Model training method, model prediction method, molecule screening method and device thereof

Abstract

The invention provides a model training method, a model prediction method, a molecule screening method and devices thereof. The model training method comprises the steps that a plurality of initial molecular data sets are obtained, and the initial molecular data sets are constructed according to different positive and negative sample proportions; an initial model is constructed, the initial model is trained by using the plurality of initial molecular data sets, a plurality of training models and a prediction result of each training model are obtained, and each initial molecular data set corresponds to one training model; obtaining an evaluation index of the training model based on the prediction result; and selecting the training model meeting a preset requirement as a final prediction model according to the evaluation indexes of the training models. Through the mode, the terminal equipment selects the model capable of meeting the specific requirements as the final output model through the performance of different data sets on the model, and the pertinence and efficiency of model training can be improved.

Description

technical field [0001] The present application relates to the technical field of computational chemistry, in particular to a model training method, a model prediction method, a molecular screening method and a device thereof. Background technique [0002] Drug safety is an important issue in the drug development process. The main reasons for clinical trial failure in the 21st century are lack of efficacy and safety (approximately 30%). Cardiotoxicity, hepatotoxicity, genotoxicity, and phototoxicity are frequently observed toxicities. Cardiotoxicity and hepatotoxicity were reported in 30% to 40% of clinical trials and post-approval studies, respectively. hERG is a key factor in drug-induced QT prolongation and cardiac arrhythmias (also known as torsades de pointes). Due to its special molecular structure, hERG can be inhibited by structurally diverse compounds. Terfenadine, cisapride and sertindole have been withdrawn from the market due to their inhibition of hERG. There...

AI Technical Summary

Problems solved by technology

At the same time, the selection of training sets and descriptors of different models is different, so that the accuracy of the final model is also different.

Due to the differences in data formats and ontologies of data sets from different sources, it is very difficult to integrate various databases. In most studies, the data sets used to build hERG prediction models come from a single database, and the size of the data set is less than 3000. Using a single database to build predictive models may result in limited predictive performance and applicability

Images