Computational antibody modelling and prediction – SAbPred singularity platform
Antibodies are proteins that form the backbone of the human immune system. These molecules are highly designable allowing them to be harnessed as therapeutics. This makes antibodies the most prominent class of biologics, with a current market value of ~£70b.
At present, antibody therapeutics are developed using experimental pipelines that are very expensive, which in turn pushes up the price of the resulting drug. Facilitating this process by reducing the number of required experiments has the potential to accelerate and reduce the price of drug design.
Recent advances in understanding antibody structure and function have allowed the development of software that facilitates antibody engineering in this manner. Research led by Charlotte Deane, Professor of Structural Bioinformatics, is pioneering the development of such software and its dissemination to the pharmaceutical industry.
The Oxford protein informatics group
Prof Deane’s research group have developed the Structural Antibody Database (SAbDab) and Structural Antibody Prediction Platform (SAbPred) for the computational design of antibodies. The former collects and curates structural information on antibodies and the latter uses this data to perform modelling and prediction tasks that support antibody design. The software helps to prioritize which antibodies should be investigated further, and even to design entirely new antibodies.
The platform is currently used by leading biopharmaceutical companies to improve the design of antibody-based therapeutics.
The SAbDab-SAbPred platform is currently available for download and plug and play use as a Singularity container. Singularity uses a lightweight form of virtualisation which allows a Linux distribution with all its software to run in a special environment on top of an existing Unix-based operating system, such as Linux, macOS, or the Windows Subsystem for Linux.
The SAbDab-SAbPred Singularity container provides all the SAbDab-SAbPred tools in a single, portable format. Using the container, it is possible to run any of the SAbDab-SAbPred on any system with a Singularity installation without the need for any additional software installation, and the container can be easily managed and transferred between systems.
Unlike a virtual machine, software in a Singularity container runs at close to bare metal speed, without the performance overheads typically associated with virtualisation, resulting in optimal performance across all compatible systems.
about this technology