This project is a reimplementation of the models introduced in the following papers:
-
"Multiobjective de novo drug design with recurrent neural networks and nondominated sorting." (paper). Official GitHub repository.
-
"REINVENT 2.0: An AI Tool for De Novo Drug Design." (paper). Official GitHub repository.
-
"Hierarchical generation of molecular graphs using structural motifs." (paper). Official GitHub repository.
-
"Mol-CycleGAN: a generative model for molecular optimization." (paper). Official GitHub repository.
-
"Multi-objective de novo drug design with conditional graph generative model." (paper). Official GitHub repository.
-
"Graph convolutional policy network for goal-directed molecular graph generation." (paper). Official GitHub repository.
Specifically, the code is a slightly updated version of that published by the authors in their projects.
Important: Even if the code presented in this repository is almost entirely based on the code published by the authors in their works the results might differ for some reason. Therefore, for any benchmark test to be performed on the models of the paper, please refer to the original code.
Yasonik, Jacob. "Multiobjective de novo drug design with recurrent neural networks and nondominated sorting." paper Journal of Cheminformatics 12.1 (2020): 1-9.
Blaschke, Thomas, et al. "REINVENT 2.0: An AI Tool for De Novo Drug Design." paper Journal of Chemical Information and Modeling (2020).
Jin, Wengong, Regina Barzilay, and Tommi Jaakkola. "Hierarchical generation of molecular graphs using structural motifs." arxiv International Conference on Machine Learning. PMLR, 2020.
Maziarka, Łukasz, et al. "Mol-CycleGAN: a generative model for molecular optimization." paper Journal of Cheminformatics 12.1 (2020): 1-18.
Li, Yibo, Liangren Zhang, and Zhenming Liu. "Multi-objective de novo drug design with conditional graph generative model." paper Journal of cheminformatics 10.1 (2018): 1-24.
You, Jiaxuan, et al. "Graph convolutional policy network for goal-directed molecular graph generation." paper arXiv preprint arXiv:1806.02473 (2018).