(361j) Extracting Thermodynamic and Fluorescent Properties of Intercalating Dyes from Temperature-Programmed PCR Measurements with Modeling and Optimization

Fluorescent dyes that intercalate with DNA are commonly used in real-time Polymerase Chain Reaction (PCR) instruments. In this talk, we present a mathematical model which provides a quantitative relationship between the measured fluorescence signal and underlying biochemical phenomena. The model accounts for the partitioning of dye between solution and DNA, as well as the fluorescence of intercalated dye. A fixed number of DNA strands, subject to a series of step decreases in temperature, is considered. The model predicts a measured signal to be linear in total dye concentration. The experimental measurements possess trends that deviate from linearity, due to noise both within the same 96-well plate and between different plates. Numerical optimization allows for ascertainment of the linear trends in the experimental measurements, accounting for error both in signal and total dye concentration. Uncertainty quantification and propagation of optimal solutions associated with different DNA concentrations allows for calculation of dye partitioning coefficients and adsorbed molar fluorescences. It also sheds new light on the role of noise in total dye concentration in this important measurement technology. The temperature dependence of the partitioning coefficients, in turn, allows for calculation of thermodynamics of intercalation. All properties calculated separately for single-stranded and double-stranded DNA are consistent with biochemical intuition.