HiSeq 3000/4000 2x150 data

[scrosby]

Hello wise ones,

Has anyone out there had any paired-end long read experience. We'd love to see the run data (Q30 graphs, etc).

S

[luc]

I do not believe there are any 3000's/4000s out there in the wild yet.

[Brian Bushnell]

I would not expect any useful data from "Q30 graphs"; my understanding is that Illumina is currently planning mandatory Q-score quantization on the new platforms, as they do on NextSeq, which will render the quality scores mostly useless.

[GenoMax]

There are some examples of HiSeq 4000 data in public area on BaseSpace. Don't know if they are long read since the description does not say anything specific.

[scrosby]

Checked BaseSpace already. Just 2x76.

[Leighb]

We just received our HiSeq 4000 today and it will be a couple more months until we can get validation runs going.

[Genohub]

Would you be open to listing your HiSeq 4000 on Genohub.com? There are many researchers hoping to get access. Are access to flow cells still a limiting factor?

[GenoMax]

@scrosby: UC Davis lab had posted some 3000 data. http://seqanswers.com/forums/showthread.php?t=58353

[QazSeDc]

As I believe many companies are still undergoing 4000 validation testing, but so far the graphs on 4000 raw data that i have seen are all pretty nice and stable beside the duplication problem everyone has already mentioned.
And by the way I think Q20 is enough to determine the quality of the sequence in most cases.

[NextGenSeq]

Increased number of duplicate reads, preferential amplification of small fragments and shorter sequencing libraries?

Is this worth 900K?

[Brian Bushnell]

And by the way I think Q20 is enough to determine the quality of the sequence in most cases.

...no, absolutely not. That's like determining the flavor of a bottle of wine by reading the label.

Increased number of duplicate reads, preferential amplification of small fragments and shorter sequencing libraries?

Is this worth 900K?

Not to me; I'd prefer the 2500, except when you need immense volume of short-insert overlapping fragments. For example, a time-and-space-series deep metagenome, or when sequencing a country's population...