A genetic algorithm for assigning individuals to populations using multi-locus genotyping

This paper reports a genetic algorithm (GA) for individual assignment using multi-locus microsatellite genotyping. Its performance has been compared with existing frequency, Bayesian and distance-based methods using simulated as well as actual data. Simulated data has been generated with SIMCOAL program. Actual data has been generated from genotypes of four cattle breeds from India. The GA showed lower accuracy while assigning individuals from simulated data. Its performance was comparable to that of existing methods using actual data.