Atlas home || Conferences | Abstracts | about Atlas
Host: University of British Columbia
Sponsor: PIMS
Homepage: http://www.pims.math.ca/sections/activities/geno.html
Organizers: David Sankoff, Michael Waterman
Description:
Genomics is concerned with the characterization and analysis of genetic material from a wide range of organisms including man. Within the next ten years, the complete genome sequences
of human and many other organisms will be determined. It is an exciting time for genomics with the current success at generating huge amounts of basic data. There is a continuing concern
with generating this data in ever more efficient and inexpensive ways. In addition the big new challenge must be met: to decode the information in the genome. For example, just to find all
the genes is a much harder problem than it was imagined in the early days of molecular biology. After the genes are identified, it is necessary to determine how the gene products function
and interact. In addition, our ideas of evolution will be developed by the data coming from genomics. This session will bring together people from a variety of disciplines including
molecular biology, evolutionary biology, discrete mathematics, and statistics. New technologies will be developed and applied to obtain new genetic data which in turn must be interpreted
and analyzed by new techniques from the mathematical sciences.
One of the most interesting areas in mathematical biology is modeling evolutionary processes. Bringing theory into the mass of sequence data is one of the most challenging problems we have. The details of modeling triplet repeats (that cause many human diseases) are not yet entirely worked out. Using homologous sequences to infer evolutionary relationships continues to be of great interest, and new simulation techniques are being employed in this area. Sankoff and others have brought into prominence the distance between genomes problems. The most elegant of these is to compute the reversal distance between two gene orders.
Finding patterns in sequences is a major activity. The database search methods in BLAST and its variants use sophisticated computer science and statistics. Estimating the statistical significance of matches between words (molecular sequences) requires some sophisticated tools. Hidden Markov Models (HMMs) are used to estimate more sophisticated statistical patterns. The area of determining DNA coding for proteins is called ``gene finding'' and has become very active. The importance in the era of genome sequencing cannot be overestimated.
Finally we mention the area of DNA structure. (Protein structure is another major area of activity, but we have not made that part of this proposal as it seems to enlarge the scope too much.) DNA structure, particularly that of closed circular duplex DNA, is central to understanding the general chemistry and biology of DNA. Of relevance are topological, geometric and elastic properties. Twist, writhe, and linking numbers have been studied. X-ray crystallography and NMR are powerful methods that have given important information about protein structure, but are not well suited to studying the dynamic processes of the cell. Applications of non-obvious topological and geometrical methods have already yielded some valuable insights.
Speakers: Nadia El-Mabrouk, Joseph Felsenstein, Terry Gaasterland, Philip P. Green, Tom Hagedorn, Steven Henikoff, Susan Holmes, Leroy Hood, Richard M. Karp, John Kececioglu, Gary D. Stormo, Elisabeth Tillier
Date received: May 06, 1999
© 2008 Atlas Conferences Inc.