The CP value is the central value in real-time PCR applications. Everything is related to this single point. But not much effort has been put into standardizing and optimizing the determination of this parameter that is so central to quantification. Most software use the so-called 'threshold cycle method' or 'fit point method' and measure the CP at a constant fluorescence level. But there are other possibilities and options to consider. Let us first think about the background:
• What kind of background fluorescence is evident, a noisy, a constant, a rising or a decreasing background?
• Does the software show me my real raw-fluorescence-data or are the data already manipulated, e.g., additional ROX adjustment?
• What about the curve smoothing of the fluorescence data?
• Which kind of fluorescence background correction and/or subtraction is applied?
Most real-time platforms show pre-adjusted fluorescence data and pre-adjusted CP. After doing an automatic background correction the CP value are determined by various methods, e.g., at a constant level of fluorescence. These constant threshold methods assume that all samples have the same DNA concentration at the threshold fluorescence. But measuring the level of background fluorescence can be a challenge. Often real-time PCR reactions with significant background fluorescence variations occur, caused by drift-ups and drift-downs over the course of the reaction. Averaging over a drifting background will give an overestimation of variance and thus increase the threshold level (Livak, 1997, 2001; Rasmussen, 2001; Wilhelm et al., 2003). The threshold level can be calculated by fitting the intersecting line at 10 standard deviations above baseline fluorescence level. This acquisition mode can be easily automated and is very robust (Livak, 1997, 2001). In the fit point method the user has to discard uninformative background points, exclude the plateau values by entering the number of log-linear points, and then fit a log line to the linear portion of the amplification curves. These log lines are extrapolated back to a common threshold line and the intersection of the two lines provides the CP value. The strength of this method is that it is extremely robust. The weakness is that it is not easily automated and so requires a lot of user interaction, which are more or less arbitrary (Rasmussen, 2001, LightCycler® Software, 2001).
The real problem lies in comparing numerous biological samples. The researcher will have problems in defining a constant background for all samples within one run or between runs. These sample-to-sample differences in variance and absolute fluorescence values lead to the development of a new and user-friendly CP acquisition model. As discussed in the previous section there are several mathematical models to determine the amplification rate, using a logistic or sigmoidal model. These mathematically fit models can also be used to determine the optimal CP (Table 3.1). They are more or less independent of the background level or calculated on the basis of the background fluorescence and implement the data in the CP determination model (Tichopad et al., 2004; Wilhelm et al., 2003).
In LightCycler® (Roche Applied Science) and Rotor-Gene™ (Corbett
Research) software packages these approaches are already implemented. In second derivative maximum method the CP is automatically identified and measured at the maximum acceleration of fluorescence (Ramussen, 2001; LightCycler® Software, 2000). The exact mathematical algorithm applied is still unpublished, but is very comparable to a logistic fit. In the Rotor-Gene family using comparative quantification the 'take of point' is also calculated on basis of a sigmoidal model. Both the sigmoidal and polynomial curve modeto work well with high agreement (P<0.001; r>0.99) (Tichopad et al., 2004; Liu and Saint, 2002a; Liu and Saint, 2002b; Rutledge, 2004). The sigmoidal exponential function was the more precise and could increase the exactness and precision of the CP measurement as well as the amplification efficiency rate (Wilhelm et al., 2003). Peirson further discusses the importance of threshold setting in relative quantification in Chapter 6.
Was this article helpful?