PhD defense by Christian Theil Have
Wednesday February 6, 2013 at 13.00 – 16.00 in room 43-2.29
Christian Theil Have defends his thesis:
Efficient Probabilistic Logic Programming for Biological Sequence Analysis
Abstract
This thesis is concerned with the application of probabilistic logic programming to biological sequence analysis. Probabilistic logic programming is a declarative programming paradigm which enables convenient and concise expression of a wide range of statistical models including common models from biological sequence analysis. A main advantage of this approach is that logic and control is separated – this enables that general inference algorithms can be reused for any model which is expressed using probabilistic logic programming. The dissertation demonstrates the usefulness of probabilistic logic programming for biological sequence analysis through applications in bacterial gene finding and through programming abstractions that enable convenient expression of constraints from the domain of biological sequence analysis. To deal with central limitations with regard to the efficiency of probabilistic logic programming, a number of optimizations of the approach are introduced. With these optimizations, probabilistic logic programming is now efficient enough to deal with a wide range of problems from biological sequence analysis.
Assessment Committee
Troels Andreasen, Associated Professor, Computer Science, CBIT, RUC, DK (chairman)
Marina De Voos, Senior Lecturer of Computer Science, University of Bath, UK
Anders Krogh, Professor, Bioinformatics, University of Copenhagen, DK
Head of defense
Keld Bødker, Head of Doctoral School, CBIT, Roskilde University, DK
Supervisor
Henning Christiansen, Professor, Computer Science, CBIT, Roskilde University, DK
Thesis is available for examination here.
After the defense CBIT will host a reception.