Clonal amplification/ Kinetic Exclusion parameters

[Dornfield]

Reposted from Library Genertation Forum, where I somehow managed to mispost it...

Newer generation Illumina platforms with patterened flowcells use kinetic exclusion to achieve a high degree of clonal chastity. From what I understand, the amplification method is still bridge, but is driven by isothermal RPA, so there is no need for the chemical denaturation and renaturation steps of old.
My question is, can anyone comment on the final number of copies of the clonally amplified material within any given cluster? An although it possibly differs from instrument to instrument, what is the feature size of any given patterned location: what the density of the amplified molecule would therefore be?
Does the signal generated go through a 'photomultiplier' prior to analysis, which I imagine would reduce the need for a strong raw signal, and so would better accommodate a lower level of amplification/ template density during the clonal amplification?

Thanks for any insight.

[Dornfield]

Perhaps I could come at this from a different direction (or invite educated guesses?). I am not actually an Illumina user, so I'm not familiar with the workflow that generates the clusters in the patterned flow cells: how long (minutes) does this step take, or is this a variable dependent on the insert length? Does this length of time give anyone a clue as to how many copies there might be per cluster?

Thank you for any input.

[GenoMax]

If you look at the images in the patent for patterned flowcells you can get some idea of size of the "features". Someone can chime in about the actual time required for the ExAmp part.

@nucacidhunter has posted a couple of links about ExAmp in this post.

[Dornfield]

Thanks. I had seen that patent before, but not thought to do some very back of the evenlope calculations:

250nm diameter features means a surface ares of ~49,000nm^2. My rule of thumb for a DNA polymerase is to think of it as being globular (think spherical) and about 5nm in diameter, so ~19nm^2. So, worst case you could have a lawn of DNA polymerase on the surface of the feature of about 2500 DNA polymerases, before steric interactions become really bad. So even if the lawn of P5 and P7 oligos was denser than that, I'm tempted to think that each feature would be hard pressed to cater for more than a couple of thousand template strands before templates start preferentially annealing to each other, and enzymes start fighting for space.

2500? Which was kind of my ballpark/ gut feel.

It would still be good to know how long the ExAmp step takes in the real world.

[GenoMax]

What is described in the patent and what happens in practice is anyone's guess. Much of the process likely remains a trade secret (exact chemical modification used etc).

It takes about 3 hours to cluster a 8 lane HiSeq 4000 FC on a cBot.