I am trying to find the efficiency of my primers by doing a serial dilution of my sample, but the last 1 or 2 dilution points fail to amplify (or my triplicate points don't fall in the same Ct value). This leads to a failed efficiency if I keep all 5 dilution points. Is there a rule about omitting points? I know sometimes the most concentrated point may inhibit the reaction, and I've read that some times you can omit the first point. What is the case for the dilutions that follow?
The fuzziness of the Cq values at the least concentrated part of your curve can be due to near-single copy reactions. You would expect this in such reactions. Are you using a SYBR Green method, hydrolysis probe method, and what are the concentrations of your samples in each of the 5 standard rxns? An interesting thing about "nil" analysis: once you find out the average Cq value at which you get enough on-off behavior of enough of the technical replicate Cq values for such a reaction: you have then identified your 1-copy Cq, and then can proceed to speak of your results in absolute copy terms.
Also, make sure your primer/probe designs are 100% specific for the target here. Have you BLASTed their sequences against all data bases to prove the fidelity of your sequences for their intended target? Never absolutely trust a published primer/probe sequence; sleuth out for your self where they actually bind on the sequence of interest in each case. Leaving simple unknowns like this uninvestigated can be the source of a problem, and also the way that poor oligo designs are handed down through the ages in the literature... e.g., once published, some details are considered sacred. Not so with primer and probe designs - always distrust them until you see for yourself where they were designed to on the exact sequence of interest you wish to interrogate.
Hey there Jack,
I am using SYBR Green for relative quantification. I am working with time course samples. My starting RNA concentration was 100ng (I know, this is very low, but I didn't have enough to work with) for each time point sample. After cDNA synthesis I pooled all the time course samples and performed a 1:5 serial dilution. I then took 1ul per dilution (in triplicates) to test my primers.
I did a BLAST of my sequences and they are unique to my organism and picked the primer location based on bioinformatics data.
I agree with you about publised primers. I started with some that I found on literature, but were not that great, so I designed my own.
My guess is that maybe the starting concentration of my RNA was too low? What puzzles me is that when using this pooled sample I was able to get amplification of my first point for one of my genes. I then tested a single time point for one of those genes with an RNA concentration of 500ng and didn't get any amplification that time.
So the cDNA concentration (in ~ng cDNA/uL) in your qPCReactions was?
The pooling is a good idea to pick up all targets - and yes you can have a sample that is not expressing a target of interest - but, that may also be confused with loading too much sample in a qPCReaction and shutting Taq down. Sounds like you are being very careful with design - which is good~!
- Badges
- Science method