@techreport{oai:ipsj.ixsq.nii.ac.jp:00216915, author = {Dian, Ma and Nobuaki, Yasuo and Masakazu, Sekijima and Dian, Ma and Nobuaki, Yasuo and Masakazu, Sekijima}, issue = {5}, month = {Mar}, note = {This study develops a new deep learning-based extendable multiple-objective molecular generator (MOMolGen). This generator integrates a recurrent neural network (RNN) to generate molecules and Pareto Multi-Objective Monte Carlo Tree Search (Pareto MOMCTS) to decide search direction. This generator is validated by generating compounds for specific target proteins with evaluation on the drug-like properties and docking score., This study develops a new deep learning-based extendable multiple-objective molecular generator (MOMolGen). This generator integrates a recurrent neural network (RNN) to generate molecules and Pareto Multi-Objective Monte Carlo Tree Search (Pareto MOMCTS) to decide search direction. This generator is validated by generating compounds for specific target proteins with evaluation on the drug-like properties and docking score.}, title = {In silico drug design by Molecular Generative Model and Docking}, year = {2022} }