Computational Algorithms, Methods, and Tools for Antibody Design

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications, trials & patents

Abstract

One potential solution for how to elicit antibodies against specific targets is precise immunogen design. Design and improvement of antibodies against targets that are immunorecessive, cryptic, or transient in their native context is a challenge for vaccine design. One potential solution is computational-aided antibody design and improvement. The ability of structural biology to provide atomic-level detail of antibody-antigen interactions and of computational biology to manipulate protein structure has raised the possibility of altering the antibody surface to improve antibody properties such as antigen binding and neutralization. Recent successes in paratope transplantation onto a protein scaffold and immune focusing using computer-aided antibody design highlights the usefulness of computational methodology. Rather than relying on a small set of well-defined tools, however, increasing evidence suggests that antibody design will require the systematic development of numerous computational algorithms, methods, and tools associated with the analysis, design and manipulation of protein structure. This also includes manipulation of quaternary protein structure and protein-ligand interactions. Moreover, introduction of machine learning to extract patterns that govern antibody functionality will prove to be essential in future antibody design efforts. Here we address a number of technological issues that are expected to impact antibody design and improvement. Specifically, we describe the development of new protein design and structure-modeling algorithms, methods, and tools that are particularly suited for antibody design.