Extensive comparison of protein sequence-based bioinformatics applications for predicting lysine succinylation sites: a comparative review

Lysine succinylation represents a critical post-translational modification (PTM) that alters the charge of lysine residues from positive to negative, thereby influencing protein structure and physiological functions in both eukaryotic and prokaryotic cells. Given the biological significance of succinylation, identifying these modification sites is essential for understanding cellular regulation. While experimental methods are precise, they remain resource-intensive and time-consuming. Consequently, computational approaches leveraging machine learning and deep learning have emerged to predict succinylation sites directly from primary protein sequence data. This project provides a comprehensive comparative analysis of existing sequence-based bioinformatics applications designed for lysine succinylation prediction. It systematically evaluates the architectural frameworks, feature extraction techniques, and classification algorithms employed by

various models. Furthermore, the study addresses critical computational challenges, including dataset imbalance, feature selection redundancy, and model generalization across diverse biological species. By benchmarking performance metrics such as sensitivity, specificity, and Matthews correlation coefficient, this research direction offers structured guidance for selecting optimal predictive models and outlines future development pathways for robust, high-throughput post-translational modification prediction tools.

Message: Hi FE, I need help with “Extensive comparison of protein sequence-based bioinformatics applications for predicting lysine succinylation sites: a comparative review” in “Biotechnology & Biomedical Engineering”