Evolutionary Algorithm that Designs the DNA Synthesis Procedure

Abstract

Chemical synthesis of nucleotide chains is very erroneous for long sequences. Often a gene is constructed from short fragments joined with the use of complementary helper chains. The number of possible potential solutions for a long gene synthesis is very large, therefore a fast automated search is required. In the presented approach a modified method of long DNA construction is proposed. A computer program that searches for an optimal solution in the space of potential synthesis methods has been developed. This software uses an evolutionary algorithm for global optimization and a hillclimbing algorithm for local optimization. The long DNA construction method was tested on random sequences. The results are very promising. The next step is to perform experiments in a biotechnological wet laboratory involving DNA strand synthesis using the method designed by the presented software.