Protein conformation change is a fundamental problem to both computational biology and drug design.The function of a protein is highly correlated with its structure and conformational change,and the binding site of target protein could undergo significant transformation along with the protein conformation change.However,it is a great challenge to simulate large scale protein conformational change nowadays due to enormous computer resource is required for the simulation.In order to improve the efficiency of protein conformation change simulation,we developed a new approach,namely,NUMD,which could efficiently simulate the conformational change of protein and predict the free energy profile along the conformational transition pathway.The application of the new approach on some protein system revealed that the simulation results by NUMD is similar to that by conventional molecular dynamics(MD)simulation.[1,2] However,the computational resources required for NUMD simulation is only about 2%of that for conventional MD.Meanwhile,we also optimized other enhanced sampling method(e.g.,REMD)to increase the computational efficiency for same purpose.