Genetic Algorithms Brandon Andrews Topics What are genetic

Genetic Algorithms Brandon Andrews

Topics �What are genetic algorithms? � 3 steps �Applications to Bioinformatics

What are genetic algorithms? � Invented and published in 1975 by John Holland � Cells have DNA which define properties � Reproduction crosses DNA from both parents merging properties from both • During this step random mutations can occur � A test of the fitness of the organism is performed • Scores the organism against others based on criteria for survival • Essentially evolution

3 Steps �Selection step • Based on the calculate fitness �Reproduction step • Mutations • Strategies for crossing �Termination step • When the goal is met

Steps Expanded � 1) Generate random properties (chromosomes) for N entities � 2) Calculate their fitness and discard ones that fall below the threshold • Can be determined through a simulation � 3) Randomly cross over pairs that survive the selection step • Also randomly choose properties and mutate them. This could be as simple as jittering them � 4) Go to step 2 until a goal is reached • Return the best set of properties

Fitness Function �Could be anything �The goal is to minimize or maximize the fitness function normally after each step

Crossover Probability � How often crossovers happens • 0% represents if no crossover and both parents are simply moved to the next step • 100% represents that all of the parents are crossed and only their children are move to the next step � The idea is that hopefully the good properties of both parents are merged or the good parent is preserved completely if it has no flaws that can be fixed via a crossing pair

Mutation Probability � The probability that part of the chromosome is changed after a crossing • 0% if none of it is changed �Not useful since variety is needed to approach the best solution or you’re stuck with the first generated properties • 100% if all of it is changed �Not useful since it negates the point of crossing at all, causes a random search essentially � The concept is to stop the algorithm from halting at a local maximum. The mutations have a chance to generate small better changes

Termination �When the expected error is low • Sometimes it’s hard to calculate an error since the solution isn’t known �Or when the results stop minimizing for a few iterations or stops increasing depending on the problem

Approximate Solutions �Might be obvious, but genetic algorithms are by design approximate solutions since they attempt to optimize to a solution • Perfection is only as good as the fitness function and the number of iterations, crossing and mutation probabilities

Applications � Multiple Sequence Alignment • Initial generation – random generation of an alignment based on the alignments of the given sequences �No authors agree on the initial size of the population • Selection via a tournament style pairing crossing the possible alignments • The fitness function �“Sum of pair” Objective Function (everyone uses a different one) • The survival rate is different for each alignment �Sum all alignment scores together and take a percentage for each alignment �Basically better alignments have a higher percentage to survive

• Reproduction �Crossing uses a “one-point crossover” �Takes the first half of the first alignment and cross if with the second half of the second parent �ABCD and EFGH -> ABGH �Or “point-to-point crossover” �Random index is chosen �ABCD and EFGH -> ABCH • Mutation �Remove or insert a gap into the alignment

References � Obitko M. (1998). Genetic Algorithms. Retrieved from http: //www. obitko. com/tutorials/genetic‑algo rithms/ � Radenbaugh A. (2008). Applications of genetic algorithms in bioinformatics. Retrieved from http: //scholarworks. sjsu. edu/cgi/viewcontent. cgi? article=4491&context=etd_theses