Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Nonlinear Modeling of Real-Time Reverse Transcription Pcr Data from Multiple Independent Samples for Statistically Verified Determination of Relative Mrna Expression Changes

Author
item Bayles, Darrell

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: January 31, 2006
Publication Date: May 21, 2006
Citation: Bayles, D.O. 2006. Nonlinear modeling of real-time reverse transcription pcr data from multiple independent samples for statistically verified determination of relative mrna expression changes. Meeting Abstract. CD-ROM ISBN 1-55581-389-5

Technical Abstract: Quantitative measurement of gene expression through relative real-time reverse transcription PCR (RT-PCR) has become a leading tool for sensitive determination of changes in mRNA transcript levels for a variety of molecular biology experiments. The methods currently available for analyzing the RT-PCR fluorescence data potentially bias the analyses due to various assumptions which include: i) that amplification efficiency is equal for both reference and target genes, ii) selection of which data points adequately describe the linear portion of a log-linear transformation of a RT-PCR growth curve, iii) there is little experimental variation, and iv) there are no effects due to arbitrary selection of cycle threshold for the analysis. We have developed an approach that does not require these assumptions by using an appropriate level of experimental replication followed by background subtraction of the raw RT-PCR data and fitting of a reparameterized Gompertz exponential growth model to the logarithmic transformation of the data. The SAS/STAT NLMIXED procedure is used to fit the model to the fluorescence data curves and to provide parameter estimates that allow the user to verify that amplification efficiencies for the specific genes are indistinguishable between treatments. Checking for indistinguishable amplification efficiencies between treatments provides quality control for the real-time PCR step. Our program code provides the estimate of the relative expression ratio for the target gene along with the 95% confidence interval. To verify our analysis method we used six independent mRNA samples, isolated from Listeria monocytogenes grown in broth, to build a model system whereby the mRNA of a target gene (htrA or pgpH) differed either by 2- or 4-fold, but where the reference gene (rpoB) mRNA concentration remained constant. After reverse transcription and RT-PCR using a fluorescent reporter dye, our method quantitatively detected both the 2- and 4-fold changes in relative expression. This analysis approach is generally applicable, RT-PCR platform independent, and independent of the mRNA source.

Last Modified: 12/21/2014
Footer Content Back to Top of Page