AodPred

AodPred predicts antioxidant proteins from amino acid sequence information using a support vector machine classifier to support identification of proteins involved in mitigating cellular and DNA damage caused by free radical intermediates.

Key Features:

• Support Vector Machine-Based Prediction: Employs a support vector machine (SVM) classifier for sequence-based discrimination of antioxidant versus non-antioxidant proteins.
• Optimal 3-Gap Dipeptide Encoding: Encodes protein sequences using an optimal 3-gap dipeptide representation to capture discriminative sequence features.
• Validated Accuracy: Demonstrated overall accuracy of 74.79% assessed by jackknife cross-validation.

Scientific Applications:

• Pharmacology: Supports exploration of therapeutic applications of antioxidant proteins in disease prevention and treatment.
• Molecular Biology: Aids elucidation of antioxidant protein roles in mitigating cellular and DNA damage caused by free radical intermediates.
• Bioinformatics: Provides a machine-learning approach for identifying antioxidant proteins from sequences to support computational analyses.

Methodology:

Protein sequences are encoded with the optimal 3-gap dipeptide method, a support vector machine classifier is trained on known antioxidant proteins, and performance is evaluated using jackknife cross-validation.