This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 681
Type: Contributed
Date/Time: Thursday, August 5, 2010 : 10:30 AM to 12:20 PM
Sponsor: Section on Statistical Computing
Abstract - #309066
Title: Exact Distribution of Prediction Error Rates for Protein and Domain Interactions
Author(s): Donald E.K. Martin*+ and John Aston
Companies: North Carolina State University and University of Warwick
Address: 4272 SAS Hall, Raleigh, NC, 27695,
Keywords: protein and domain interactions ; sum-product algorithm ; recursive computation ; computational statistics
Abstract:

The study of interactions of proteins and of the domains that comprise them helps with understanding protein functions. High-throughput methods of measuring protein interactions have high error rates; however structural similarities between proteins may be exploited to make model-based predictions of their interactions, and of interacting domains. A Bayesian network may be used to model protein and domain interactions in a unified fashion, and the sum-product algorithm allows the computation of marginal probabilities of interactions. Proteins with probabilities above a threshold are predicted as interacting. Using this setup, we give an algorithm to compute model-based prediction error rate distributions, conditional on noisy measurements. The methodology applies to hidden states of undirected and directed graphical models represented using conditional random fields and factor graphs.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 JSM Online Program Home

For information, contact jsm@amstat.org or phone (888) 231-3473.

If you have questions about the Continuing Education program, please contact the Education Department.