Predicting protein structural classes for low-similarity sequences by evaluating different features
作者：
Highlights：
• A novel method is developed to predict protein structural classes.
• The protein samples were formulated by integrating various knowledge.
• An overall accuracy of 96.7% was obtained on a strict benchmark dataset.
摘要
•A novel method is developed to predict protein structural classes.•The protein samples were formulated by integrating various knowledge.•An overall accuracy of 96.7% was obtained on a strict benchmark dataset.
论文关键词：Protein structural class,Feature fusion,Low-similarity sequence,Machine learning method
论文评审过程：Received 2 March 2018, Revised 2 September 2018, Accepted 4 October 2018, Available online 15 October 2018, Version of Record 21 November 2018.
论文官网地址：https://doi.org/10.1016/j.knosys.2018.10.007
