Method for predicting protein and ligand molecule binding free energy based on convolutional neural network
A technology of convolutional neural network and ligand molecules, which is applied in the field of predicting the binding free energy of proteins and ligand molecules based on convolutional neural networks, can solve problems such as limited application, difficult generalization of homologue ligands, and difficulties in large-scale application , to achieve the effects of small error, good generalization ability, and precise protein-ligand binding free energy calculation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0048]Example 1
[0049]Binding free energy prediction for a kinase target:
[0050]First collect the small molecule inhibitors of the target, perform 3D conformation preparation and molecular docking calculation, then call the molecular descriptor module to process these molecules and proteins, and input them into the convolutional neural network model. Such asimage 3 As shown, all molecular data is divided into 5 parts, one of which is selected as the validation set each time, and the other four parts are used as the training set. In this way, the model is trained for 5 rounds, that is, 5-fold cross-validation, and finally all the validation scores are taken The average value is used as the final verification score.Figure 4 ,Figure 5 Represents the performance on the validation set during the model training, fromFigure 4 It can be seen that the error in the initial stage of the model is relatively large, but after about 50 rounds of training, the error is reduced to about 2.3, and it re...
Example Embodiment
[0053]Example 2
[0054]Optimize the structure of the lead compound for a target:
[0055]Some seed compounds with good initial activity are obtained through virtual screening, and the convolutional neural network model is used to optimize the structure of this batch of seed compounds to obtain lead compounds.
[0056]Firstly, the molecular docking of the seed compound and the target is carried out, and then they are encoded and molecular descriptors are calculated. Input it into the convolutional neural network model to predict the free energy of binding between the compound and the target. According to the binding mode between the seed compound and the target, as well as the binding free energy value and spatial hierarchical structure information predicted by the model, the structure of the seed compound is modified and optimized, and groups matching the spatial hierarchical structure of the target are added to the seed compound , To form a better structural complementation with the target...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap