Self-adaptive Evolutionary Algorithm for DNA Codeword Design
Abstract
DNA has emerged as a new computational resource for data encoding and processing. The fundamental problem of DNA Codeword Design (CWD) calls for finding effective ways to encode and process data in DNA. The problem has shown to be of interest in other areas as well, including computational memories, self-assembly and phylogenetic analysis, among others. In prior work, a framework to analyze this problem has been developed and simple versions of CWD have been shown to be NP-complete using any single reasonable metric that approximates the Gibbs energy, thus practically making it very difficult to find a general procedure for finding optimal efficient encodings. We present a Self-adaptive Evolutionary Algorithm for CWD (SaEA-CWD) as an extension of the Hybrid Adaptive Evolutionary algorithm (HAEA). SaEA-CWD is a parameter adaptation technique that automatically adapts the rates of its genetic operator applications to exploit structural properties of the search space to improve the speed and quality of the solutions. An implementation and preliminary results are evaluated in spaces where searches are already prohibitive to ordinary methods (such as 8- and 10-mers) due to the combinatorial explosion of the solution DNA space. Applications to other problems are suggested, such as a general technique for dimensionality reduction based on SaEA-CWD.
Publication Title
2018 IEEE Congress on Evolutionary Computation, CEC 2018 - Proceedings
Recommended Citation
Prieto, J., León, E., & Garzon, M. (2018). Self-adaptive Evolutionary Algorithm for DNA Codeword Design. 2018 IEEE Congress on Evolutionary Computation, CEC 2018 - Proceedings https://doi.org/10.1109/CEC.2018.8477827
