Virtual Cell Program, Department of Systems Biology, Harvard Medical School

Applying machine learning to problems in bio-informatics

Yoav Freund
Center for Computational Learning Systems
Columbia University

Biography

Dr. Freund is an internationally known researcher in the field of machine learning, a field which bridges computer science and statistics. He is best known for his joint work with Dr. Schapire on the Adaboost algorithm. For this work they were awarded the 2003 Gödel prize in Theoretical Computer Science.

Abstract

In the last year I have devoted most of my time to the application of machine learning techniques to problems in bio-informatics. In this talk I will show how machine learning methods are used in Columbia University to extract information from a variety of experimental data sources. I will cover:

Boosting and Alternating trees.
Using boosting to analyze gene regulation networks in yeast.
Using machine learning for image analysis, with emphasis on using selective sampling to reduce human labor.

I might cover some of:

Using stochastic models and selective sampling to improve the accuracy of information extraction from journal texts.
Mapping the space of protein structures using the Information Bottleneck method.

Through these examples of my work and the work of others in CU, I will demonstrate the great potential of machine learning tools for advancing experimental methods in systems biology.

References

Manuel Middendorf, Anshul Kundaje, Chris Wiggins, Yoav Freund and Christina Leslie, "Predicting Genetic Regulatory Response Using Classification", to appear in Proceedings of the International Conference on Intelligent Systems for Molecular Biology 2004. PDF

Yoav Freund and Robert E. Schapire, "A short introduction to boosting", Journal of Japanese Society for Artificial Intelligence, 14:771-780, 1999. PDF.