Related Experiment Video
Updated: Sep 19, 2025
A Spin-Tip Enrichment Strategy for Simultaneous Analysis of N-Glycopeptides and Phosphopeptides from Human Pancreatic Tissues
Published on: May 4, 2022
An effective statistical moment-based feature extraction technique to identify the phosphoglycerylation sites from
Md Sohrawordi1, Md Ali Hossain2, Md Al Mehedi Hasan2
1Department of Computer Science and Engineering, Rajshahi University of Engineering and Technology, Rajshahi, Bangladesh; Department of Computer Science and Engineering, Hajee Mohammad Danesh Science and Technology University, Dinajpur, Bangladesh.
This study introduces a new method, statistical moment of physicochemical properties (SMPP), to identify lysine phosphoglycerylation sites. SMPP significantly improves prediction accuracy, aiding research into related diseases.
Area of Science:
- Biochemistry
- Computational Biology
- Genomics
Background:
- Post-translational modifications (PTMs) are crucial in cell biology.
- Lysine phosphoglycerylation, a reversible PTM, impacts glycolytic enzyme activity and is implicated in heart failure, arthritis, and nervous system disorders.
- Current machine learning methods for identifying phosphoglycerylation sites can be enhanced with novel feature extraction techniques.
Purpose of the Study:
- To develop an efficient feature extraction technique for identifying phosphoglycerylation sites.
- To create a computational model for predicting phosphoglycerylation sites using the proposed technique.
- To evaluate the predictive performance of the new method compared to existing approaches.
Main Methods:
- Proposed a novel feature extraction technique: statistical moment of physicochemical properties (SMPP).
- Utilized the statistical moment procedure and physicochemical characteristics of amino acids to generate numerical protein features.
- Developed a computational model employing a Support Vector Machine (SVM) classifier.
Main Results:
- Achieved high predictive accuracy: 98.26% on 10-fold cross-validation and 99.40% on an independent test set.
- The SMPP feature extraction method outperformed existing techniques in identifying phosphoglycerylation sites.
- Demonstrated the efficacy of SMPP in enhancing the prediction of lysine phosphoglycerylation.
Conclusions:
- The SMPP method offers a powerful and efficient approach for identifying phosphoglycerylation sites.
- This advancement has significant implications for understanding and diagnosing diseases associated with altered phosphoglycerylation.
- The developed web server, dataset, and source code are publicly available for further research and application.
