Fig. 5From: Combining genetic algorithm with machine learning strategies for designing potent antimicrobial peptidesFitness distributions of selected generations. Fitness score distribution across selected generations as violin plots. The center of the violin plot is a box plot and the sides of the violin plot are the probability density kernels. The gray line is the score of zero, a perfect score representing all design targets being met. Generations are sequentially produced until a limit of non-improving generations is reached. A generation is non-improving if none of the members of the generation have an improved fitness score from the previous generationsBack to article page