Construction method of molecular prediction model and computing equipment

Abstract

The invention discloses a construction method of a molecular prediction model, which is suitable for being executed in computing equipment, a feature type set, a feature engineering set and a prediction model set are stored in the computing equipment, the method comprises the following steps: collecting a plurality of pieces of molecular data of a specific molecular type, and the molecular data comprises performance data and at least one kind of feature data; one or more feature types, feature engineering and prediction models are selected from the feature type set, the feature engineering setand the prediction model set for traversal combination, and multiple combination modes are obtained; training each combination mode in combination with the plurality of pieces of molecular data and the feature type, the feature engineering and the prediction model in each combination mode to obtain a trained model and a model evaluation index; and selecting an optimal combination mode of the feature type, the feature engineering and the prediction model according to the model evaluation indexes to predict the properties of the same type of molecules. A computing device adapted to perform themethod is also disclosed.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to view more

AI Technical Summary

Benefits of technology

The technology described by this patented allows researchers to extract relevant characteristics from large amounts of material with different physical or chemical processes involved during manufacturing. These results help identify potential sources of new drugs more efficiently than current methods like X-ray crystallography alone. This helps improve pharmaceutical development while reducing costs associated therewith.

Problems solved by technology

The technical problem addressed by this patented method is how fast and accurate identification of highly specific chemical structures that are involved in diseases such as cancer or Alzheimer's disease through computational methods like machine learning techniques (ML).

Images