Find in Library

Simulation studies in population genetics play a crucial role in better understanding of different evolution scenarios and effects of different genetic models on genetic diversity. forward-in-time method starts with an initial population and follows the entire evolution under various genetic models within multiple generations. Artificial neural networks represent a formidable method for genetic simulation and prediction. In this study, we wanted to compare and corroborate results obtained with forward-in-time simulation with results attained from a specially designed strategy based on artificial neural networking. As input data, alleles of 13 microsatellite loci from 187 specimens representing autochthonous Adriatic haplotype of Salmo trutta L. from the Neretva River were used. The main goal of this study was to compare precision and reliability of these two methods. Our results are in concordance with other reports from literature which indicate that both of these approaches can be used as a reliable simulation tools. However, it is believed that artificial neural networks can represent more powerful simulation tools