Succinylation Site Prediction Based on Protein Sequences Using the IFS-LightGBM (BO) Model

Area of Science:

• Biochemistry
• Computational Biology
• Proteomics

Background:

• Succinylation is a crucial post-translational modification affecting protein structure and cellular functions.
• Dysregulated succinylation on lysine residues is implicated in various diseases.
• Accurate identification of succinylation sites is vital for mechanistic studies.

Purpose of the Study:

• To develop a highly accurate computational model for predicting protein succinylation sites.
• To enhance understanding of the role of succinylation in biological processes and diseases.

Main Methods:

• Developed the IFS-LightGBM (BO) model integrating Incremental Feature Selection (IFS), LightGBM, and Bayesian Optimization.
• Employed features including pseudo amino acid composition (PseAAC), PSSM, disorder status, and CKSAAP.
• Utilized LightGBM and IFS for optimal feature subset selection and Bayesian optimization for model parameter tuning.

Main Results:

• The IFS-LightGBM (BO) model demonstrated superior performance in predicting succinylation sites.
• The model achieved high accuracy, recall, precision, MCC, and F-measure.
• Feature selection and parameter optimization strategies improved predictive power and computational efficiency.

Conclusions:

• The developed IFS-LightGBM (BO) model offers a robust and efficient tool for succinylation site prediction.
• This method facilitates deeper insights into the functional roles of succinylation in health and disease.
• The study highlights the potential of integrated computational approaches in post-translational modification research.