SEPPA

SEPPA predicts spatial epitopes on protein antigens to identify conformational epitope regions recognized by antibodies.


Key Features:

  • Unit Patch of Residue Triangle: Introduces the "unit patch of residue triangle" to represent local three-dimensional arrangements of surface residues.
  • Clustering Coefficient: Incorporates a clustering coefficient that quantifies the spatial compactness of surface residues to identify likely epitope regions.
  • Performance Metrics: Validated on independent testing datasets with an average Area Under the Curve (AUC) value over 0.742 and a successful pick-up rate of 96.64%.
  • Comparison with Other Methods: Demonstrates improved performance relative to CEP, DiscoTope, and BEpro in benchmark comparisons.
  • Threshold Scores for Confidence Levels: Provides threshold scores corresponding to specific levels of accuracy, sensitivity, and specificity.

Scientific Applications:

  • Vaccine Design: Assists identification of antigen surface regions for vaccine antigen selection.
  • Antibody Development: Aids identification of target epitopes for antibody engineering and selection.
  • Immunotherapy: Supports identification of therapeutic targets and characterization of immune responses at the molecular level.
  • Epitope Mapping of Discontinuous Regions: Enables analysis of conformational/discontinuous epitope regions involved in complex protein–antibody interactions.

Methodology:

Defines a "unit patch of residue triangle" to describe local surface residue arrangements, computes a clustering coefficient to measure spatial compactness, applies threshold scores for confidence calibration, and was validated on independent testing datasets and benchmarked against CEP, DiscoTope, and BEpro.

Topics

Details

Tool Type:
web application
Added:
2/14/2017
Last Updated:
12/10/2018

Operations

Data Inputs & Outputs

Epitope mapping

Publications

Sun J, et al. SEPPA: a computational server for spatial epitope prediction of protein antigens. Nucleic Acids Res. 2009; 37:W612-6. doi: 10.1093/nar/gkp417

PMID: 19465377