Nextera DNA library - low yield?

[sweetph3]

I just ran the Nextera DNA sample prep for the first time, following the protocol with the exception of using a Zymo column instead of the spin plate. Input was 50 ng genomic DNA (nuclear and chloroplast, resuspended in 10mM Tris-Cl pH 8.5, integrity checked on agarose gel). The final yield was 0.5 ng/ul in 30 ul final volume (15 ng total library yield) as measured by Qubit. I can't find any posts or other indications of what a typical yield should be, but this seems low to me. I have not run it on the Agilent Bioanalyzer yet, as I am waiting to receive another DNA sample from a collaborator and run several samples at once.

Can anyone give me an idea of what a typical good yield is for Nextera DNA libraries? And any hints on what I could have done wrong if this is indeed low would be helpful, too.

[Innovelty]

I was told that typical yield should be between 3 and 8 ng/ul.

I don't know what caused this problem, because I have exactly the same problem. In my case, it was 50 ng good quality cDNA input, suspended in pure water. One library came out at 0.7 ng/ul, and the other 0.5 ng/ul. I had problems earlier like this because (as it turned out) I actually had been under-loading the input DNA. But I fixed that, and had gotten one library made last week that worked beautifully, though the concentration was only 2.96 ng/ul. I have no idea what went wrong this time.

- Possibly the AMPure purification? Is it possible that, if I had over-tagmentation, I would lose a lot of shorter fragments when I did the 0.5x volume protocol for the MiSeq 2x250?
- Is the Zymo column very different from the spin plate? Should I modify the spin steps to account for it?

And why, oh why, is the NPM the limiting reagent? It feels so sad to have little aliquots of enzyme left and no NPM.

[Simone78]

I don´t know your problem in detail but I would never use 50 ng for tagmentation. That´s the upper limit and the Tn5 transposase won´t be able to cut it all in 5 mins at 55 degrees, so you will end up with a lot of uncut (and unusable because doesn´t have adaptors at both ends) DNA. With the Nextera kit I never use more than 5 ng and with the XT kit I try to stay far below 1 ng of input (100pg is plenty if you amplify 12 cycles).
I wouldn´t be worried about the beads. A 0.5:1 ratio won´t get rid of your short fragments. The Tn5 has a lower limit of about 300 bp, which means that you won´t get shorter fragments the more enzyme you use (or the longer you incubate). That´s how the enzyme works. It cuts the DNA and attach the adaptors. Once it´s "unloaded" it´s inactive. I use 0.6:1 ratio and get nice libraries of around 300 bp.
Actually, to save money and time you could skip the Zymo columns altogether. Just add the NT buffer from the XT kit (and set up the tagmentation in 20 ul instead of 50) and go on with the PCR. Your libraries will show broader peaks but the quality of data you get won´t change. When I start from 5 ng I usually get conc of 5-8 ng/ul (tot elution vol is 15 ul) when using the Nextera kit.
I can´t write all the details here but very soon we might have a paper accepted where we explain how to perform ALL the steps of the library prep (from single cell isolation to final library) at a fraction of the actual cost and without buying ANY kit (Nextera, SMARTer, etc). check it out!

[FWOS]

Nextera Rapid fixed the issues I've had

The Nextera Rapid kit fixes issues with tagmentation efficiency using 50ng input gDNA. I have toyed with the number of PCR cycles required to yield sufficient .tagmented libraries for Exome prep following Nextera library prep and always yield >600ng. Average yield after 7 cycles of PCR is 1ug.

Also, with the new kit, Illumina has 3x more enzyme which creates a beautiful BA profile (attached) that shifts dramatically when input is mis-quantitated or purposely underloaded. I have attached Bioanalyzer 1000 chip traces of 25ng input, 50ng input and 100ng input gDNA, all underwent 7 cycles of PCR, which shows how well the assay performs and how sensitive it is to input.

Hope this helps! I agree that the old Nextera kit was suboptimal at best, but it looks like the adjustment in Enzyme concentration is a big help.

Not sure if the XT protocol has also been updated, but have heard that most people find .5ng input to be sufficient. If you use the XT kit, remember that your final libraries are single stranded and can not be quantitated via Picogreen. You can either heat denature and run ribogreen, or stick with Kappa qPCR (make sure to calculate molarity based on ssDNA instead of dsDNA)

[rwinegar]

With the Nextera kit I never use more than 5 ng and with the XT kit I try to stay far below 1 ng of input (100pg is plenty if you amplify 12 cycles).

When you use 100 pg input with XT do you scale back the volume or the amount of ATM or is the DNA quantity the only thing you change? Thanks so much.

[Simone78]

When you use 100 pg input with XT do you scale back the volume or the amount of ATM or is the DNA quantity the only thing you change? Thanks so much.

never tried that. I use the same amount of enzyme and still get a peak with avg size of 400-500 bp. Actually, I use the Nextera XT kit just to compare it with the modifications I made to the protocol. Therefore I don´t want to introduce yet another variable and run the XT protocol as the manual suggests

[sweetph3]

I had some sort of inhibitor in my genomic DNA. I finally got some library after cleaning it up with MoBio Power Clean DNA Clean-Up Kit. My yield was 7.20 ng/ul. I'm also finally getting fragments detected on the Bioanalyzer ~300bp to >1kb using this cleanup kit. Other kits and protocols to remove inhibitors didn't work. I'd recommend using this MoBio product if you are having similar problems. Thank you all for your advice and suggestions! It's also good to know that I can safely lower the amount of input DNA and get good library, some of our collaborators do not provide much in their samples.

[FredericRaymond]

We had some problems with failling Nextara libraries. We found out that some zymo columns (almost 1 out of 10) were not well packed and that the liquid just went through the columns without binding. After we started controling visually for that, our library prep fail rate dropped to 0.

[wjyzidane]

I don´t know your problem in detail but I would never use 50 ng for tagmentation. That´s the upper limit and the Tn5 transposase won´t be able to cut it all in 5 mins at 55 degrees, so you will end up with a lot of uncut (and unusable because doesn´t have adaptors at both ends) DNA. With the Nextera kit I never use more than 5 ng and with the XT kit I try to stay far below 1 ng of input (100pg is plenty if you amplify 12 cycles).
I wouldn´t be worried about the beads. A 0.5:1 ratio won´t get rid of your short fragments. The Tn5 has a lower limit of about 300 bp, which means that you won´t get shorter fragments the more enzyme you use (or the longer you incubate). That´s how the enzyme works. It cuts the DNA and attach the adaptors. Once it´s "unloaded" it´s inactive. I use 0.6:1 ratio and get nice libraries of around 300 bp.
Actually, to save money and time you could skip the Zymo columns altogether. Just add the NT buffer from the XT kit (and set up the tagmentation in 20 ul instead of 50) and go on with the PCR. Your libraries will show broader peaks but the quality of data you get won´t change. When I start from 5 ng I usually get conc of 5-8 ng/ul (tot elution vol is 15 ul) when using the Nextera kit.
I can´t write all the details here but very soon we might have a paper accepted where we explain how to perform ALL the steps of the library prep (from single cell isolation to final library) at a fraction of the actual cost and without buying ANY kit (Nextera, SMARTer, etc). check it out!

Hi Simone78,

Has your paper published? I am looking forward to read it as there seems to be some problems in my Nextera library. Thanks.

Jingyi

[Simone78]

Hi Simone78,

Has your paper published? I am looking forward to read it as there seems to be some problems in my Nextera library. Thanks.

Jingyi

the first half of the story has been published here:

http://www.ncbi.nlm.nih.gov/pubmed/24385147

http://www.ncbi.nlm.nih.gov/pubmed/24056875

but that is only about RT and the first PCR. The other half of the story, with all the details about tagmentation (the one you are probably most interested in) is going to be submitted next week. It took much longer than expected because we found a lot of interesting things...I hope it´s getting published soon, I believe it will be very useful for all people out there wasting their money buying Nextera kits one after the other!
Best,
Simone