ITR/IM+AP: Automated Compilation and Computational Analysis of Regulatory Networks

EIA-0121687

Rzhetsky, Andrey

Columbia University

ITR/IM+AP: Automated compilation and computational analysis of regulatory networks

This project will provide advanced information management systems as well as software applications relevant to the fields of bioinformaticis, molecular biology, and medicine. It will contribute to fundamental Information Management by developing advanced algorithms for the efficient retrieval of information about biological pathways from research articles (using natural language processing techniques), machine-learning-assisted data mining in biological texts, and information management via a uniquely designed database with two levels of representation of biological information.

The specific aims of the project are the following: to implement and test natural language processing techniques for the automatic retrieval of signal-transduction pathways from the research literature; to develop general and portable text-mining techniques for the automatic recognition of genes, proteins, and other domain terms, and of relationships between them; and to develop a mathematical framework for the statistical analysis of heterogeneous data on regulatory pathways.