Science at the Shine Dome 2010
Symposium: Genomics and mathematics
Friday, 7 May 2009
Professor Terry Speed FAA
Terry Speed completed an undergraduate mathematics and statistics degree at the University of Melbourne (1961-64), and a PhD in mathematics at Monash University (1965-69). He has taught mathematics or statistics at the University of Sheffield (1969-73), the University of Western Australia (1974-82), and the University of California, Berkeley (1987-2009), and has had a spell with CSIRO's then Division of Mathematics and Statistics (1983-87).
Since 1997 Terry has been with the Walter and Eliza Hall Institute of Medical Research, initially 50:50 with UC Berkeley, but full-time since mid-2009. He has been interested in genetics since studying that subject as an undergraduate, and these days his research is in the application of statistics and mathematics to problems in genetics and molecular biology, which includes genomics and several other omics.
Integrating science, technology, data and mathematics
Many of us will be familiar with Moore’s law, an empirical relationship describing exponential trends in the history of computing hardware. Initially concerning the number of transistors on an integrated circuit, it was later seen to be applicable to increases in processing speed, memory capacity, and other aspects of digital electronics. Less well known, but of importance for biology, are similar exponential trends in the amount of DNA sequence and related genomic data in public repositories, and in the capacity of devices to generate such data. Less well known still is one focus of today’s symposium: the need for the co-evolution of computational, mathematical and statistical methods for the storage, retrieval, display, analysis and interpretation of genomic data to keep pace with the rapidly growing amounts of such data. While we are unlikely to see the headline ‘Mathematician cures cancer!’ or ‘Mathematician cures malaria!’ any time soon, the mathematical sciences are becoming increasingly important in areas such as cancer or malaria research, due to the nature and quantities of data now being collected. These data are becoming available as a result of incredible advances in genomic data-generating technologies. Indeed, in the last two years the rate of genomic data generation has shot well above the steady exponential levels seen over the last quarter of a century, challenging not only the mathematical sciences, but also the digital processing and data storage capacities mentioned at the outset. These are exciting and challenging times for genome science and the mathematical sciences. In my talk, I will try to convey a sense of this excitement and these challenges as an introduction to the symposium.
