Author: Kaifu Gao; Duc Duy Nguyen; Rui Wang; Guo-Wei Wei
Title: Machine intelligence design of 2019-nCoV drugs Document date: 2020_2_4
ID: 1qniriu0_16
Snippet: MathPose is a 3D pose predictor that converts SMILES strings into 3D poses with references of target molecules. For a given SMILES string, about 1000 3D structures are generated by several common docking software tools, i.e., Autodock Vina, 26 GOLD, 27 and GLIDE. 28 Additionally, a selected set of known complexes is re-docked by the aforementioned three docking software packages to generate at 100 decoy complexes per input ligand as a machine lea.....
Document: MathPose is a 3D pose predictor that converts SMILES strings into 3D poses with references of target molecules. For a given SMILES string, about 1000 3D structures are generated by several common docking software tools, i.e., Autodock Vina, 26 GOLD, 27 and GLIDE. 28 Additionally, a selected set of known complexes is re-docked by the aforementioned three docking software packages to generate at 100 decoy complexes per input ligand as a machine learning training set. In this training set, the calculated root mean squared deviations (RMSDs) between the decoy and native structures are used as machine learning labels. Then, we set up MathDL models and apply them to pick up the top-ranked pose for the given ligand. The MathPose-generated top poses are fed to the MathDL for druggable property evaluation. Our MathPose was the top performer in D3R Grand Challenge 4 in predicting the poses of 24 beta-secretase 1 (BACE) binders. 18
Search related documents:
Co phrase search for related documents- calculated root and docking software tool: 1
- calculated root and input ligand: 1
- calculated root and machine learning: 1
- calculated root and native decoy: 1
- calculated root and pose predictor: 1
- calculated root and select set: 1
- calculated root and SMILES string: 1
- calculated root and training set: 1, 2
Co phrase search for related documents, hyperlinks ordered by date