Chemical estrogen receptor activation activity prediction model and a screening method

Abstract

The invention discloses a chemical estrogen receptor activation activity prediction model and a screening method. The establishment method of the chemical estrogen receptor activation activity prediction model comprises the steps of obtaining chemical data of known estrogen receptor activation activity, and the chemical data comprises SMILES codes of chemicals; converting the SMILES code to obtain an M * N digital matrix; dividing the obtained data of the known chemicals into a training set and a verification set, and constructing a convolutional neural network model taking an SMILES digital matrix as input; and training the convolutional neural network model by using the training set, and determining an optimal hyper-parameter of the convolutional neural network model by using the verification set to obtain an optimal convolutional neural network model. The method is suitable for screening the estrogen activation activity of large-scale chemicals, is simple, rapid and efficient, and has a wide application prospect in the fields of chemical risk evaluation, environmental safety evaluation and the like.

Description

technical field [0001] The invention relates to the technical field of environmental health risk assessment of chemicals, and more specifically relates to a prediction model and screening method for estrogen receptor activation activity of chemicals. Background technique [0002] With the development of industry, human beings are more exposed to environmental chemicals. Environmental endocrine disruptors interfere with the normal function of human hormones and cause reversible or irreversible biological effects on the human body, which has attracted great attention from the government and researchers. As early as 1999, the US Environmental Protection Agency (EPA) implemented an endocrine disruptor screening program. In 2000, the National Natural Science Foundation of my country also invited bids for the key project "The Mechanism of Environmental Hormones Affecting Human Health", and started large-scale research on environmental hormones in my country. Environmental estroge...

AI Technical Summary

Problems solved by technology

At present, CNN has not been applied in the field of environmental estrogen activity evaluation.

[0004] In summary, although the quantitative structure-activity relationship mathematical prediction model based on traditional machine learning algorithms has greatly improved the process of chemical evaluation and rapid screening of properties, due to the limitation of available descriptors, it cannot be used in more complex systems. It is difficult to achieve sufficient prediction results; and the calculation and collection of descriptors requires a certain amount of time, computing resources, and a certain discipline foundation, which also limits the application of prediction models to a certain extent

Images