A genetic algorithm for designing microarray experiments

• Published in: Computational statistics Vol. 31; no. 2; pp. 409 - 424
• Main Authors: Latif, A. H. M. Mahbub, Brunner, Edgar
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2016; Springer Nature B.V
• Subjects: Algorithms; Arrays; Computer simulation; Design of experiments; Economic Theory/Quantitative Economics/Mathematical Methods; Experiments; Factorial experiments; Factorials; Gene expression; Genes; Genetic algorithms; Heuristic; Mathematics and Statistics; Optimization; Optimization techniques; Original Paper; Probability and Statistics in Computer Science; Probability Theory and Stochastic Processes; Searching; Statistics; Studies; Texts; Design criteria; Microarray experiments; Experimental design; Genetic algorithm
• ISSN: 0943-4062; 1613-9658; 1613-9658
• DOI: 10.1007/s00180-015-0618-2

Heuristic techniques of optimization can be useful in designing complex experiments, such as microarray experiments. They have advantages over the traditional methods of optimization, particularly in situations where the search space is discrete. In this paper, a search procedure based on a genetic algorithm is proposed to find optimal (efficient) designs for both one- and multi-factor experiments. A genetic algorithm is a heuristic optimization method that exploits the biological evolution to obtain a solution of the problem. As an example, optimal designs for 3 × 2 factorial microarray experiments are presented for different numbers of arrays and for various sets of research questions. Comparisons between different operators of the genetic algorithm are performed by simulation studies.