Predoctoral Trainee

Joseph P. Bockhorst Computer Sciences Doctoral Training Program Dept. of Computer Sciences Faculty Supervisor: Mark Craven Email: joebock@biostat.wisc.edu Computer Sciences web page: http://www.cs.wisc.edu/~joebock/ 265-8612

My research involves the study of machine learning algorithms with a focus on those applicable to the analysis of bioinformatics problems. In particular, I am interested in investigating the properties of relationships among machine learning tasks. This entails identifying and defining different types of relationships and in developing and evaluating a set of machine learning methods that exploit these relationships to yield more accurate models. Currently, I am exploring how the relationships among regulatory elements in bacteria, such as promoters, terminators and operons, can be best used to train probabilistic models of these elements.

I will discuss methods for prediction of regulatory elements in bacterial genomes. Supervised learning techniques, in which models are induced from training examples, are used to build models of operons and terminators. Our operon prediction method constructs a probabilistic operon model from a variety of data sources including gene expression and sequence data. Our terminator prediction method constructs a probabilistic sequence model trained to discriminate sequences containing terminators from those that do not. Both methods have been shown to make accurate predictions. One of the interesting aspects of these prediction tasks is that they are strongly related in that predictions of one model can provide useful information to the other and vice-versa. I will also discuss how these relations may be exploited to learn more accurate models.

Resume

Resume 2002 (.pdf format, Download Reader)

Publications

Bockhorst J and Craven M(submitted). Markov networks for detecting overlapping elements in sequence data. 20th International Conference on Uncertainty in Artificial Intelligence 2004.

Bockhorst J, Craven M, Page D, Shavlik J and Glasner J. A Bayesian network approach to operon prediction. Bioinformatics 19(10): 1227-35, 2003.

Bockhorst J, Qiu Y, Glasner J, Liu M, Blattner F and Craven M. Predicting bacterial transcription units using sequence and expression data. Bioinformatics 19 Suppl 1: i34-43, 2003.

Bockhorst J, Qiu Y, Glasner J, Liu M, Blattner F, and Craven M. Predicting bacterial transcription units using sequence and expression data. Proceedings of the 11th International Conference on Intelligent Systems for Molecular Biology, 2003.

Bockhorst J, Qiu Y, Glasner J, Liu M, Blattner F, and Craven M. Predicting bacterial transcription units using sequence and expression data. Bioinformatics 19 Suppl 1: I34-I43, 2003.

Bockhorst J, Craven M, Page D, Shavlik J, and Glasner J. A Bayesian network approach to operon prediction. Bioinformatics 19: 1227-1235, 2003.

Bockhorst J and Craven M. Exploiting relations among concepts to acquire weakly labeled training data. Proceedings of the 19th International Conference on Machine Learning, 2002.

Craven M, Page D, Shavlik J, Bockhorst J, and Glasner J. A probabilistic learning approach to whole-genome operon prediction. Proc. Int. Conf. Intell. Syst. Mol. Biol. 8: 116-127, 2000.

Bockhorst J and Craven M. Refining the Structure of a Stochastic Context-Free Grammar. Proceedings of the 17th International Joint Conference on Artificial Intelligence. Seattle, WA. Morgan Kaufmann, 2001.

Search for other publications by Joseph Bockhorst (Pub Med, Cite Seer)

Feedback, questions or accessibility issues.
Graduate Program  |  Postdoctorate Program  |  Short Term Research Program

File last modified: August 14, 2003
Copyright © 2003 The Board of Regents of the University of Wisconsin System.