OGFE_RAAC

OGFE_RAAC predicts 2OG oxygenases from protein sequences using machine learning and a 673 amino acid reduced-alphabet feature representation to analyze sequence properties related to polarity and hydrophobicity.

Key Features:

• Machine Learning Application: Applies machine learning classification to identify 2OG oxygenases from protein sequences.
• Optimal Feature Representation: Recodes protein sequences using a 673 amino acid reduction alphabet to generate feature representations for prediction.
• High Predictive Accuracy: Achieves 91.04% accuracy under 10-fold cross-validation with independent dataset testing confirming robustness.
• Functional Feature Insights: Relates predictive features to sequence properties such as polarity and hydrophobicity to inform catalytic mechanism and substrate interaction hypotheses.

Scientific Applications:

• Bioinformatics and Molecular Biology: Supports computational identification of 2OG oxygenases for sequence-based functional annotation.
• Disease-related Studies: Applies to analyses of 2OG oxygenases implicated in disease contexts to inform biological hypotheses.
• Experimental Prioritization: Guides selection of candidate proteins for experimental validation, helping to reduce experimental workload and cost.
• Therapeutic Target Identification: Assists in identifying and characterizing enzyme candidates relevant to therapeutic research and enzyme–substrate interaction studies.

Methodology:

Uses machine learning classification on protein sequences recoded with a 673 amino acid reduced alphabet, evaluated by 10-fold cross-validation and validated on independent datasets.