Optimizing HP Model Using Reinforcement Learning
Protein structure prediction has always been an important issue in bioinformatics field. This paper proposes an HP model optimization method based on reinforcement learning, which is a new attempt in the area of protein structure prediction. It does not require external supervision as the agent can find the optimal solution from the reward function in the training process. And the method also decreases computational complexity through making the time complexity of the algorithm has a linear relationship with the length of protein sequence.
KeywordsReinforcement learning HP model Structure prediction
This paper is supported by the National Natural Science Foundation of China (61772357, 61502329, 61672371), Jiangsu 333 talent project and top six talent peak project (DZXX-010), Suzhou Foresight Research Project (SYG201704, SNG201610) and the Postgraduate Research & Practice Innovation Program of Jiangsu Province (SJCX17_0680).
- 3.Mann, M., Backofen, R.: Exact methods for lattice protein models. Bio-Algorithms Med-Syst. 10(4), 213–225 (2014)Google Scholar
- 5.Qin, Y.F., Zheng, X.Q., Wang, J., et al.: Prediction of protein structural class based on linear predictive coding of PSI-BLAST profiles. Open Life Sci. 1, 11–15 (2015)Google Scholar