So I co-extracted DNA and RNA from decomposing root material using Mobio's RNA powersoil kit and now I'm trying to amplify the fungal ITS7/ITS4 region for fungal community analysis. The DNA that comes off of this kit is good quality, 260/280 around 1.8-2, 260/230 around 1.9. I'm using published PCR conditions which have worked well for me in the past, and barcoded primers that have also worked well. For some unknown reason, the majority of these samples will not amplify. The nanodrop readings are good, so I thought maybe it was RNA carryover somehow contaminating my samples and inhibiting PCR, so I treated the samples with RNAse and ran them through a phenol-chloroform extraction. Good quality DNA came out the otherside, again. I'm using 25 uL reactions with 100ng of template, and using dreamtaq master mix (4 mM MgCl2, 0.4 mM dNTP each, plus the buffer). I've varied the amount of template (10, 50, 100, 200 ng) and still no amplification. In the last run I just did, 6 of 16 samples successfully amplified, but the samples that worked look no different in terms of quality then those that don't amplify. I'm starting to think there is some kind of demon sabotaging things while I'm not looking! Any help/suggestions would be greatly appreciated
---
Edit:
Thanks for your responses. I forgot to mention that my DNA also looks great on a gel, no degradation. As for the MgCl2 concentration - yes, it is high in this master mix, but wouldn't that just reduce specificity by increasing the activity of Taq? I assume that's why they call it 'DreamTaq' - it'll amplify anything with such a high MgCl2 concentration (at the expense of specificity). I think I'll just bump up the cycle number and dilute even further and see where that gets me. And throw in some positive controls, plus some revert back to the 50 uL rxn volume recommended by the manufacturer. I'll report if anything works for those also struggling with PCR (also known as witchcraft).
---
Edit 2:
Dilution seems to be the key to getting good amplification. Doubling the reaction volume and diluting the template seemed to solve the problem. For my most difficult samples, 2ng/uL did the trick. I went through a phenol-chloroform clean-up my samples, but I think that was unnecessary - I should have just diluted the sample from the start before cleaning up.