Prediction of Breast Cancer Candidate Drug Activity Based on Bi- LSTM
Konferenz: ISCTT 2021 - 6th International Conference on Information Science, Computer Technology and Transportation
26.11.2021 - 28.11.2021 in Xishuangbanna, China
Tagungsband: ISCTT 2021
Seiten: 4Sprache: EnglischTyp: PDFPersönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Zhang, Le (Shenyang University, College of Intelligent System Science and Engineering, Shenyang, Liaoning, China)
Li, Jinhua; Yang, Hong; Yin, Gang (Shenyang University, College of Information Engineering, Shenyang, Liaoning, China)
Breast cancer is currently one of the most common and fatal cancers in the world, so research on the prediction of the activity of its drug candidates is very important. In this paper, a QSAR model is constructed through the Bi-LSTM network to predict the activity of the drug. This process cleans the data obtained from the public database, discards or completes the missing values, and then obtains 20 molecular descriptors with higher correlation through correlation analysis. After obtaining 2024 pieces of data of dependent variable (IC50) and variable (descriptor), carry out serialization and normalization processing. By analyzing the mean square error (MSE) and the coefficient of determination (R2), the Bi-LSTM model has a higher accuracy than the RNN model. The Bi-LSTM model has good performance in predicting the activity of anti-breast cancer drug candidates.