A Bayesian method for calculating real-time quantitative PCR calibration curves using absolute plasmid DNA standards

Area of Science:

• Molecular Biology
• Biostatistics
• Quantitative Analysis

Background:

• Real-time quantitative PCR (qPCR) uses calibration curves for DNA concentration estimation.
• Differences between plasmid and genomic DNA amplification are often overlooked.
• A statistical method is needed to address uncertainties in calibration curve parameters.

Purpose of the Study:

• To develop a Bayesian approach for generating calibration curves in qPCR.
• To estimate DNA concentrations from genomic samples using plasmid DNA standards.
• To account for various sources of uncertainty in calibration.

Main Methods:

• Utilized a Bayesian statistical approach.
• Employed Monte Carlo Markov Chain (MCMC) estimation for calibration curves.
• Used WinBUGS software for simulations and posterior distribution generation.

Main Results:

• Generated single, master, and modified calibration curves using MCMC.
• Point and interval estimates for parameters like intercepts, slopes, and DNA concentrations were derived from posterior distributions.
• The Bayesian method provided robust parameter estimates.

Conclusions:

• The Bayesian approach enables accurate DNA concentration estimation from environmental samples using qPCR.
• It effectively addresses uncertainties from experiment-to-experiment variation and replicate measurements.
• This method accounts for variability introduced by using plasmid DNA standards.