EpiDope

EpiDope predicts linear B-cell epitopes from protein sequences using a deep neural network to support epitope identification for serodiagnostics and immunological research.

Key Features:

• Linear B-cell epitope prediction: Predicts regions of antigenic protein sequences likely to form linear B-cell epitopes.
• Sequence-level analysis: Operates on individual protein sequences to identify epitope candidate regions.
• Deep learning implementation: Uses a deep neural network model implemented in Python.
• Performance metrics: Evaluated by ROC analysis with reported AUC of 0.67 ± 0.07 and improved AUC10% (AUC at false-positive rate < 0.1) compared to other methods.

Scientific Applications:

• Serodiagnostic assay development: Supports selection of linear B-cell epitopes for serodiagnostic assays.
• Therapeutic optimization: Aids identification of epitope candidates relevant for optimizing medical therapies.
• Epitope mapping and immunology research: Facilitates mapping of antibody-binding regions and related immunological studies.

Methodology:

Analysis of individual protein sequences using a deep neural network implemented in Python with evaluation by ROC analysis reporting AUC (0.67 ± 0.07) and AUC10% (AUC at false-positive rate < 0.1).