iProtGly-SS

iProtGly-SS predicts protein glycation sites and identifies lysine glycation sites by integrating sequence and structural protein information with a Support Vector Machine (SVM) classifier to support investigation of glycation-related mechanisms in Parkinson's disease, Alzheimer's disease, and diabetes mellitus vascular complications.

Key Features:

• Prediction target: Predicts protein glycation sites, describing glycation as the non-enzymatic attachment of sugar molecules, and specifically identifies lysine glycation sites.
• Data integration: Leverages sequential (sequence) and structural protein information to enhance prediction accuracy.
• Algorithm: Employs a Support Vector Machine (SVM) classifier for site prediction.
• Benchmark evaluation: Assessed on three benchmark datasets with reported accuracies of 81.61%, 93.62%, and 92.95%.
• Performance improvement: Reports higher identification accuracy of lysine glycation sites compared to previous computational methods.

Scientific Applications:

• Posttranslational modification analysis: Enables identification of glycation sites to study effects of this PTM on protein function.
• Neurological disease research: Supports investigation of glycation-related mechanisms in Parkinson's disease and Alzheimer's disease.
• Diabetes complications research: Supports study of vascular complications related to diabetes mellitus involving protein glycation.
• Therapeutic development: Informs development of strategies targeting glycation modifications.

Methodology:

Integrates sequence and structural protein features and applies a Support Vector Machine (SVM) classifier; performance was assessed on three benchmark datasets yielding accuracies of 81.61%, 93.62%, and 92.95%.