PAbFold: Linear Antibody Epitope Prediction using AlphaFold2.

Area of Science:

• Immunology
• Structural Biology
• Bioinformatics

Background:

• Defining antibody binding epitopes is crucial for understanding antibody-antigen interactions and molecular functions.
• Current empirical methods for determining linear antibody epitopes are labor-intensive, time-consuming, and costly.

Purpose of the Study:

• To develop a computational pipeline leveraging AlphaFold2 for rapid and accurate prediction of linear antibody epitopes.
• To assess the PAbFold pipeline's ability to identify known and novel antibody epitopes.

Main Methods:

• Developed PAbFold, a pipeline using AlphaFold2 to predict antibody-target peptide complex structures.
• Grafted antibody complementarity determining regions (CDRs) onto single-chain variable fragments (scFv) with varying frameworks.
• Validated PAbFold predictions experimentally for a novel anti-SARS-CoV-2 nucleocapsid antibody.

Main Results:

• PAbFold accurately identified known linear epitopes for HA and Myc targets.
• The pipeline successfully predicted the epitope of a novel anti-SARS-CoV-2 nucleocapsid antibody.
• Methodological details like peptide length and AlphaFold2 versions influence prediction accuracy.

Conclusions:

• The PAbFold pipeline enables rapid and accurate identification of linear antibody epitopes from protein sequences alone.
• This computational approach offers a cost-effective alternative to traditional experimental methods.
• PAbFold has significant implications for antibody research, drug development, and diagnostics.