Concentrating NexteraXT libraries

[docbio]

Just wondering if anyone has any experience with this. I am still fairly new to preparing NexteraXT libraries. I have had good luck with the bead based normalization and running pools on the MiSeq, but now I want to send some pools to a HiSeq core lab.

My HiSeq core wants pools to be around 10nM. Illumina recommends pooling the double stranded libraries from the CAN plate and then using that for the HiSeq (vs the single stranded libraries coming out of the bead based normalization). I used a Qubit to quantitate one of my first CAN plates and the concentrations are fairly low, in the 500-3000 pg/uL range. The average size is around 800bp on Bioanalyzer. I can pool them but the pool volume is going to be large and the concentration very low, so I need to concentrate the pools.

Illumina recommends SpeedVac of the samples without heat, but tech support warned me if I used the buffer in the kit for the elution into the CAN plate (and why wouldn't have I?), then the salts in the buffer will concentrate as well, so they recommend not reducing the volume by >50%. They say in the future I can elute with H20 instead of the buffer in the kit, but that doesn't help me now. Any other column methods, while possible, would run the risk of sample loss. Ethanol precipitation might be an option, but the quantities are so low, I fear being able to see a pellet or achieve success with that method.

I haven't sat down to do the math yet to figure out my theoretical volumes, but just wondering if anyone else had any experience or advice for running NexteraXT libraries on the HiSeq.

Best,
DB

[pmiguel]

AMPure.

--
Phillip

[docbio]

I have AmpureXP beads in house but they are a couple of months out of date - not sure how quickly they degrade but not sure I want to risk it. I have fresh AmpureXP on order.

If I were to go the AmpureXP route, would you follow the standard PCR cleanup protocol from Beckman with a 1.8x bead ratio, or something more specific to concentrating libraries? My library size ranges from around 200bp to 1kb with a peak around 800bp. I've been looking around but can't find a consensus.

Thanks for the help.

[pmiguel]

0.8x ampure works in our hands to get rid of much of the stuff below 300. Of course this may be somewhat batch-dependent.

Keep your first supernatant as a fail-safe. If you need to, you can add additional ampure to recover all the DNA.

--
Phillip

[docbio]

My pool volumes wound up being rather large (~500uL) and I didn't have any un-expired AmpureXP on hand, so I decided to call an audible and try an Amicon 0.5ml Ultracel-30K concentrator that we had on hand. It seems to have worked fairly well with around an 80% recovery of my calculated input (too dilute to detect by Qubit). Looks good on the Bioanalyzer, too, although the size range does tend towards the large end in one pool in particular.

I have heard that the quantitation on the Bioanalyzer is not reliable, compared to the Qubit. In looking at a few individual libraries that I ran on the higher end of the concentration spectrum as well as my pools, it seems the Bioanalyzer calculated concentration is consistently around one third to one half of the Qubit value. Is this consistent with what others see?

Best,
docbio