Before discussing the actual preliminary analysis procedure, it is wise to have a good understanding of what constitutes the best possible assay. Information on primer and probe design will help with this task if you are designing your own assay. Often it is more convenient to use a published assay, however, keep in mind that just because a primer/probe set is published, it is no guarantee that an optimal assay will result. Also, typographical errors in published DNA sequences of any type abound, so BLAST (basic local assignment tool) all sequences before using them to insure they are correct. If the information is not included in the publication, it is a good idea to check the sequence and melting temperatures (Tm) of the primers and the probe just to assure that the assay meets basic design requirements, before going to the expense of purchasing the primer/probe set. Checking the relationship of the primers/probe to intron/exon boundaries will make you aware of the importance of having controls that determine genomic DNA contamination in RNA preparations. Similarly, checking amplicon size will prevent any unpleasant surprises. Perhaps these measures are 'overkill' but they can save time and money in the long run. The assay should always be validated using the conditions in your laboratory.

How to determine if an assay is optimal? The standard curve is a very useful tool for determining the qualities of an assay. Using a defined template, such as a plasmid containing a relevant portion of the gene of interest, PCR product, synthetic oligonucleotide or transcribed RNA to perform a standard curve, will allow determination of the PCR efficiency of the assay along with the sensitivity and dynamic range independently of any variables associated with the sample preparation and/or reverse transcription.

How is a standard curve generated? Begin with determining the OD260 of the template sample.

The amount of the template should be expressed in molecules. Convert the mass to molecules using the formula:

Mass(ingrams) x Avogadro'sNumber .

Airi 111 — molecules or dina

Average mol. wt. of a base x template length

For example, if a synthetic 75-mer oligonucleotide (single-stranded DNA) is used as the template, 0.8 pg would be equal to 2 x 107 molecules:

0.8 x 10-12 gm x 6.023 x 1023 molecules/mole 330 gm/mole/base x 75 bases

For double-stranded templates, use 660 gm/mole/base. To generate a standard curve, prepare seven 10-fold dilutions, starting with 1 x 107 template copies and ending with 10 copies. Be sure to use a carrier such as tRNA to minimize loss of template at low concentrations. Use a previously determined optimized primer/probe concentration that gives the highest ARn and the lowest Ct.

Most platform software will plot a standard curve using designated wells or Microsoft Excel® can be used to draw xy plots with the log template amount as the x value and the cycle threshold (Ct) as the y value. A line representing the best fit is calculated for a standard curve using the least squares method of linear regression:

y — m x + b where y — Ct, m — slope, x — log10 template amount, and b — y-intercept.

The coefficient of determination, r2 should also be noted.

What does each parameter tell us? Assay efficiency is based on the slope of the line, calculated by the formula:

Efficiency is primarily an indication of how well the PCR reaction has proceeded.

The integrity of the data fit to the theoretical line is described by the r2. This is a measure of the accuracy of the dilutions and precision of pipetting. The y-intercept is an indication of the sensitivity of the assay and how accurately the template has been quantified.

Figure 2.1

Standard curve cartoons. Exaggerated depictions of a 'perfect' standard curve (closed circles) and the effect that low efficiency (open circles) or inaccurate template quantification or degraded template has on the y-intercept (closed diamonds).

What constitutes a good assay? Figure 2.1 is an exaggerated depiction of the standard curve of a 'perfect' assay and what influence an alteration of the parameters will have on the outcome. By expressing the template amount in molecules, the formula below can be used to determine how many cycles are necessary to produce a specified number of molecules.

n = Log(Nn) - log(No)/log(1+E) Where Nn = number of copies of the template after n cycles, N0 = number of copies of the template at cycle 0, E = efficiency and n = number of cycles.

Taking into account that the signal detection limit of free FAM is 1010 -1011 molecules on most platforms, it is possible to determine that at 100% PCR efficiency, a single copy of template should be detected between 33.3-36.5 cycles. Therefore, a perfect assay would have a slope of -3.32

Standard Curre - Flu7

Standard Curre - Flu7

10" t 10" 2 10" 3 10-4 10-5 10" Ê 10*7 10" S 10" 9

Starting quantity

Figure 2.2

10" t 10" 2 10" 3 10-4 10-5 10" Ê 10*7 10" S 10" 9

Starting quantity

Figure 2.2

Standard curve of degraded template. Real example of a standard curve using a template that has degraded during storage.

(100% efficiency), a y-intercept between 33 and 37 cycles and an r2 of 1.00. Figure 2.1 demonstrates that as the slope of the line increases (becomes more negative), the efficiency decreases, the y-intercept increases and the sensitivity of the assay suffers because more starting copies are needed before the detection limit is reached. Or, more cycles are required to detect the same number of template molecules. This may be a distinct disadvantage if dealing with genes that have low expression. Make a spreadsheet using the formulas above and see for yourself what effect alteration of the efficiency has on detectable copy number or the cycle at which the sample will be detected.

An example where efficiency may not be optimal, even though the primers and probes have been properly designed, is the case of a non-linearized plasmid being used as a template for the standard curve. Because of super-coiling, 100% of the copies of the plasmid are not available for initial PCR resulting in low efficiency at the start, but once the new ampli-cons outnumber the plasmid, efficiency is 100%. The net result is an underestimation of copy number. This is why plasmids should always be linearized before using them as templates for standard curves. However, efficiency is not necessarily the only measure of assay quality. An assay can have an apparently acceptable efficiency of 95-100% but if the y-intercept is substantially higher than 37 or lower than 33, it indicates that the amount of template has not been correctly determined. In the case of a high y-intercept, deterioration of the template is likely (Figure 2.2). Degradation of template is a common result of too many freeze-thaw cycles of the standard or storing the template at too low a concentration without a carrier.

Once the assay has been validated using a standard curve, then it can be used with cDNA or a universal RNA standard in a similar manner to optimize sample preparation and reverse transcription by monitoring assay efficiency. Of course, in reality, not every assay will be perfect, but an attempt should be made to develop the best assay possible or at least know what the weaknesses of an assay are.

Preliminary analysis flowchart

Examine amplification curves

Check/adjust baseline

Check/adjust threshold

Check no template controls

Check positive controls

Check -RT controls

Check samples

Perform experiment specific analysis

Figure 2.3

Flow chart for preliminary analysis. Parameters to check during preliminary analysis of data obtained from a real-time PCR run.

Was this article helpful?

## Post a comment