Model Classification Using an Extreme Gradient Boosting Algorithm to Predict the Post-Translational Modification (PTM) Glycosylation in Sequence O

Abstract

Glycosylation is one kind of post-translational modification. The most challenging modification to proteins is glycosylation. Glycosylation has also been connected to a number of diseases, such as diabetes, cancer, and the flu. It is important to anticipate glycosylation as a result. The manual lab method for calculating glycosylation requires costly upkeep and equipment. Computer technology is required to improve it by making glycosylation predictions faster. The purpose of this work was to increase sequence O post-translational modification glycosylation prediction accuracy. This study employs a more effective gradient augmentation approach called extreme gradient boosting. Experiments with the following feature extraction types—AAindex, hydrophobicity, solubility, composition, CTD, and PseAAC—help to improve this accuracy. MRMR is the method used for feature selection. The study's findings show that sequence O's post-translational modification glycosylation can be predicted with 100% accuracy.