Summary: The research of protein thermostability has been vigorously studied in the field of biophysical and biological technology. What is more, protein thermostability in the level of amino acid sequence is still a challenge in the research of the protein pattern recognition. In this paper, we try to use new integrated feedforward artificial neural network which was optimized by particle swarm optimization (PSO-NN) to recognize the mesophilic and thermophilic proteins. Here, we adopted Genetic Algorithm based Selected Ensemble (GASEN) as our integration methods. A better accuracy was got by GASEN. So, the integrated methods were proved to be effectual.