Standard Multi-Layer Perceptron on Positive - Unlabeled Glycosylation Site Dataset
Procedia Computer Science(2023)
摘要
Abstract—The implementation of computational approaches for protein glycosylation site prediction is becoming popular since the experimental-validated glycosylation data became more abundant. Some of the data were found to be wrong after the experiment was again carried out with more sophisticated technology. To solve this issue, the latest state-of-the-art model trained the model based on a positive-unlabelled algorithm. The aim of this research is to explore the possibility of an approach applying a simple neural network algorithm and still achieve high classification performance. The model proposed in this research gave competitive results with fewer preprocessing steps. Increasing the accuracy of glycosylation site prediction can complement laboratory-based methods and is very useful for understanding the role of glycosylation.
更多查看译文
关键词
Glycosylation,Site Prediction,PTM,Standard Multi-Layer Perceptron
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要