Molecular Structure Evolutionary Search for Cost-Efficient Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional molecular structure optimization in drug development is costly and inefficient due to reliance on biochemical experiments and expert knowledge.
Innovation Solution
An evolutionary-algorithm-based method that generates, filters, evaluates, and iteratively learns molecular structures to achieve optimization objectives, reducing costs and improving efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional biochemical experiments and expert knowledge are used for molecular structure optimization, then reliable molecular structures can be obtained, but the process becomes costly and inefficient
Solution Approach 1:
The patent replaces the mechanical system of biochemical experiments with an evolutionary algorithm-based computational system. The algorithm iteratively generates, evaluates, and optimizes molecular structures through automated computational processes, eliminating the need for manual experimentation while maintaining reliability through systematic search and evaluation mechanisms.
Solution Approach 2:
The evolutionary algorithm performs self-directed optimization by automatically generating candidate molecular structures, evaluating their properties, and selecting improved structures for the next generation. This self-service mechanism eliminates dependency on expert knowledge for each optimization step while systematically improving molecular structures according to defined objectives.
2Measurement precision
If conventional biochemical experiments are used for molecular structure optimization, then accurate results can be achieved, but implementation costs increase significantly
Solution Approach 1:
The patent creates computational copies of molecular structures and their properties, allowing virtual evaluation and optimization without physical experimentation. The evolutionary algorithm generates and evaluates numerous candidate structures in silico, replacing expensive wet-lab experiments with computationally inexpensive simulations that maintain accuracy through sophisticated modeling.
Solution Approach 2:
The patent substitutes expensive biochemical experimentation with computational algorithms. The evolutionary process uses mathematical models and computational chemistry methods to predict molecular properties, replacing costly laboratory procedures with automated computational assessments that maintain precision while dramatically reducing implementation costs.
3Reliability
If expert knowledge is relied upon for molecular structure optimization, then quality molecular structures can be obtained, but the process requires extensive time and resources
Solution Approach 1:
The evolutionary algorithm implements continuous optimization through iterative generations, where each cycle generates new candidate structures, evaluates them, and produces improved offspring. This continuous automated process eliminates interruptions associated with manual expert analysis and biochemical experimentation, maintaining quality through systematic evaluation while dramatically reducing total optimization time through parallel processing and automated selection.
Solution Approach 2:
The algorithm performs self-directed molecular structure optimization without requiring expert intervention at each step. The automated evolutionary process independently generates candidates, evaluates properties, and selects improvements, maintaining reliability through built-in evaluation criteria while eliminating time losses associated with expert availability and manual analysis cycles.
Data Source
AI summary
A molecular structure acquisition method, an electronic device and a storage medium, which relate to the field of artificial intelligence such as deep learning, are disclosed. The method may include: performing, for an initial seed, the following first processing: generating M molecular structures according to the seed, M being a positive integer greater than one; taking the M molecular structures as candidate molecular structures, and selecting some molecular structures from the candidate molecular structures as progeny molecular structures; and performing evolutionary learning on the progeny molecular structures, taking the progeny molecular structures after evolutionary learning as the seed, and repeating the first processing until convergence reaches an optimization objective, and when the convergence reaches the optimization objective, a newly selected molecular structure is taken as a desired molecular structure.