Machine learning-based nucleoside derivative gelling ability prediction model

• Main Authors: ZHAO XING, WEN YINGHUI, XU HAO, XIE LIANG, WANG KAICHAO, LI WEIQI
• Format: Patent
• Language: Chinese; English
• Published: 12.09.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTEDFOR SPECIFIC APPLICATION FIELDS; PHYSICS

The invention provides a nucleoside derivative gelling ability prediction model based on machine learning, and belongs to the field of computer prediction systems. Based on feature selection, hyper-parameter optimization and algorithm comparison, an optimal machine model for predicting the nucleoside derivative hydrogel forming ability is successfully established. The model can effectively predict whether the nucleoside derivative has gelling ability. 12 nucleoside gels with high possibility are selected from the model, the hydrogel forming ability of the nucleoside gels is verified through experiments, 10 nucleoside derivatives can form the hydrogel, the success rate of forming the hydrogel is 83.33%, and it is indicated that the nucleoside derivative gelling ability prediction model is high in prediction accuracy. The machine model provides a tool for predicting nucleoside derivatives with hydrogel forming ability. 本发明提供了一种基于机器学习的核苷衍生物成胶能力预测模型，属于计算机预测系统领域。本发明基于特征选择、超参数优化和算法比较，成功建立了预测核苷衍生物水凝胶形成能力的最优的机器模型。该模型