What if there are no curves or curves that look unusual? This situation must be dealt with before any further analysis is possible. The first parameter that should be checked is that the correct dye layer or reporter has been designated. This seems like a very simple proposition, but if there are no amplification curves or strange curves this is the most likely explanation. This problem occurs most often on multi-user instruments and when templates are routinely used. The result of choosing the wrong template with the incorrect dye layer can be as drastic, as very few amplification curves above the threshold as shown in Figure 2.4A (Ct of a = 33) when TAMRA is chosen for the dye layer for a FAM probe or very subtle as the instance when a SYBR® Green I template is chosen when using a FAM probe, as shown in Figure 2.4B, (Ct of a = 17). If the spectral calibration curves are viewed (Figure 2.4D), it is easy to see that the TAMRA spectrum is quite different from FAM. However, because the FAM and SYBR® Green I have overlapping spectra, the amplification curves will appear quite respectable and the Ct values may be indistinguishable (Ct of curve 'a' is 17 in Figure 2.4B and 2.4C). Only by checking that the correct dye layer has been chosen will the discrepancy be detected. One other way to detect/verify this problem is to select a well and look at it in the Raw Spectra view with the best fit option turned on. The curve of the raw spectra for the well and the fit curve should be superimposed on each other. If they are not this is an indication that the wrong reporter is being monitored (or there is contamination with another dye on the plate or in the sample block). What can be done to correct this problem if it occurs? Go to the set-up view, select the proper dye layer, reassign the wells and reanalyze to get the proper Ct values as shown in Figure 2.4C. One way to minimize the chances of the wrong dye layer being used is to reboot the data collection computer between individual runs, which resets template/software parameters to their previously assigned values.
What if the Amplification view looks like the one shown in Figure 2.5A, with very erratic amplification plots and an unusual threshold? This occurs on the ABI 7700 (Applied Biosystems, Foster City, CA, USA) when it encounters wells that have no fluorescence in them. This may be a result of mistakenly designating wells as containing a sample when they are actually not being used, or the over zealous newbie who wants to include a water blank. The troublesome wells can be identified by selecting all wells on the plate and observing them in the Raw Spectra view. They will be most apparent if looked at after 40 cycles of amplification. The empty wells will have flat fluorescent traces, as shown in Figure 2.5B. The lack of fluorescence from probe and/or reference dye can be verified by examining the individual wells in the Multicomponent view (Figure 2.5C) which allows viewing of the raw fluorescence in an individual well. The problem is resolved by switching to the set-up view and designating the offending wells as 'not in use' and reanalyzing the data without the blank wells. Now the curves look quite respectable, as shown in Figure 2.5D and an example Ct (b) is corrected from 29 to 17.
One other anomaly that can generate very weird curves is having ROX designated as the reference dye when in fact there is no ROX in the reaction. The amplification curves will look similar to Figure 2.5A. Turning
Effect of including wells with no fluorescence. A. Appearance of the amplification view if wells have been selected that contain no fluorescence. (a) artifact, (b) real amplification. B. Raw data view of all samples.
C. Multicomponent view of (a) verifying the absence of fluorescence.
D. Appearance of the amplification view after removing the wells with no fluorescence and reanalyzing. Sample b has a Ct of 17 compared to a Ct of 29 in panel A.
off ROX as a reference and reanalyzing, restores the curves to their proper configuration.
Was this article helpful?