We have done RAD-based libraries with samples which were extracted using PureGene, Qiagen, Phenol-chloroform... I don't think it really matters as long as you have pure DNA of a high molecular weight. You will also need fairly high yields as you will only recover a small proportion of fragments following size selection, and after ampure cleanup, which removes the very small fragments (usually we digest 1,000 ng and recover 10-30% after digest and ampure). Typically we use extractions with most fragments greater than several kb...

Another consideration is the purity of the DNA.. Many restriction enzymes are salt-inhibited, I have had some extractions which failed to digest.. This was easily fixed by purifying those samples (Ampure XP)- but this does create more work.

Degredation is an issue in that it decreases efficiency of shearing, but also in that highly fragmentary extractions decrease the probability of recovering homologous stretches of DNA across individuals... Additionally you would expect to generate on average smaller contigs... However I do not think that you necessarily need "super quality" DNA as I would think the method should be robust to moderate degredation.

What yields and quality do you expect from your extractions? Do you have some extracted sampled on-hand?

Here's a paper: http://onlinelibrary.wiley.com/doi/10.1111/1755-0998.12404/epdf

We have done RAD-based libraries with samples which were extracted using PureGene, Qiagen, Phenol-chloroform... I don't think it really matters as long as you have pure DNA of a high molecular weight. You will also need fairly high yields as you will only recover a small proportion of fragments following size selection, and after ampure cleanup, which removes the very small fragments (usually we digest 1,000 ng and recover 10-30% after digest and ampure). Typically we use extractions with most fragments greater than several kb...

Another consideration is the purity of the DNA.. Many restriction enzymes are salt-inhibited, I have had some extractions which failed to digest.. This was easily fixed by purifying those samples (Ampure XP)- but this does create more work.

Degredation is an issue in that it decreases efficiency of shearing, but also in that highly fragmentary extractions decrease the probability of recovering homologous stretches of DNA across individuals... Additionally you would expect to generate on average smaller contigs... However I do not think that you necessarily need "super quality" DNA as I would think the method should be robust to moderate degredation.

What yields and quality do you expect from your extractions? Do you have some extracted sampled on-hand?

Here's a paper: http://onlinelibrary.wiley.com/doi/10.1111/1755-0998.12404/epdf

Tyler's answer is in line with our experience, and a useful guide.  Several different types of extractions will work, depending on the species and the quality of the starting material.  Some taxa are difficult- like cacti with all their mucilage.  But Tyler's guidelines will work.  A minimum quality is roughly 10 ng/ul, and not too sheered.  But I would say that is a rather crude threshold, and you would much prefer better extractions than that.

Dehydrated leaf tissue (60/100 mg) from each sample was homogenized in a 2-ml microcentrifuge tube containing a 5-mm steel bead cooled with liquid nitrogen using Mixer Mill 300 (Qiagen, Hilden, Germany). Genomic DNA was extracted and purified using a DNeasy96 Plant Kit (Qiagen) and stored at −20 °C. For any other detail about the  quantity of DNA and procedures, please contact Paola Pollegioni, Research Gate member. Regards

Thanks everyone so far!!  I expect high yields- I have had good success in the past with CTAB.  Yields in the 100s of ng.  So that is not a problem.  I expect two problems: 1) with shearing (we do use a 'bead beater' type machine to pulverize tissue- does that cause shearing?).  But we will run on gels to check that.  2) purity, as you mention salts and other impurities might interfere with the restriction or sequencing.  And that is likely species specific.  Does a nanodrop reveal these impurities?  Tyler could you explain a little more about Ampure?  And thanks for the reference on degredation- we don't expect quite so badly degraded DNA, just perhaps some shearing from the pulverizing.

Our lab uses a CTAB extraction protocol, and since our tissue is very tough (C4 grass) we freeze dry it and grind it vigorously in a bead beater before extraction.  Double-digest RAD-seq has worked fantastically for us.  Ideally we want 50 ng/ul DNA, but it still works with 10 ng/ul and with poor quality samples (phenolics etc.).  If I have a lot of good quality samples and a lot of poor quality samples, I make libraries that exclusively contain good samples or exclusively poor samples; then the poor ones don't get outcompeted by the good ones so much in terms of read depth.

Protocol available here: http://openwetware.org/wiki/Sacks:RAD-seq

Thanks Lindsay for the suggestion!  And thanks everyone.  Please let's keep the conversation going- its probably useful to lots of people.

Hello Sean,

I am in the same lab as Tyler and noticed you asked about the AMPure XP cleanup. In many column methods of extractions (i.e. Qiagen DNeasy) you can sometimes get high salt concentrations which can reduce the activity of the restriction enzymes. In NEB's troubleshooting guide they suggest to avoid this DNA solution should not be more than 25% of the reaction volume and that DNA solutions may need to be cleaned prior to digest. Most of the time we do not have trouble with salt inhibition but occasionally we will have a sample with excess salt or have a low yield sample in which more DNA is need or the sample has to be concentrated which may result in a high concentration of salt in the final solution.

If we do you salt inhibition we just clean up the sample with AMPure XP beads at a 1.5 to 1 (AMPure to DNA solution) ratio. This removes the salt from the solution by pulling out the DNA and washing away the salt and has worked great for us. However, this does not typically happen (maybe 1 out of every 1,000 samples) and is more of a concern for column based extraction methods, which we tend to use.

Hope this helps also below is NEB's trouble shooting guide. It may help if you have any problems with digestion.

https://www.neb.com/tools-and-resources/troubleshooting-guides/restriction-enzyme-troubleshooting-guide

Best Regards

Qiagen tissue kit (slightly improved for small peaces of ear biopsies) and PhenolChlo protocols worked fine for us on small primates (lemurs)  ear biopsies.

Thanks Max!  That troubleshooting guide will be really helpful!

Do you think is there any problem if I used ATMAB instead of CTAB for RADseq? Or if I used diclhloromethane instead of chloroform? I work with trees and sometimes the DNA extraction is very difficult. Thanks!

When targeting HMW DNA, I recommend avoiding extraction methods that depend on spin columns due to the potential for shearing and degradation. Phenol Chloroform will give great results but ultimately, I have found the PureGene extraction kit from Qiagen to be superior for obtaining high yields of HMW DNA without the need for dangerous chemicals.