A nonlinear mixed-effects model for estimating calibration intervals for unknown concentrations in two-color...

Area of Science:

• Bioinformatics
• Statistical Genomics
• Molecular Biology

Background:

• Accurate normalization is crucial for microarray data analysis.
• Spiked-in experiments are essential for validating normalization methods.
• Existing methods may lack robust confidence interval estimation.

Purpose of the Study:

• To develop a new calibration method for preprocessing spiked-in microarray data.
• To estimate normalized absolute expression values using nonlinear mixed-effects models.
• To construct confidence intervals for estimated expression values.

Main Methods:

• Utilizing nonlinear mixed-effects models for data preprocessing.
• Employing a spike-in calibration curve to estimate expression values.
• Leveraging asymptotic properties for confidence interval construction.
• Performing simulations to validate the accuracy of confidence intervals.

Main Results:

• The proposed method accurately estimates normalized absolute expression values.
• Confidence intervals constructed using the method achieve desired coverage probabilities.
• Simulations demonstrate the robustness of the confidence interval approximations.
• The method was successfully applied to two public cDNA microarray datasets.

Conclusions:

• The developed calibration method provides accurate normalization and reliable confidence intervals for spiked-in microarray experiments.
• The method is adaptable for preprocessing one-color oligonucleotide microarray data.
• This approach enhances the reliability and interpretability of microarray expression data analysis.