Keynotes

Tuesday | Wednesday | Thursday

Tuesday, January 29

Bioinformatics is a true blend of computer science, software engineering and biology. The complexity of the information is such that there are few "off the shelf" solutions to bioinformatics problems; many problems have to be solved in place with people who understand as much of the system as possible. Only being able to understand the entire system can one tweak the code giving the correct one-off solution to the problem at hand.

Given such an emphasis on tailored solutions, it is not surprising that open source libraries have a massive impact on bioinformatics. These libraries range from the NCBI toolkit (home of the workhorse of bioinformatics, BLAST) through the scripting language projects like BioPerl to advanced projects such as BioLisp and hidden Markov model libraries. The open nature of this software development is perfectly aligned with openness in science and data release in genomics.

Ewan Birney shows how open source development has had an impact from the bench scientist to genome data management. He’ll describe how the open source movement in bioinformatics has matured from a group of twenty-somethings rescuing a 486 from a dumpster to be our server through to a non-profit organization marshalling over 200 active developers worldwide.

Lincoln Stein discusses the fragmentation of bioinformatics protocols, technologies and standards, that together created a landscape of confusing and poorly integrated web sites and other services. He reviews past and current efforts to reduce fragmentation, and speculates on ways to move past the medieval city-state mentality toward a nation of cooperative bioinformatics services.

Wednesday, January 30

Data integration is a growing field in bioinformatics, as researchers combine information from multiple diverse data sets to learn about and explain natural processes. Methods have been developed to integrate insights from hybridized cDNA and oligomer microarray data with genome sequence annotations. The integrated data enables the use of genome annotations to explain gene expression patterns and to compare gene expression patterns for orthologous genes from different organisms.

Terry Gaasterland discusses the integration project, with examples from the genomes of Trypanosoma brucei (a human and cattle parasite), six strains of Staphylococcus aureus (a pathogenic bacterium), human and mouse.

The Biomolecular Interaction Network Database, BIND is the fastest growing database in Bioinformatics. A torrent of new information from the yeast model organism has been added thanks to two large scale experiments in mass spectromety published in the Jan 12 2002 issue of Nature. The combined dataset of all known yeast interactions together with a new complex-finding algorithm recently published in the Jan 11 2002 issue of Science, gives a compelling glimpse of the overall structure of the eucaryotic nucleolus, the large nuclear organelle that assembles ribosomes, and is one of the largest structures in a eucaryotic cell. The BIND database sits atop significant infrastructure in Bioinformatics including the NCBI data model, an integrated database called SeqHound and high-performance cluster computing to compute protein sequence neighbors. An overview of the new scientific results obtained by combining this data together over the past two weeks will be presented as well the global synthesis of molecular assembly information and the underlying architecture of the BIND system.

Thursday, January 31

Since its founding in 1988, the National Center for Biotechnology Information (NCBI) has grown from a small group of 12 people to one of the largest and most heavily used information resources in biomedicine. James Ostell discusses some of the underlying principles used to organize and prioritize NCBI's efforts over that last decade. Those principles are illustrated by an architecture overview of the Entrez retrieval system, and examples of using NCBI resources to make discoveries. Finally some of the newer NCBI resources will be touched upon briefly and some future directions considered.

The Human Genome Project has altered the view and practice of biology and has led to several paradigm changes in systems biology, and predictive and preventive medicine. I will discuss these changes and consider the analysis of two biological systems: galactose metabolism in yeast, and sea urchin development, using integrative systems approaches.