Accurately predicting nitrosylated tyrosine sites using probabilistic sequence information

Area of Science:

• Biochemistry
• Computational Biology
• Bioinformatics

Background:

• Post-translational modification (PTM) involves enzymatic changes to proteins after synthesis.
• Nitrotyrosine, a key PTM, is linked to diseases involving inflammation and cell damage.
• Current experimental methods for identifying nitrotyrosine sites are time-consuming and expensive.

Purpose of the Study:

• To develop a novel machine learning approach for accurate nitrotyrosine site prediction.
• To provide a computationally efficient alternative to experimental methods.

Main Methods:

• Utilized sequence coupling information from amino acid neighbors of tyrosine residues.
• Employed a support vector machine (SVM) as the core classification algorithm.
• Developed PredNitro, an online prediction tool.

Main Results:

• PredNitro achieved 98.0% accuracy in predicting nitrotyrosine sites.
• Demonstrated high performance with MCC > 0.96 and AUC > 0.99.
• Outperformed previous prediction methods in cross-validation tests.

Conclusions:

• PredNitro offers a highly accurate and efficient method for identifying nitrotyrosine sites.
• The tool can accelerate research in diseases associated with nitrotyrosine modification.
• PredNitro is available as a public online predictor.