Abstract: Abstract: To improve efficiency of predicting protein structures, a series of improved search strategies is investigated based on the standard genetic algorithms (GA) for the hydrophobic-polar (HP) protein folding problem. The new strategies include the followings: to control the number of identical individuals for maintaining the diversity of populations; to combine with competing between parent and offspring at a single point crossover to enhance the survival level; a local search algorithm based on systematic mutation coupled with the best individuals, and so on. The experiments show that the improved genetic algorithms lead to significant improvements in the efficiency and success of the GA, and GA has a significant potential for predicting protein configuration.