No Detail Found!

Accurate prediction of Snare Protein Sequence using Machine Learning

Author(s)
Dani Bux Talpur Center for Computing Research, Department of Computer Science and Software Engineering, Jinnah University for Women,Karachi 74600,Pakistan
Salahuddin Shaikh Center for Computing Research, Department of Computer Science and Software Engineering, Jinnah University for Women,Karachi 74600,Pakistan
Ashfaque Khowaja Department of Computer Science and Engineering Central South University,Changsha,China
Saifullah Adnan Center for Computing Research, Department of Computer Science and Software Engineering, Jinnah University for Women,Karachi 74600,Pakistan
Ali Ghulam Computerization and Network Section, Sindh Agriculture University,Tandojam,Pakistan
Abstract
In recent years, researchers have become increasingly interested in machine learning. Machine learning has been widely used to solve multiple challenges that span a wide range of industries with industry-leading performance. This research study aims to use machine learning to determine the key biochemical roles of proteins called SNARE proteins. Numerous studies have investigated human diseases as shown a loss of functional SNARE proteins (e.g., neuro degeneration, mental illness, cancer, etc.). Therefore, developing an accurate model to identify the role of these diseases is important to understand them, and developing therapeutic targets to treat them is an important task. We proposed a computational method based on SNARE-XGBoost, which contained on 1D-dimensional convolutional neural network and dipeptide deviation form expected mean (DDE) vector score matrix is used. We examined 10-fold cross-validation data set, then we obtained sensitivity is 76.6%, specificity is 93.5%, accuracy is 89.7%, and MCC is 0.70%. In addition, we used experimental test to check the degree of overfitting of the model by evaluating independent data sets, and the results showed that we had eliminated overfitting. Our proposed approach has yielded considerable overall excellent performance across all criteria compared to previous cutting-edge technologies. In the proposed study, we have created an excellent model for the identification of snare proteins and laid the foundation for broader research, which will facilitate the development of in-depth learning techniques, especially protein function prediction.
Publication Details
Page(s) 1414-1422
DOI DOI not available
Published Journal: Bioscience Research, Volume: 19, Issue: 3, Year: 2022

Request the full text from PASTIC

Keywords
machine learning, XGBoost, protein peptide prediction, SNARE Proteins, Computational Method
0

Citations

0

Downloads

49

Views