A computational method to predict carbonylation sites in yeast proteins

Area of Science:

• Biochemistry
• Computational Biology
• Proteomics

Background:

• Protein carbonylation is an oxidative stress-induced post-translational modification (PTM) and a biomarker of oxidative stress.
• Only specific amino acid residues (lysine, arginine, threonine, proline) are susceptible to carbonylation.
• Identifying carbonylation sites experimentally is costly, time-consuming, and limited in scope.

Purpose of the Study:

• To develop a computational method for predicting protein carbonylation sites in yeast.
• To address the lack of bioinformational tools for predicting yeast protein carbonylation.
• To provide an accessible tool for researchers studying oxidative stress.

Main Methods:

• Development of a computational prediction model for yeast protein carbonylation sites.
• Utilized amino acid residue susceptibility (lysine, arginine, threonine, proline) in the model.
• Validated the method using 10-fold cross-validation and analyzed site-flanking residue features.

Main Results:

• The computational method achieved high prediction accuracies: 86.32% for lysine, 85.89% for arginine, 84.80% for threonine, and 86.80% for proline.
• 10-fold cross-validation confirmed the robustness of the prediction results.
• Analysis of position-specific residue composition provided insights into carbonylation site characteristics.

Conclusions:

• The developed computational method offers an efficient and accurate approach for predicting yeast protein carbonylation sites.
• This tool can aid in the study of oxidative stress and its impact on proteins.
• Associated datasets and software are publicly available for research use.