the cause of this?

Any help or suggestion is appreciated

[rhall]

This would indicate that the on chip DNA concentration is too high. Values like these would result in very poor data quality.

[raibiotech]

Thank you so much for your reply ... when I contact tech support they said it is too much unbound polymerases there so P2 is high.
My doubt is polymerase will not give any signal alone from ZMWs there has to be some DNA fragments also... is there any possibilities of errors during samples prep like Damage repair is not good or Polymerase was added too much??

[tacana]

Both things, cell being overloaded or having to much unbound polymerase, can cause the phenotype that you show. Have you tried lowering the concentration of your DNA? I think that this is a first good step to trying to figure what what is going on. If you lower the concentration and you still see the same results, then you may have other issues. What is the concentration and size of your library?