abstract |
The invention discloses a method for predicting the cytotoxicity of disinfection by-products. With the help of the molecular structure and physicochemical properties of compounds, a method based on a machine learning algorithm is used to predict the cytotoxicity of DBPs. The method process includes: collecting cytotoxicity values of DBPs and establishing a database; converting all DBPs into SMILES; calculating molecular fingerprints of all DBPs samples, standardizing and normalizing sample data; constructing toxicity prediction based on multiple machine learning algorithms model, and select the optimal model; after inputting the SMILES expression of the DBPs to be tested, the predicted cytotoxicity value of the DBPs to be tested is directly output. |