University of Colorado at Denver and UC Health Sciences Center Center for Computational Biology

Workshop Announcement

To register and obtain more information, visit

Bioinformatics: Inference in High-throughput Molecular Biology

Imran Shah, UCHSC Dept. Preventive Medicine & Biometrics / Dept. Pharmacology

February 22, 2002

Room 480, CU Building, 1250 Fourteenth St., Denver

Build collaboration prospects between researchers in molecular biology and medicine with computer scientists, statisticians, probabilists and other quantitative people.

Target audience:
Computer scientists, statisticians, probabilists and computationally inclined scientists who want to learn the challenging problems in molecular biology with medical applications (no background in molecular biology is assumed)

Biomedicine is entering a new era of high-throughput data production. The macromolecular sequence databases are doubling in size every 18 months or so, and now contain more than 7 million sequences representing more than 9 billion nucleotides. Gene chips allow for the simultaneous assaying of the expression levels of thousands of genes at a time. Related technologies allow the identification of millions of point mutations in particular individuals, the simultaneous screening of tens of thousands of compounds for drug-like binding affinity, and the relatively rapid determination of three-dimensional macromolecular structures. The challenges inherent in analyzing this onslaught of extraordinarily interesting data are defining the new field of bioinformatics. Making biologically relevant inferences from massive data sets is a key aspect of the field, drawing on techniques from statistics, stochastic processes, pattern recognition, and machine learning. In this overview, several representative problems and solution techniques will be presented, including hidden Markov models, support vector machines, Bayesian networks, and pathway inference.