ElliPro: A Tool for Predicting Antibody Epitopes
Author Information
Author(s): Ponomarenko Julia, Bui Huynh-Hoa, Li Wei, Fusseder Nicholas, Bourne Philip E, Sette Alessandro, Peters Bjoern
Primary Institution: San Diego Supercomputer Center, University of California, San Diego
Hypothesis
Can a new web-tool effectively predict antibody epitopes using a structure-based method?
Conclusion
ElliPro is a useful tool for predicting antibody epitopes, outperforming other methods in terms of accuracy.
Supporting Evidence
- ElliPro achieved an AUC value of 0.732, indicating strong predictive performance.
- The tool was tested on a benchmark dataset of 39 epitopes.
- ElliPro's predictions ranked in the top three for over 70% of proteins analyzed.
Takeaway
ElliPro helps scientists find parts of proteins that antibodies can recognize, which is important for making vaccines.
Methodology
ElliPro uses a web-based application to predict antibody epitopes based on protein structure and a residue clustering algorithm.
Limitations
ElliPro's predictions may not always correlate well with known epitopes, especially for larger or multi-domain proteins.
Statistical Information
P-Value
5.55E-10
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website