Related Experiment Video
Updated: Apr 5, 2026
12:39
A Novel Bayesian Change-point Algorithm for Genome-wide Analysis of Diverse ChIPseq Data Types
Published on: December 10, 2012
11.8K
A Bayesian hierarchical model for inference across related reverse phase protein arrays experiments
Riten Mitra1, Peter Müller2, Yuan Ji3
1ICES, University of Texas at Austin, Austin, TX, USA.
Journal of Applied Statistics
|August 7, 2015
Summary
This study introduces a new hierarchical model for analyzing functional proteomics data. The model helps understand protein interactions over time, even with limited experimental data for each drug condition.
Area of Science:
- Proteomics
- Systems Biology
- Statistical Inference
Background:
- Functional proteomics experiments track protein activation dynamics.
- Understanding protein dependence structures is crucial for biological insights.
- Limited data per drug/dose hinders traditional inference.
Purpose of the Study:
- To develop a robust statistical method for inferring protein dependence structures.
- To address data scarcity challenges in functional proteomics.
- To enable meaningful analysis of protein activation over time.
Main Methods:
- A novel hierarchical modeling approach is proposed.
- The model incorporates shared dependence structures.
- Latent binary indicators of protein activation are utilized.
Main Results:
- The hierarchical model facilitates inference on protein dependence.
- It effectively handles limited data by sharing information across conditions.
- Enables robust analysis of complex protein activation patterns.
Conclusions:
- The proposed hierarchical model is a powerful tool for functional proteomics.
- It overcomes data limitations inherent in drug perturbation studies.
- Provides a framework for deeper understanding of protein network dynamics.
