Troubleshooting with the Hive Mind

Late last week, my PCRs stopped working. One day I was able to amplify DNA from multiple different templates using different primers, and the next day I couldn’t. This is a major setback for me — a huge chunk of the remaining work I need to complete for my PhD involves doing PCR. If I can’t get my PCRs to work, I’m royally screwed.


As soon as I couldn’t get any PCR products, I went into troubleshooting mode. I had just made a new batch of dNTPs, so those were the first thing to get replaced. No dice. I thought there might be something wrong with my water, so I grabbed a new bottle and made another batch of dNTPs. Nothing. I replaced my buffer solution. No luck. I made a new dilution of each primer. Nope. I bought new Taq DNA polymerase and new stock solutions of each dNTP, and I remade my dNTP solution again. Nada. I prepped a new dilution of my template DNA (and I had been testing my PCRs using multiple templates that had worked in the past), but that didn’t help. I tried multiple other primers, two different thermal cyclers (I’m currently trying a third machine tonight), different containers of tubes and tips, but none of it helps. When I run my gels, I still get no bands in any of my lanes.

I can pretty much rule out the possibility that I’m getting amplification, but my gels are screwy for two reasons. First of all, my ladder runs perfectly fine in all the tests. Secondly, I reran some PCR products from a couple of weeks ago (back when I was able to perform a successful PCR), and they looked a-ok.

So I’m turning to you guys. Now that I have replaced every reagent, every solution, and my template, what can I do? Is there anything I haven’t thought of (tinkering with Mg2+ concentration, perhaps?) that will fix this problem? Keep in mind, I went from really nice PCR products to none at all, and my primers are designed to the genome sequence of the species and strain I’m amplifying. What’s a helpless grad student to do?


  1. #1 techne
    May 31, 2007

    Here’s some top-of-head ideas. I assume you’re coming from genomic DNA, and not cDNA?

    –Order new primers, both new seqs and new batches of ones you have already used.

    –Try touchdown PCR.

    –Try using some of your old product as template.

    –Try some of those all-in-one-tube, E-Z-just-add-DNA-and-primers formulations. (You’ll never go back.)

    –Have a person from a different lab give you tubes of their DNA and primers to run on your bench.

  2. #2 Spaniard Habsburg
    May 31, 2007

    Poor lazy laking of skills american guy… I’m so sorry for you. But while I was reading your pathetic post I realized that you don’t have any idea of what is a PCR (if your PHD depends on the PCR technique you have to go back to your first graduate year) … How many of those did you do? 10, 20?… So, in your first little problem you are asking for help… Work hard and don’t be such a stupid… By the time that you will have done a couple of hundred reactions please talk about PCR but for the moment just shut up, because you are showing your own stupidity.

  3. #3 Kimball
    May 31, 2007

    Is there any positive control template that amplifies successfully? (mix with sample template to detect inhibition)

    “Try using some of your old product as template” – but nest the new primers inside the old.

    Have you tried tweaking the thermal cycler parameters? (lowering the annealing temperature perhaps?).

    Good luck…

  4. #4 Peggy
    May 31, 2007

    Is anyone else in your lab successfully performing PCR? If so, you could ask to borrow their reagents and see if they work in your hands. That would at least narrow down whether it’s your reagents or something else going wrong.

    Also, did you clean out your pipetman? Maybe it’s contaminated with something that is inhibiting the reaction.

    I’m sure you’ll get it working again, even if it seems like you’ve been cursed by the PCR gods.

  5. #5 John H. McDonald
    May 31, 2007

    I’d suggest double-checking the calculations you used in diluting dNTPs to make the dNTP stock solution–maybe you were supposed to make a solution of 10 mM total dNTP, and you made a solution of 10 mM each dNTP (40 mM total). dNTPs chelate magnesium, so if you have a little too much dNTP in the PCR reaction, all the magnesium gets chelated and it doesn’t work at all.

  6. #6 RPM
    May 31, 2007

    Spanish Guy, Of course I’m a stupid — I can’t get my PCR to work (a fairly simple reaction), and I’m asking for help on my blog.

    As for the more intelligent responses:

    – I’m the only one in my lab doing PCR, but asking someone from another lab for their reagents is a good idea.

    – The exact same reactions that aren’t working today were working a week or two ago. I’ll try the nested primers since I have some of those hanging around already.

    – I actually just cleaned and recalibrated my pipetmen a couple of days ago.

  7. #7 apalazzo
    May 31, 2007

    Rule of thumb:

    If things stop working – throw everything away and start from scratch – replace everything – DNA, oligos, Taq, H2O … everything. Don’t try to figure out what exactly went wrong, it will drive you crazy.

    If that fails, give your stuff to someone else whose PCR is working and see if they can do it and then use their reagents.

    And another tip, don’t use Taq, use Phusion:


    It is a helicase-Taq fusion enzyme that beats the crap out of Taq or any other thermophilic DNA polymerase.

  8. #8 qetzal
    May 31, 2007

    You might also try using a dilution of one of the ‘failed’ reactions in place of your normal template. Maybe you’re getting partial amplification, but for some reason the step-wise efficiency got worse. A small loss at each step can obviously end up causing a major reduction in final product.

    If that’s the case, you probably need to re-optimize your conditions (adjust Mg conc. and temperature conditions). Your conditions could have been near the edge of functionality, and for whatever reason now they’ve slipped beyond it.

    P.S. Spaniard, you should take your own advice (“for the moment just shut up”). You couldn’t look stupider if you tried.

  9. #9 Sandra Porter
    May 31, 2007

    I agree with Alex.

    I have seen too many problems that were caused by malfunctioning water distillation apparati and I have seen people spend way, way, way too much time trouble-shooting those only to find out that filter needed to be changed and no one knew how to do it, or they thought that deionized and distilled water were the same thing.

    Go borrow reagents from someone else whose reactions are working, along with a bottle of distilled water.

    If all else fails, burn some nitrocellulose over your bench at midnight, while playing really loud obnoxius songs from the Kinks. I like “Pressure, Pressure, I’ve got pressure!”
    It may not help your experiments, but it’s kind of fun. 🙂

  10. #10 Sandra Porter
    May 31, 2007

    Oh yeah, a couple of other things.

    Check that the program you’re using on your thermocycler is really hitting the temperatures that you think it’s hitting. Sometimes the thermostat goes awry or people reprogram the machine and assign the program the number that you were using.

    And one, last thing – have you checked your templates at all? Do you know that they’re not degraded?

  11. #11 great_ape
    May 31, 2007

    “I prepped a new dilution of my template DNA (and I had been testing my PCRs using multiple templates that had worked in the past)…”

    Don’t worry; you’ll get it working again. Happens to all of us from time to time. (Except, perhaps, the Spaniard buffoon above). All of the above concerning exchanging all reagents and finding a new water source is good advice. Also consider the possibility that some DNa-ase contaminant has digested your templates, even your stock. (Did your PCRs stop working after you cleaned your pipets or before…make sure you didn’t contaminate things with the nucleic-acid purification solution usually used for decontamination). So you may also seek out a fresh template that you have never touched with your own current pipet set. And then use someone else’s pipet set for the entire rxn.

  12. #12 ERV
    May 31, 2007

    *warning* *blind leading the blind*

    I have no idea what youre doing– but do you grab the wrong master mix tube and get one with UNG?

  13. #13 sparc
    June 1, 2007

    “Try using some of your old product as template” – but nest the new primers inside the old.

    This wouldn’t help because RPM wants to do the same reaction again. IMO ist should be sufficient to dilute the old product 1:100 to 1:1000 to use it with the exact samen primers.
    If you should have enough of all old meterial you could set up a reation with the only the old material and additional reactions in which you replace one of the components at a time with the new reagents.

    BTW, Spaniard Habsburg: If your recations allways work you may have contaminated your lab with some template.

  14. #14 sparc
    June 1, 2007

    One additional remark: I solve all my primers and DNAs in 5mM Tris pH8 (not TE because EDTA may hinder some reactions), because in water DNA is prone to autolysis. I experieced severe problems with water our clinic purchased to set up infusions. DNA isolated from agarose gels was degraded within this water within hours. Water from our destille was not as bad but still one should freeze primers deluted in pure water. My experience is that in buffered solutions primers are quite stable at RT. However, this may have just been good luck.

  15. #15 Kim
    June 1, 2007

    I’m no grad student, but I have done a fair amount of PCR so far. In our lab we use nuclease-free water instead of distilled. First thing I would do would be to grab a sample of something we know we have (E.coli is forever on hand for such instances) and test out the primers. I can’t see the gel too well, but it looks like you have the primers running off just below where the ladder ends, which means it’s not binding at all. I would also consider quantifying the product to make sure the sample hasn’t degraded.

    All else fails, the start-from-scratch approach would be best. New sample, new primers, new mastermix. Good luck!

  16. #16 Kevin
    June 2, 2007

    If you are doing PCR I am assuming you have seen Palumbi’s “Fools guide to PCR.” Personally, pound for pound the best advice I think they offer in that manual is if you work for a week and can not get the reaction working “Go to Kona for the weekend-when you get back there is a good chance it will start.” I have done quite a few PCR’s (for the spanish guy with the big mouth ~4000 samples for 8 loci each) and I always seem to find that a day or two break changes the dynamic. Give it a shot you have tried everything else.

  17. #17 RPM
    June 4, 2007

    I think it was my pipette tips (and maybe my ep tubes).

    I swapped them out, remade all my solutions with the new tips & tubes and my PCR worked. I think the tips (and maybe tubes) got DNase on them, but I’m not sure how. Is it possible to have DNase get on tips when they’re autoclaved? The new tips haven’t been autoclaved and they work fine, so I don’t think I’ll be autoclaving pasticware any longer.

New comments have been disabled.