Computer aided de novo design of nanobodies

Antibodies have become major players in the pharmaceutical industry and were valued at 0.16 US billion in 2023.

Traditionally, antibodies are obtained after immunisation of different animals and then produced in relevant cells for use in research, diagnostics and therapy.

In recent years, there is a growing public opinion in Europe to ban the use of animals for biomedical research, and therefore there is an increasing pressure to move from animal produced antibodies to design and produce them in vitro.

Antibody engineering has another important advantage, which is the possibility of targeting a precise epitope and not rely on serendipity as when injecting an animal with an antigen.

In recent years, there have been significant advances in protein design based on the use of artificial intelligence and precise force fields.

Despite this, the majority of the companies that work on antibody design combine rational engineering with massive proprietary screening methods, and so far, there are no reported cases of fully de novo design of an antibody with nM affinity against a defined epitope.

Using an interleukin receptor as a case study, we have shown that we can indeed fully design de novo a nanobody that recognizes the target with nM affinity, using our proprietary protein design software FoldX and ModelX.

Experts consulted to date indicate that the results obtained so far are truly impressive, prompting us to continue validating and optimising our process.

Our proposal has two main objectives: First, to fully automate our pipeline that involves epitope selection, antibody framework selection and docking, backbone move and side chain search.

Second, demonstrate that our optimised pipeline can design fully de novo nanobodies against a defined target in a fast and cost-effective way.

Success in both objectives will open the way to fully de novo antibodies with desired properties, and position ourselves in the search for funding and spin-off incorporation.