Protocol Online logo
Top : New Forum Archives (2009-): : PCR, RT-PCR and Real-Time PCR

Is "mini-mixing" acceptable/standard practice? - (Jul/25/2013 )

I had a debate with my PI the other day about whether I should be making mini-mixes for RT-PCR. I'm going to define a "mini mix" as adding your template to one big reaction and then splitting it three ways to make three replicates for the PCR run. The alternative, of course, is to prepare three small reactions and then add the template individually to each. This may seem like not a big deal, but you essentially reduce variability due to pipetting error if you prepare your replicates as a mini-mix (it makes the data more consistent between replicates). At the moment, I'm looking for ways to make my data prettier/more reliable in future experiments. Anyhow, my PI thinks that making a mini-mix is cheating and does not account for the pipetting error that occurs when you make the mini-mix (any error in pipetting for a sample would look like a consistent change in expression across all replicates if you made a mini-mix). The technicians in my lab are convinced that making a mini-mix is the best way to make sure that any issues during amplification are not due to differences in template addition. All of these people have little or no experience with RT-PCR, so I'm posting here to see what other people are doing. Is it standard practice? Or is it a bad idea?


If you are using single primer forward and reverse and the same template DNA why would you want to create three replicas of them? PCR is essentially amplifying for you. 
What I do and everyone else I know does is creating a master mix ( PCR mix) in which if required adding different set of primers. This master mix doesn't include the template DNA. Just before running the PCR we add the template because the master mix already has taq.
Even if you do the mini mix thing you are not accounting for the error in splitting the reaction into three parts. since all the ingredients and the DNA sample are in very minute amount, any small disturbance from the standard would make the whole reaction wrong and eventually give contradictory results. Your PI is right.
Best wishes

-Ragesh Nair-

Great question. Had the same discussion with my PI here.

In my old lab we used a mini-mix, now I make them as individual reactions and add the 2.5ul of cDNA to the samples. I actually have to say that with the later method I get way less variation and it takes half the time to do.

Would be interesting to hear what people say about this.


In our genomics centre they tell you to make mini-mixes and also their robot for the 384 pipettes triplicates out of a 96 well plate.


Well I'm not a real-time expert but in my view a "mini-mix" is not a replicate for the template amount and master mix variability and any possible errors/variability occurring here, as it's all the same but divided by three. Only the effects that may come from other sources of variability such as well position or pcr machine variation and all "possible issues during amplification" you can control with this.

Anyway if the latter points are major sources of variability, then I'd say it's useful to do these mini-mixes, if not then I'd perhaps prefer "real" replicates with each template added singly (and if template amount differences are a major source of variability that needs to be controlled).


Okay, thanks! This was pretty much in line with what I was thinking. I've been setting up individual reactions and will continue to do so.