Inferring Protein Associations Using Protein Pull-Down Assays

• Activity Number: 104
• Type: Contributed
• Date/Time: Monday, August 7, 2006 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Julia Sharp*+ and Kevin K. Anderson and Don S. Daly and Deanna L. Auberry and John Borkowski and William R. Cannon
• Companies: Montana State University and Pacific Northwest National Laboratory and Pacific Northwest National Laboratory and Pacific Northwest National Laboratory and Montana State University and Pacific Northwest National Laboratory
• Keywords: protein association ; likelihood ratio test ; Monte Carlo simulation

Abstract

One method to reveal protein-protein interactions is a "bait-prey pull-down" assay using a protein affinity agent and a protein detection method that is prone to various errors. A pull-down study generates a presence/absence matrix wherein each column or row corresponds to a bait protein sample or prey protein, respectively. Our method evaluates the presence/absence pattern across a row with a Likelihood Ratio Test (LRT) and simulated LRT test statistic distributions, checking the statistical assumptions with simulated binomial random variates. Each prey protein is assigned a category (specific, non-specific, or not associated) that is then appraised with respect to a pull-down study's goal and design. The LRT p-values objectively reveal specific and ubiquitous prey, as well as potential systematic errors. The LRT screen reduces the bias introduced when applying ad hoc filtering rules.