The Parallel Genetic Algorithm for Designing DNA Randomizations in a Combinatorial Protein Experiment
Evolutionary methods of protein engineering such as phage display have revolutionized drug design and the means of studying molecular binding. In order to obtain the highest experimental efficiency, the distributions of constructed combinatorial libraries should be carefully adjusted. The presented approach takes into account diversity–completeness trade–off and tries to maximize the number of new amino acid sequences generated in each cycle of the experiment. In the paper, the mathematical model is introduced and the parallel genetic algorithm for the defined optimization problem is described. Its implementation on the SunFire 6800 computer proves a high efficiency of the proposed approach.
KeywordsGenetic Algorithm Phage Display Parallel Genetic Algorithm Length Versus Protein Library
Unable to display preview. Download preview PDF.
- 1.Steipe, B.: Evolutionary Approaches to Protein Engineering. In: Famulok, M., Winnacker, E.-L. (eds.) Current Topics in Microbiology and Immunology: Combinatorial Chemistry in Biology, vol. 243, pp. 56–81 (1999)Google Scholar
- 6.Dziurdza, B., Błażewicz, J., Markiewicz, W.T.: Artificial evolution. Algorithms for searching for amino acid sequences. In: RECOMB 2003. Currents in Computational Molecular Biology, pp. 151–152 (2003)Google Scholar
- 8.Dziurdza, B., Błażewicz, J., Markiewicz, W.T.: Working papers in Institute of Computing Science, Poznan University of Technology (2004)Google Scholar
- 10.Wall, M.: GAlib C++ library (2005), http://lancet.mit.edu/ga/
- 11.MPICH, free Message Passing Interface library, http://www-unix.mcs.anl.gov/mpi/mpich/