Disagreement between bioanalyzer and agarose gel for PCR amplicon library

mbirnb · 11-04-2012, 05:10 PM

Hello,

We are setting up for an illumina MiSeq run using a PCR amplicon-based library (adding the Illumina adaptors using nested PCR). Our results look quite clean via agarose gel (expected size ~310nt), but when our facility ran bioanalyzer we had quite variable results (ranging between 400 and 500nt, depending on the sample). There are seven samples, each with a different barcode, labeled 1-7.

I'm attaching both the gel and the bioanalyzer report - has anyone seen anything like this before?

Thanks,
Michael

pmiguel · 11-05-2012, 04:47 AM

I can only speculate. Maybe the agarose gel was run either without EtBr, then post run stained, or no EtBr was put in the gel or loading solution, but was put in the buffer? In either of these cases, the products would migrate mostly or entirely in the absence of EtBr.

Why is that important? Well, I have not tested this myself, but Dave Cook at Sage Science told me that in the absence of EtBr (or other DNA binding dyes) the difference in mobility between fully double-stranded amplicons, and "bubble products" (double stranded only at the adapter ends) does not happen. I.e. everything basically runs at the "correct" size.

So, if this is the issue you see, your amplicons are fine, but consist of variable amounts of normal and bubble-products. The Bioanalyzer chip faithfully displays the difference with the bubble products migrating slower than their true molecular weight.

Again, just a guess.

--
Phillip

DNA_Dan · 11-13-2014, 02:47 PM

Make sure you aren't overloading the chip. The first 3-5 samples look really hot relative to the standard. You could also try a DNA1000 chip since your products are so small.

tomhooven · 07-30-2015, 04:45 PM

Hi Michael--
I was curious how this worked out for you. I recently prepared an amplicon library for TnSeq and had a similar discrepancy between my gel (the expected 190-bp band) and my bioanalyzer results (around 220 bp). I couldn't figure out why this happened.
-Tom

DNA_Dan · 08-03-2015, 10:28 AM

I think this has to do with the mobility of the y-forked adapters. Agarose is more difficult for migration because it's a solid phase, whereas the Bioanalyzer chip's gel matrix is much more fluid. I think the DNA 7500 chip gives the best representation of the "real" size.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Non-specific band after gel purification of PCR products	omnivore	Sample Prep / Library Generation	3	04-05-2017 07:19 PM
Amplicon sequencing - gel cut purification	JKistler	454 Pyrosequencing	7	11-28-2012 12:14 PM
Fluidigm PCR amplicon Library prep	yog77	Sample Prep / Library Generation	2	10-30-2012 05:36 AM
Library invisible on gel, despite Nanodrop results. Gel extraction problem?	thdybwf	Sample Prep / Library Generation	5	09-24-2012 05:41 AM
Library on gel but not on Bioanalyzer	pauji	Sample Prep / Library Generation	6	05-13-2012 06:30 PM

11-05-2012, 04:47 AM	#2
pmiguel Senior Member Location: Purdue University, West Lafayette, Indiana Join Date: Aug 2008 Posts: 2,317	I can only speculate. Maybe the agarose gel was run either without EtBr, then post run stained, or no EtBr was put in the gel or loading solution, but was put in the buffer? In either of these cases, the products would migrate mostly or entirely in the absence of EtBr. Why is that important? Well, I have not tested this myself, but Dave Cook at Sage Science told me that in the absence of EtBr (or other DNA binding dyes) the difference in mobility between fully double-stranded amplicons, and "bubble products" (double stranded only at the adapter ends) does not happen. I.e. everything basically runs at the "correct" size. So, if this is the issue you see, your amplicons are fine, but consist of variable amounts of normal and bubble-products. The Bioanalyzer chip faithfully displays the difference with the bubble products migrating slower than their true molecular weight. Again, just a guess. -- Phillip

11-13-2014, 02:47 PM	#3
DNA_Dan Member Location: Montana Join Date: Nov 2008 Posts: 21	Make sure you aren't overloading the chip. The first 3-5 samples look really hot relative to the standard. You could also try a DNA1000 chip since your products are so small.

07-30-2015, 04:45 PM	#4
tomhooven Junior Member Location: New York City Join Date: Jul 2015 Posts: 1	Hi Michael-- I was curious how this worked out for you. I recently prepared an amplicon library for TnSeq and had a similar discrepancy between my gel (the expected 190-bp band) and my bioanalyzer results (around 220 bp). I couldn't figure out why this happened. -Tom

08-03-2015, 10:28 AM	#5
DNA_Dan Member Location: Montana Join Date: Nov 2008 Posts: 21	I think this has to do with the mobility of the y-forked adapters. Agarose is more difficult for migration because it's a solid phase, whereas the Bioanalyzer chip's gel matrix is much more fluid. I think the DNA 7500 chip gives the best representation of the "real" size.