Bayesian Approach to Protein Lysate Array Quantification

• Activity Number: 347
• Type: Contributed
• Date/Time: Tuesday, July 31, 2012 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Ji-Yeon Yang*+ and Xuming He and Feng Liang
• Companies: MD Anderson Cancer Center and University of Michigan and University of Illinois at Urbana-Champaign
• Keywords: Protein lysate array ; Dilution series ; Bayesian hierarchical model

Abstract

The protein lysate array (Reverse phase protein array) is an emerging technology for quantifying protein concentration ratios in multiple biological samples. Most previous studies of protein array quantification are based on the assumption that the errors are independent. However, the nature of the experiments and the examination of the residuals obtained under the i.i.d. error assumption warrant the existence of correlation of the data.

This paper considers the complexity of the correlation structure and introduces a new model that can allow for the dependence structure of the errors. We propose a Bayesian hierarchical model for statistical inference. The results from simulations and real data indicate that the hierarchical model outperforms the existing methods if the i.i.d. error assumption does not hold, and still present comparable results even if the i.i.d. error assumption holds.