Molecular Generation Method Based on Data Mining

Abstract

The invention discloses a method for generating molecules based on data mining, comprising: acquiring a data set, selecting N from the data set p number of molecules; encode all the molecules to obtain the corresponding molecular matrix, put N p A molecular matrix is ​​used as the initial population, and chemical restriction mask operation is performed on each molecular matrix in combination with chemical prior knowledge to indicate whether each atom in each molecule is saturated; based on the molecular matrix and the result of chemical restriction mask operation, within the population Perform crossover and mutation operations to generate multiple new molecular matrices, and select N from the molecular matrix in the population and the new molecular matrix through the fitness function p The molecular matrix of each is used as the next-generation population; the above operations are performed on the molecular matrix in the next-generation population until the population converges or the number of iterations exceeds a preset upper limit. The above can solve the disadvantages of low efficiency and expensive calculation of deep learning models, greatly improve the efficiency and effect, and produce legal molecules.

Description

technical field [0001] The invention relates to the field of automatic molecule generation, in particular to a method for generating molecules based on data mining. Background technique [0002] Molecular generation is a very important task for the field of chemical pharmacy. In the past chemical pharmaceutical industry, it took billions of dollars and more than ten years from the discovery of a drug to the actual production of drugs. In recent years, the technology of using computers to accelerate molecular pharmacy has received more and more attention. Many articles in the top journals of computer and chemistry have begun to focus on this aspect of research, which shows its importance. [0003] The traditional pharmaceutical paradigm starts with an existing chemical material concept, then synthesizes candidate materials based on this concept, then tests the material in the entire system, and finally evaluates the material's properties and properties. However, this paradi...

AI Technical Summary

Problems solved by technology

However, deep learning models usually require a large amount of labeled data, a lot of computing resources, and a very complex training model, which brings great difficulty to practical applications.

At the same time, some people regard this problem as a black-box optimization problem and use the perspective of evolutionary calculation to solve this problem, but the effect and efficiency need to be improved. For example, the process of generating molecules will produce illegal molecules

Images