PCR in real-time

Definition

The chain reaction by polymerase in real-time is a technology having many applications, based on a enzymologic reaction, PCR and on the continuous measurement of its product.
Il exists different Appareils from PCR in real-time.
A each cycle of amplification, the quantity of total DNA or Amplicon is measured thanks to a fluorescent marker. Obtaining the complete kinetics of the reaction of polymerization makes it possible to obtain an absolute quantification of the initial quantity of target DNA, which was very difficult to obtain without skew in PCR in final point. From the enzymatic point of view, it has there no theoretical difference between these two types of PCR. You are thus invited to consult to consult the general page on the Chain reaction by polymerase. Thanks to the extraordinary power of the technique of amplification of the DNA by PCR, it is possible to today establish a genetic profile starting from negligible quantities of DNA.

History

1985 : Invention by Kary Mullis in 1985 of the technique of PCR (PCR).

1991 : first use of probe (probe of hydrolysis) in PCR published by Holland PM and collaborators.

1992 : the first detection of the product of PCR in real-time using a marking with the Study Bureau published by Higuchi R and collaborators.

Why quantify of many cycles?

The basic difference of the PCR in Real-time with the PCR in final point is that the entirety of the measurable kinetics (above the background noise) is quantified. The data of fluorescence can thus be expressed in logarithm in order to easily identify the exponential and measurable phase, which takes a linear appearance then. This part, then called “quantifiable segment”, makes it possible to calculate the quantity of initial DNA. Graphic of left (A):

the kinetics of theoretical PCR of three samples (K, L and M) of decreasing concentration of initial DNA (let us consider of a factor 10 each time) are represented in a reference mark semi logarithmic curve. Measurements of the background noise or too influenced by him (phase 1 and 2 of measurable kinetics of PCR) are not represented.

(1) Zone of the quantifiable segments of each sample (phase 3D' measurable kinetics of PCR). The other phases are represented in gray and do not have interest for this chapter.

(2) Each quantifiable segment makes it possible to define an equation of the type Y = aX + B . This equation of right-hand side models the exponential phase of the PCR after transformation logarithmic curve, even the part which is not measurable because of the background noise. The exponential phases of the samples K, L and M are represented respectively by the full lines navy blue, turquoise and mauve. The slope “has” is derived from the effectiveness of PCR. Since same the amplicon is detected for each sample, it is a constant and the lines are parallel. The ordinate in the beginning “B” corresponds to the quantity of DNA to cycle 0. the modeling of the quantifiable segment thus allows theoretically to directly determine the quantity of initial DNA.

(3) Actually, experimental measurements always have error, such a weak is it, by definition imprédictibles and answering stochastic phenomena. So samples of concentrations K, L and M were amplified several times, one would obtain as many different kinetics, although very close relations (their quantifiable segments are respectively represented in red, pink and orange). Each one of these measurements makes it possible to establish a new line equation represented in dotted lines of the same color. Different ordered in the beginning (B) are then measured. It will be considered that the two lines represented for each sample represent the maximum error of the technique. The difference between their B (EK, EL and EM) thus represents the uncertainty of measurement for each sample. This uncertainty is considerable. In the example, it is sufficient so that one can measure L more concentrated qu K or M that L but actually, it is often much larger than that (two to four orders of magnitude).

(4) Remarquez that uncertainty to the measure is not identical for each sample (EK is lower than EL, itself smaller than EM). A projection of the equation of right-hand side on the y-axis or one of its parallels gives an uncertainty concentration of initial DNA dependant .

Graphic of right-hand side (B):

(1) the equations determined starting from the quantifiable segments can be extrapolated to the x-axis, even if one obtains a mathematical point without biochemical reality.

(2) the lines modelling the experimental error (dotted lines and red, orange or pink) then make it possible to define new uncertainties EK, EL and EM. Notice that these uncertainties are larger than in the graph (A) but of size identical between them. The error to the measure thus became concentration of initial DNA independent (3).

(4) It is possible to project the equations of right-hand side on a parallel with the x-axis cutting the quantifiable segments by the medium. This segment will be called “ threshold of detection ”.

(5) the values in X (of many cycles) of these intersections are generally named CT (of English Cycle Threshold for " cycle seuil"), but sometimes also CP (of English Crossing Not for " not croisement"). They are mathematical values definite on the space of positive realities and not of the positive entireties (although a fraction of cycle does not have any experimental reality). These values are inversely proportional to the quantity of initial DNA, and uncertainty to the measure is minimized to the maximum (in general lower than 5%).

The quantification while passing by a mathematical value of number of cycle (CT) thus makes it possible to obtain reliable results, but is not exploitable directly. In order to obtain the quantity of initial DNA, it will be necessary to carry out new mathematical transformations which require to know the effectiveness of PCR. The latter is generally given thanks to a range of calibration.

Range of calibration

Graphic of left (A):

Last nine samples (F, G, H, I, J, K, L, M and NR) of concentration in initial DNA decreasing (of an order of magnitude each time) were amplified by PCR in the same experiment. Each kinetics made it possible to determine its own CT, given of number of cycle. The concentrations in DNA are known of many copies per tube (F ≈ 70 million and NR ≈ 0,7). The data represented correspond to experimentation “N”, example among a great number of duplicated independent in time, for the batches of the reagents and the experimenters. Fluorescence is expressed in arbitrary units and background noise was standardized. Note that no amplification was obtained for the N_n sample.

Graphic of right-hand side (B):

the averages of the CT according to the initial quantity of DNA of all duplicated range were deferred in a semi-logarithmic reference mark . The average calibration line was modelled using a linear Regression. Variability represented is not the SEM but the dispersion (or wide) of measurements. An average range was represented in order to illustrate the limits of the method, but each individual range is nevertheless extremely robust, with the significant figure of the coefficient of determination (R ²) generally to the fourth decimal after the comma (example: 0,999 6 )! It is thus useful to carry out the calibration lines in PCR on spreadsheet, because the software of the majority of the thermocycleurs is much less discriminating and can model only one individual range. This average range provides several information:

(1) detectable and quantitative phase the “of the PCR”: All the samples are detectable and are aligned with the other points of their individual range. This phase generally includes/understands concentrations of initial DNA energy of the hundred to the hundred million copies. In lower part, the stochastic phenomena become very perceptible but can be compensated by a multiplication of measurements. With the top, the phase of “background noise” is not enough any more important to be able to be correctly given. That could be compensated by a protocol of PCR having a very low effectiveness but it is generally simpler to dilute the sample.

(2) sometimes detectable but not-quantitative phase the “of the PCR”: It includes/understands the concentrations of initial DNA about the copy with ten copy. The percentage of detected samples (for this example) was indicated for the concentrations M and NR, is 83% for an average concentration of 7 copies and 28% when three tubes out of four contain a copy (concentration 0,7 or -0,15 in log). The dispersion of measurements (and thus the margin of error) become much more important. Note that a certain number of measurement M and NR is aligned suitably with their individual range, but this is with the chance. Only a multiplication of the ranges thus makes it possible to determine with precision this phase.

(3) a range standard of PCR becoming a line in a semi-logarithmic reference mark, it is possible to model it by an equation of the type Y = aX + B where:

*Y is the CT measured by the thermocyclor.

*La slope (A) is a function of the effectiveness of PCR, the latter being able to be calculated by the equation:

E = (index \; \; logarithm) ^ {\ frac {- 1} {slope}} = \ sqrt {index \; \; logarithm}

Notez that the diagram corresponds to the most frequent case where the concentration is expressed in decimal logarithm (= of index 10). This slope is often regarded as a constant for the amplification of a particular amplicon with a given experimental protocol. The slope or the effectiveness can be employed to quantify the samples.

*X is the concentration of initial DNA expressed in log (of copies/tube, ng/µl, arbitrary units, etc).

*B is a mathematical point which does not have any experimental reality (log 0 = 1/∞). It can nevertheless be used to gauge the experiments of PCR (often named “run”) between them. If the various ranges had been gauged by their intercept in the beginning (B), dispersion would have seemed much weaker, except for the points M and NR.

(4) the straight regression line does not pass in the center of dispersion various measurements for the concentrations M and NR, one notes a “damping” of the slope. If one considers CT average L with NR, they would be modelled better by a polynomial of the second degree. Certain software associated with the thermocycleurs makes it possible to take into account this “damping”. It is however advisable to note that:

*Cet “damping” is extremely variable of an individual range with another, and the modelled correction probably does not correspond to what occurs in the quantified sample. The commercial software models this “damping” on only one range, even if this one comprises obvious skews, and can thus induce very important errors that the user beginner will not detect inevitably.

*L' inaccuracy on the quantification with these concentrations is so important that a modeling of “damping” is not very relevant.

*Cet “damping” corresponds, when average ranges are carried out, to a lowering of the slope, therefore with an increase in the effectiveness of PCR, whereas the increase in the stochastic effects should lower it. It should be noted whereas this “damping” is generated mainly by the weakest concentration (NR, is 0,7 copies). However no amplification will be able to be made if there is not at least a molecule of initial DNA. The skew which results from it in the Gaussian distribution from the error is likely to cause this “average damping”.

It is necessary to be aware that if the range of calibration shows the quantitative aspect of the experimental protocol, it is difficult to avoid all skews potentiels, like a difference in chemical composition between solvent of the samples (complex medium of complementary DNA, presence of ARN, proteins, etc) and the thinner of the points of the range (generally of water).

Transformation of CT (or CP)

Maximum the of derived one second

Applications

developed of starters
detection of specific changes
proportioning of GMO in products for human consumption
quantification
specific and significant detection the pathogenic ones of veterinary interest - Quantification of the bacterial, viral or parasitic load

Random links: