ampure purification of illumina libraries

[greigite]

I'm hoping for some advice/protocol sharing on the use of Ampure beads for purification of Illumina libraries post-PCR. I ran a calibration of my beads with a titrated ladder, which suggested an 0.85:1 bead:template ratio to get rid of adapter dimers. However, following repeated tries at purification I am still not able to completely get rid of a peak at ~130-140 bp (see attached trace). In multiple library preps it usually remains at around 10-25% on a molar basis, still too high for sequencing, as I don't want to waste that much flow cell real estate on adapters. I have tried other bead:template ratios as well, including the standard 1.8:1 bead:template of the standard protocol, and gotten generally worse results. Any tips on optimizing the Ampure purification protocol would be greatly appreciated!

[upenn_ngs]

from my experience, it is necessary to to a gel sizeselect, especially if you are doing whole exome protocol. the peak at 130 is self-ligated adapter with pcr extension. invitrogen has a great precast gel that makes the sizeselect quick. run the gel after ligation and before pcr to separate.

[greigite]

Interesting to know that the 130 bp peak is self ligated adapter. I did actually do a gel size selection step following ligation, but it seems that was insufficient to actually eliminate self-ligated adapter. I guess that migration through the gel was imperfect such that there were still some adapters "trapped" with the higher MW DNA, and that they preferentially amplified in the PCR.

from my experience, it is necessary to to a gel sizeselect, especially if you are doing whole exome protocol. the peak at 130 is self-ligated adapter with pcr extension. invitrogen has a great precast gel that makes the sizeselect quick. run the gel after ligation and before pcr to separate.

[nextgen]

The reason is that you have a lot DNA at 130. Ampure size selection is not as tight. When you have a lot DNA in the vicinity of your target size, it will just carry over. A double selection or tighter selection will help. The problem with gel is that when you load a lot, DNA gets tangled together. The target DNA has mix from non-target DNA. That is why Saga bio tries to limit the DNA input for their automated robot.

I'm hoping for some advice/protocol sharing on the use of Ampure beads for purification of Illumina libraries post-PCR. I ran a calibration of my beads with a titrated ladder, which suggested an 0.85:1 bead:template ratio to get rid of adapter dimers. However, following repeated tries at purification I am still not able to completely get rid of a peak at ~130-140 bp (see attached trace). In multiple library preps it usually remains at around 10-25% on a molar basis, still too high for sequencing, as I don't want to waste that much flow cell real estate on adapters. I have tried other bead:template ratios as well, including the standard 1.8:1 bead:template of the standard protocol, and gotten generally worse results. Any tips on optimizing the Ampure purification protocol would be greatly appreciated!

[greigite]

The reason is that you have a lot DNA at 130. Ampure size selection is not as tight. When you have a lot DNA in the vicinity of your target size, it will just carry over. A double selection or tighter selection will help. The problem with gel is that when you load a lot, DNA gets tangled together. The target DNA has mix from non-target DNA. That is why Saga bio tries to limit the DNA input for their automated robot.

Hi nextgen,
Could you provide more info or a link about the robot you mention? Also, by double selection do you mean a post-PCR gel size selection in addition to the post-ligation gel size selection?

[nextgen]

Check this out:
http://www.genomeweb.com/sage-scienc...gen-sequencing.

Double selection means you do the same selection twice. You will lose more DNA with each selection. Ampure selection ability is limited, especially towards 500bp and above. So there is a trade off between yield, DNA size tightness and yield.

Hi nextgen,
Could you provide more info or a link about the robot you mention? Also, by double selection do you mean a post-PCR gel size selection in addition to the post-ligation gel size selection?

[gogreen]

Hi Greigite,

I see a similar band at 130 bp in my sequencing library too! Did you figure out why this happens and any way to get rid of this? I tried going down on the adaptor concentration. But it was no good. The first lane is the library prepared with the solexa adaptors and the 2nd one is a multiplex adaptor that we made. I wonder why this doesnt happen with the solexa adaptors used at a higher concentration than the multiplex!! any suggestions would be great, Thanks in advance

[greigite]

Hi Greigite,

I see a similar band at 130 bp in my sequencing library too! Did you figure out why this happens and any way to get rid of this? I tried going down on the adaptor concentration. But it was no good. The first lane is the library prepared with the solexa adaptors and the 2nd one is a multiplex adaptor that we made. I wonder why this doesnt happen with the solexa adaptors used at a higher concentration than the multiplex!! any suggestions would be great, Thanks in advance

As upenn_ngs said above this band is most likely self-ligated adapter amplified with PCR primers. Your library looks concentrated enough that gel extraction should work well to eliminate the 130 bp peak. I've heard that there are some completely ampure-based protocols now for size selection but I have not been able to get rid of that peak without gel extraction. most likely I am doing something wrong

[upenn_ngs]

As upenn_ngs said above this band is most likely self-ligated adapter amplified with PCR primers. Your library looks concentrated enough that gel extraction should work well to eliminate the 130 bp peak. I've heard that there are some completely ampure-based protocols now for size selection but I have not been able to get rid of that peak without gel extraction. most likely I am doing something wrong

we have had success eliminating excess adapters by performing two rounds of purification after the ligation. good luck!

[zerone]

Did you use two rounds of purification using Ampure beads or with gel extraction?

[silin284]

I have tried AMPure to clean up the ligation product and both 1.8 and 1.0 volume of AMPure beads work. I have never seen a 130 bp band after purification..... Is it possible that too much adapter have been used? or the homemade adapters are of poor quality therefore lots of self-ligation?

[aperera]

Has anyone tried the "Size Selector" beads from Aline Bio?

[sjcire]

One thing that may help is that Ampure XP uses PEG 8000 as a crowding reagent. If you use the quick ligation kit, the 2X quick ligation buffer has 17% PEG 6000. Therefore, Ampure XP PEG concentration ends up higher and thus the oligo selection cut off is lower. Try Ampure selection twice and dont wash with 70% EtOH but with a 6-7% PEG 8000 solution with 1.25M NaCl and some Mg2+ in there. I stepped into this same trap. DARN YOU PEG!!

[nextgen]

Sizeselector from Aline Bio

Has anyone tried the "Size Selector" beads from Aline Bio?

You may talk to Tufts University core facility. They have some good experience. It is very effective in removing adaptor contamination.

[TonyBrooks]

Here a a few tips that I've found through experience:
1) titrate your adapter to your sample input. We use 1µL of 15µL for every 500ng of DNA. For making homebrew adapters we follow the protocol in Kozarewa et al. http://www.ncbi.nlm.nih.gov/pubmed/21431778.
2) If in doubt, double Ampure. If you want to save on beads, you can use a homebrew binding buffer (20% PEG8000, 2.5M NaCl). After resuspending beads in elution buffer, just add this at the required concentration, incubate, rebind beads, wash with ethanol twice, dry and elute.
3) 1:1 is a good ratio to use for most libraries. This should be good a removing anything below 200bp. We don't bother calibrating beads against a ladder. It's mostly a waste of time and reagents.
4) Try and do the double Ampure pre-PCR. PCR will preferentially amplify smaller material, so it will favour any adapter dimer over sample. Failure to do this will lead to large dimer peaks and lower yield from your library.
4) Make sure the beads are dry after the ethanol wash. Spin down the tube/plate once you remove the second wash. Place it back on the magnet and use a low volume pipette to remove any residual ethanol. Look for cracks in the bead pellet so you know it's dry. You can also speed up drying by incubating on a 37ºC block. If you do that, check every 30 seconds to make sure you don't over-dry the beads.