Library input on the cluster station GA2

[runemoe]

Hi!

Does anyone know the optimal input on the cluster station for GDNA PE Seq?
The protocol recommends 1-4 picoMolar (pM) but we have had more success/more sequence loading 6pM, and I am a bit tempted to go even higher...

Does the optimal input vary much between protocols (both type and read length)?

[der_eiskern]

how much are you loading these days? i've heard of some people having to load 48 picomolar gDNA for PE Seq.

I'm also trying to figure out how much Illumina says you SHOULD be having to load to get optimal cluster density (on a GAIIx machine).

So everyone please comment, a consensus among the user community would be great!

_der.

[runemoe]

Illumina says that optimal input varies from instrument to instrument, and among projects. We have now loaded up to 12 pM, and 8-10pM seems optimal for gDNA PE seq (GA2x). Loading 12 pM lead to strange intensity plots. We have also had a few problems with the laser and the mode scrambler, so I'm a bit worried that adjusting or replacing the laser/ mode scrambler would affect our conclusions...

It would be nice to know what other users load on their standard GA2 run!

[der_eiskern]

Illumina says that optimal input varies from instrument to instrument, and among projects.

rune, Any idea how wide the variation they say there is?

we haven't run enough samples on our GA2x to know it that well yet.

[basickler]

You're right, the right cluster density depends on library prep and can vary quite a bit. I know we load anywhere from 3-18 depending. You can use qPCR to get a better indication if you trend qPCR and the cluster density from the Summary.html.

As for the right cluster density, that depends on what version of SCS you're using.

Brad

[runemoe]

rune, Any idea how wide the variation they say there is?

der, I am not sure, but it seem to be in the interval 6-24. Like Brad says, it is really important to know the accurate concentration of your library. For us the Qubit (invitrogen) works fine, and we are now optimizing a qPCR protocol to be more accurate. We have also validated some of our libraries in a more primitive way; by traditional PCR followed by gel electroforesis, checking at which cycle the product pops up compared to a successful library (from an earlier project) and the PhiX.

[KPR]

Hi,

why not use a Bioanalyzer to estimate the DNA concentration of your library? Does anyone has experiences?

[der_eiskern]

i've tested the bioanalyzer versus QPCR for quantifying the library and they're roughly equivalent in my hands.

[KPR]

i've tested the bioanalyzer versus QPCR for quantifying the library and they're roughly equivalent in my hands.

Good to hear, der

We are trying to compare Qubit, Nanodrop-Picogreen and Bioanalyzer for DNA quantification after DNA enrichment. Sometime we have problems to get enough reads even working with 4pM cluster solution.

[der_eiskern]

4 pM was too little for our GAIIx machine. 8 pM gave us the so-called "sweet spot" in the 180,000-220,000 clusters/tile range.

i've compared all those but Qubit, which is just a fluorimeter, right?
when dealing with low concentrations less than 30ng/µl i found the nanodrop overestimated my sample concentration. in my hands picogreen wasn't as effective as QPCR or bioanalyzer. let me know what you guys find what works for you.

-der

[runemoe]

8 pM works great for us as well, sometimes even 10pM (if the library quality is not super). We use the Qubit for quantification of our library, then we dilute this to 10nM. Next we run a qPCR comparing the new library to the PhiX and one good and one bad library we have already sequenced. If the library performs as well as the PhiX or the good library we continue, if it doesn't we discard it and reprep the library.

We have also had overestimating problems using the NanoDrop. The Qubit (yes, it is a fluorometer) is not so sensitive for contaminants like solvents and salts etc. The BioAnalyzer is not so good at quantifying, but it is more useful to get an accurate size estimate of your library, especially useful in PE runs.

[btarrier]

We have had to adjust our concentration up over time (all on GAIIx) from around 6pM to 10pM based on declining cluster numbers and recommendation from Illumina. I'm not sure what has been changing, but our standard range from 4-8pM is now too low. We are in the 8-12 pM range now.

[upenn_ngs]

qPCR to quantify whole exome

We have become interested in using qPCR over the bioanalyzer or qubit for concentration. We sequenced 2X100bp reads of exome captures and the qubit concentration produced 4-5Gb/channel across the flowcell with avg. cluster densities ranging 250k-310k.

qPCR will give a stronger measure of fragments that amplify on the flowcell vs. an intercalating dye. Any experience optimizing this method for whole exome captures?

[Eric@Kapa]

FYI - We (Kapa Biosystems) offer ready-to-use qPCR library quantification kits for the major NGS platforms.

Each kit comprises a set of six validated, quality-controlled DNA standards representing a 10-fold dilution series, qPCR ready mix, primer mix, and a recommended protocol. Our qPCR reagents are built around an engineered polymerase optimised for qPCR via directed in vitro evolution, allowing longer and more diverse targets (i.e. NGS libraries) to be amplified with high efficiency in qPCR.

I have attached some product literature in case you'd like to take a closer look.

Eric

[upenn_ngs]

This is exactly what I was looking for, thank you.