Melting Curve giving flutating Ct values - Advice on primer design? (Jul/13/2006 )
Hi all,
I am doing some QPCR experiments with SYBR Green using cDNA and my primers never seem to work properly. I re-designed it 3 times already, the first time the primers looked promising- from the melting curve there were no primer dimers but then the CT values were flutuating up and down for my dilutions. So I re-designed the primers and by this I only mean that I only changed the amplicon regions and they were even worse because both gave me primer dimers. I really feel like the first primer set should work because there was amplification and no primer dimers but I just don't understand why the CT values don't behave! I'm getting a little weary of re-designing the primers because it seems like I'm just randomly picking new primers and hoping they will work. But if I do have to re-design it again does anyone have any tips as to how to design the primers that may be more efficient? I follow general guidelines keeping the Tm at around 55-65, the GC content is ~50%, and avoiding more than 2 C and/or G's in the last 5 nucleotides. I also designed the primers pairs so that they each would lie on a different exon. Is there anything else I should consider? I heard it was better to have the primers end in C's or G's instead of A's or T's, is that true? Is it better if the primers span the exon borders or to lie on different exons? What else can I change? Ahhhhhhhh........
have you tested your efficiencies? if your amplification efficiencies are not close to 1 and similar, you'll get wacky Ct's and your data won't make sense
have you tested/controlled everything else to make sure you're not getting degradation / inhibition somewhere?
have you tested/controlled everything else to make sure you're not getting degradation / inhibition somewhere?
The efficiencies aren't great I admit, I know we have to make sure the efficiency of the housekeeping gene and the experimental gene should be similar but I can't really compare the efficiencies without an accurate Ct reading right? If my Ct's are wacky which they are then my I can't calculate my amplification efficieny since it's based on the slope of the graph which would be thrown off becuase of the flutating Ct values right? I use the formula E= 10^(-1/slope), my graph is Ct versus log of ng DNA. How can I test for degradation and inhibition? Please advise!
W-
hmmm
it's a catch22
you won't get reliable Ct's unless your efficiency is good ,or nearly good. that's why I suggested checking efficiency. if your efficiency's bad, you'll never get good reliable Ct's they will always vary and fluctuate
as far as checking for degradation or inhibition, it's a lot like straight PCR, only more so...do you see? you need to do various controls to be certain your RNA hasn't degraded at all the many steps, there aren't inhibitors in your samples, there isn't contamination in your samples...what controls do you perform?
do you see what I mean?
hmmm
it's a catch22
you won't get reliable Ct's unless your efficiency is good ,or nearly good. that's why I suggested checking efficiency. if your efficiency's bad, you'll never get good reliable Ct's they will always vary and fluctuate
as far as checking for degradation or inhibition, it's a lot like straight PCR, only more so...do you see? you need to do various controls to be certain your RNA hasn't degraded at all the many steps, there aren't inhibitors in your samples, there isn't contamination in your samples...what controls do you perform?
do you see what I mean?
hi aimikins,
Forgive me since I am new to QPCR but I am still a bit confused...you are telling me to make sure the efficiency is good before I can get accurate CT readings...but from what I understand it's the other way around...I can't check for efficiency before I have good CT readings since the efficiency is based on the CT readings aren't they? Also, I do a no RNA and no RT control to check for contamination. O, I forgot to mention I do a one step PCR where I start out with RNA as my template and add RT into the reaction so the cDNA is generated while running QPCR. Recently I tried to re-do the experiment using cDNA as my template instead and the CT readings were better although I had some primer dimer formation. Someone suggested that I should just move on with my experiments albeit the imperfect standard curve as long as a pick a template concentration that falls within a good CT range (18-30 cycles). Maybe I should play around with primer concentration some more, I have been using high primer concentrations around 500ng, but I heard I should use no more than 300 ng? Everyone in the lab uses 500ng and never had a problem though. I am running out of options to consider.
I use very minimal primer to get the best results
just because your Ct's are between 18-30 does not make them good
and yeah, you will get bad Ct's if you don't have good efficiency. trust me. do you see why? by 'good' or 'bad' I mean, Ct's that you can use as good data. if you do not have good curves, your data is not good and I would not proceed, if I were you. the data in a project is built up like a house of cards...if the lower levels aren't stable, the upper levels will fall apart...you'll spend more time chasing bad results and this isn't cool.
OK, good Ct's...basically, think about the efficiency. you want it close to 1 (or 100% if that's easier for you to grasp) and what this means is that your product is doubling at every round of PCR. you can see this because if you double your template amount, the Ct will will reduce by 1 ...do you see why? likewise, if you halve your template amount, your Ct should go up by 1. if you can see that your product doubles with every cycle, then your efficiencies are near one and you can calculate your standard curves, knowing that your Ct's are 'good'. if this is not the case, say you double the template and your Ct drops by 1.5, you know the Ct's are 'bad' and your data will suck and be unreliable
does this help, or am I making you more confused?
yes,
I understand that point completely, as my template increases I should see a decrease in my CT values because it will overcome the background and reach the threshold more quickly, which is the reason for my post in the first place because as I increased the template instead of seeing a steady decrease in CT I saw flutuating CT's which in turn also means my efficiency is no good. I have no idea why this would happen since as the melting curves did not show any primer dimer formation nor any side products so theoretically the CT's should be behaving. Could it be perhaps that because I am using RNA as my template instead of cDNA directly that this could be affecting the results? Maybe my RNA is degrading during the reaction?
is there an added RNase inhibitor in your kit? I add one separately during the RT (I split up the steps)...don't know if it's necessary, never tried it w/out
are you certain your RNA is still in good shape prior to starting the reaction? are you certain all your stuff is RNase-free?
you still also must have the right ratio of template to primer in order to get good efficiency
are you certain your RNA is still in good shape prior to starting the reaction? are you certain all your stuff is RNase-free?
you still also must have the right ratio of template to primer in order to get good efficiency
Hi aimikins,
So I ran the standard curve with less primer concentration over more dilutions using cDNA as my template with no added RNase inhibitor or RT and the curve came out perfect! However, now I'm nervous because I'm going to have use RNA as my template for the actual experiment adding RNase inhibitor and RT to my reaction and I have a feeling there is something wrong with the RNA since I had seen those flutuating values when using RNA as my template. At least now I know it's not my primers and I don't need to re-design them.
Yay!! good curves rock
have you run your RNA on a gel to check integrity?
also, what was the difference between the cDNA that gave you good results, vs the cDNA made when you put the RNA directly in? i.e., just different enzyme/kit, etc, or is the protocol different too?