AI Virtual Screening System for Crystal Complex Drug Discovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional drug design strategies face challenges in obtaining new drug scaffolds due to the limitations of existing compound libraries and patent protections, making it difficult for latecomers to develop novel compounds through simple substitutions, and relying heavily on explored compound libraries.
Innovation Solution
A virtual drug screening system for crystal complexes, comprising a visualization subsystem, evaluation tool box subsystem, AI model management subsystem, large-scale sampling subsystem, virtual screening subsystem, and data log storage subsystem, which uses AI to generate and filter compounds based on user-defined requirements, enhancing affinity to proteins and overcoming traditional library limitations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional compound libraries are used for virtual screening, then the screening process is straightforward, but the compounds lack novelty and have been extensively explored
Solution Approach 1:
The system performs preliminary actions by pre-training the AI model on existing crystal complex data and pre-defining the virtual screening workflow before actual drug discovery tasks. This allows the system to generate novel compounds on-demand rather than relying on pre-existing libraries, resolving the contradiction between process reliability and compound novelty
Solution Approach 2:
The system changes the fundamental parameter of compound generation from selecting from fixed libraries to AI-driven de novo generation. By adjusting the AI model parameters and training data, the system can generate compounds with desired properties while ensuring novelty, thus resolving the contradiction between reliable screening and novel compound discovery
2Ease of manufacture
If traditional drug design strategies with substituent replacement are used, then the design process is simple and follows established protocols, but new drugs with the same scaffold are difficult to obtain due to patent protections
Solution Approach 1:
The system inverts the traditional approach by instead of starting with a scaffold and replacing substituents, it generates entirely new scaffold structures using AI models trained on crystal complex data. This inversion allows obtaining novel scaffolds that are not constrained by existing patent protections while maintaining systematic design approaches
Solution Approach 2:
The system uses copying by training the AI model on existing crystal complex structures and binding modes, then generating new compounds that copy the essential binding features while having novel scaffolds. This allows maintaining the simplicity of structure-based design while achieving scaffold innovation through AI-generated structures
3Adaptability or versatility
If AI-generated compounds are used to expand compound library, then the space for exploration is broader and more compounds can be generated, but the system complexity increases
Solution Approach 1:
The system introduces an intermediary AI model that bridges the gap between simple input parameters and complex compound generation. The AI model acts as a mediator that handles the complexity of generating diverse, novel compounds from simplified user inputs, thus increasing library diversity while managing system complexity through the intermediary layer
Solution Approach 2:
The system implements self-service by having the AI model automatically generate and optimize compound structures based on input criteria without requiring manual intervention for each compound design. The model self-adjusts parameters and generates diverse compounds autonomously, increasing library diversity while keeping the user interface simple and the overall system manageable
Data Source
AI summary
The present invention provides a drug virtual screening system for crystal complexes, and method of using the same, comprising a visualization subsystem, an evaluation tool box subsystem, an AI model management subsystem, a large-scale sampling subsystem, a virtual screening subsystem, and a data log storage subsystem. Starting with the known crystal complexes, a batch of candidate compounds that meet the requirements are recommended after going through the visualization subsystem, evaluation tool box subsystem, AI model management subsystem, large-scale sampling subsystem, and virtual screening system in turn. Based on this system, the generation of the compound library is organically combined with the subsequent virtual screening. Users only need to describe the action mode of the drug on the protein and the requirements for the drug to generate a batch of compounds that meet the expectations. The automated system reduces user intervention and improves the efficiency of research and development.