Tutorial - Introduction to Chemoinformaticss

Presenter: Pierre Baldi, School of Information and Computer Sciences and the Department of Biological Chemistry, University of California, Irvine

Description

This self-contained tutorial will provide an overview of chemoinformatics, from foundations to state-of-the-art results and challenges. It will cover molecular and reaction data, data structures and the available algorithms for efficiently searching large repositories and annotating or predicting the physical, chemical, and biological properties of compounds and reactions with applications ranging from chemical genomics to drug discovery. The tutorial will leverage analogies and create synergies between bio and chemical informatics. It aims to bring bioinformaticians up to speed with the state-of-the-art in chemoinformatics methods, by exploring similarities and differences between bioinformatics and chemoinformatics. The two main driving forces behind the bioinformatics expansion have been: (1) the development of high throughput methods and the corresponding public availability of large repositories (GenBank, Swissprot, PDB, etc); and (2) the development of search algorithms (BLAST) and related statistical machine learning techniques to analyse the data. Mutatis mutandis, the same is true of chemoinformatics, with the caveats that large repositories of chemical data have started to become available only very recently, over the last two years or so. Many of the basic concepts in bioinformatics (similarity, search, alignments, kernels, data structures, etc) have applications in chemoinformatics. From a scientific standpoint, chemoinformatics is relevant to bioinformatics since it plays a key role in several applications such as chemical genomics and drug discovery/screening/design applications. The tutorial is introductory in the sense that it is self-contained, but it will also cover advanced topics and present the state-of-the-art, as well as the main open challenges, in chemoinformatics.

No prior knowledge is expected. However basic familiarity with organic chemistry is most desirable. Basic understanding of databases and/or statistics and machine learning is also desirable, but not necessary.

Biosketch of Presenter:

Pierre Baldi is a Professor in the School of Information and Computer Sciences and the Department of Biological Chemistry at the University of California, Irvine and the Director of the UCI Institute for Genomics and Bioinformatics. He received a PhD in Mathematics in 1986 from the California Institute of Technology.

He has held postdoctoral, faculty, and member of the technical staff positions at UCSD and Caltech, in the Division of Biology and the Jet Propulsion Laboratory. He was CEO of a startup company for a few years and joined UCI in 1999. He is the recipient of a 1993 Lew Allen Award at JPL and a Laurel Wilkening Faculty Innovation Award at UCI. Dr. Baldi's has published over 150 scientific articles and four book. Research in his group focuses on several areas at the intersection of the computational and life sciences, in particular the application of AI/statistical/machine learning methods to problems in bio and chemical informatics. Work in his group has resulted in several databases, software, and web servers that are widely used (www.igb.uci.edu/servers/servers.html). His main contributions include the development of Hidden Markov Models (HMMPro) for sequence analysis, recursive neural networks for de novo protein structure prediction (SCRATCH), Bayesian statistical methods for DNA microarray analysis (Cyber-T), informatics infrastructure for systems biology (SIGMOID) and, more recently, databases and tools in chemical informatics (ChemDB) for the prediction of molecular properties and applications in chemical synthesis, discovery, and drug design.