My guess: your parameters for the baseline correction are set wrongly.

Presumably data of the exponential phase is wrongly used to estimate a background trend. Look up these parameters and consult the software manual how to chenge these settings.

HI Jacob

That's quite a graph you have there.  I agree with Jochen, it's possible that you're baseline/ control parameters are not within range.  Also - you may want to make sure that the machine is functioning correctly.  Sounds really silly but I've had a similar situation when out-sourcing and they found the machine had missed a service and needed to be re-calibrated.  Do you have positive controls you're able to run along with these? If those also fail then it's most likely a reagent, your standards or the machine.

Good luck

Nathaniel

It really does not look as if the reagent has a problem. You can recognize that there is an amplification in most of the samples.

The problem is how the software calculated these amplification curves from the raw fluorescence signal data. Most likely the baseline correction doe not work correctly. It is surely helpful to check the raw fluorescence signals!

 Thank you for your responses, Nathaniel and Jochen!

So, out of 98 miRs (2 pos controls) this is the only one that has this type of amplicfication curve, so we believe the instrument is working allright. 

BTW, this qPCR are performed on a Fluidigm Biomark HD and analyses in Fluidigm RT-PCR analysis software. As I understand, baseline correction can only be set to a few predefined options ("linear", "linear (derivative)" and "constant") which then corrects for background. (the curve is shown with "linear (derivative)" correction.)

At least, we are not familiar with other options for this.

Maybe someone are aware of other options in the fluidigm software?

Cheers