Mar 6, 2007: Modeling the Evolution of Neutral DNA in Mammals
Filed in: Colloquium
Dr. John Karro, Pennsylvania State University
Over the course of evolution the DNA of an organism is constantly, if slowly, subject to mutations; beneficial mutations become fixed in the population, while deleterious mutations are culled out. However, a large portion of mammalian DNA is believed to have no biological function. Mutations of this neutral DNA have no effect on the organisms, and hence are not subject to the pressures of natural selection - whether they become fixed in the population becomes a matter of random chance. The ability to calculate the rate at which neutral DNA is subjected to change is of fundamental importance in our attempts to understand the structure, history and functions of the genome. But it is difficult to determine the rate of change of DNA over time when we have information on the DNA content at only one instance in time - as it exists now.
Further complicating matters, there is evidence that this rate not only changes between mammals, but it actually varies with the DNA's position in a given genome - resulting in a complex structure reflecting, and perhaps influencing, the organization of the genome.
In this presentation we will formalize the problem of understanding the neutral substitution rate of DNA and present a new method of modeling the substitution process. Through an application of Markov theory and an array of algorithmic techniques, we have developed new software for the analysis of the neutral substitution rate in mammals that allows us to not only study how this rate varies over a given genome, but also to look back at how that variation has changed over the ninety million years since the "radiation" of placental mammals. Through the application of the tool to modern data we have then been able to analyze the substitution rate and discover biological patterns that might help explain why the rate variation occurs.
While the problems we are solving are biological, the techniques and theory used to approach the problems fall into the realm of computer science - the target audience for this seminar. In the course of the talk we will also discuss the field of bioinformatics with a goal of explaining why this area can be of great interest to a computer scientist with no biological background. No knowledge of biology will be required to understand this talk, though some such knowledge may be inflicted.
John Karro is a research associate at the Pennsylvania State University. He received a Ph.D. in Computer Science from the University of Virginia, where he concentrated on the algorithmic problems associated with the physical design of VLSI-chips. Since leaving Virginia he has shifted is focus to bioinformatics, working on genome-level problems from areas including molecular evolution, comparative genomics, genome structure analysis and population genetics. To this end he spent a year at Yale University leading the development of the Yale Pseudogene Data. He is now at Penn. State under an NIH fellowship working on methods used to understand the evolution of mammalian genomes, with an eye towards developing new tools for the identification of functional regions in the human genome.
Abstract
Over the course of evolution the DNA of an organism is constantly, if slowly, subject to mutations; beneficial mutations become fixed in the population, while deleterious mutations are culled out. However, a large portion of mammalian DNA is believed to have no biological function. Mutations of this neutral DNA have no effect on the organisms, and hence are not subject to the pressures of natural selection - whether they become fixed in the population becomes a matter of random chance. The ability to calculate the rate at which neutral DNA is subjected to change is of fundamental importance in our attempts to understand the structure, history and functions of the genome. But it is difficult to determine the rate of change of DNA over time when we have information on the DNA content at only one instance in time - as it exists now.
Further complicating matters, there is evidence that this rate not only changes between mammals, but it actually varies with the DNA's position in a given genome - resulting in a complex structure reflecting, and perhaps influencing, the organization of the genome.
In this presentation we will formalize the problem of understanding the neutral substitution rate of DNA and present a new method of modeling the substitution process. Through an application of Markov theory and an array of algorithmic techniques, we have developed new software for the analysis of the neutral substitution rate in mammals that allows us to not only study how this rate varies over a given genome, but also to look back at how that variation has changed over the ninety million years since the "radiation" of placental mammals. Through the application of the tool to modern data we have then been able to analyze the substitution rate and discover biological patterns that might help explain why the rate variation occurs.
While the problems we are solving are biological, the techniques and theory used to approach the problems fall into the realm of computer science - the target audience for this seminar. In the course of the talk we will also discuss the field of bioinformatics with a goal of explaining why this area can be of great interest to a computer scientist with no biological background. No knowledge of biology will be required to understand this talk, though some such knowledge may be inflicted.
Bio:
John Karro is a research associate at the Pennsylvania State University. He received a Ph.D. in Computer Science from the University of Virginia, where he concentrated on the algorithmic problems associated with the physical design of VLSI-chips. Since leaving Virginia he has shifted is focus to bioinformatics, working on genome-level problems from areas including molecular evolution, comparative genomics, genome structure analysis and population genetics. To this end he spent a year at Yale University leading the development of the Yale Pseudogene Data. He is now at Penn. State under an NIH fellowship working on methods used to understand the evolution of mammalian genomes, with an eye towards developing new tools for the identification of functional regions in the human genome.