The “AI-ntibody” Competition: benchmarking in silico antibody screening/design

We recently contributed to a communication in Nature Biotechnology detailing an upcoming competition coordinated by Specifica to evaluate the relative performance of in vitro display and in silico methods at identifying target-specific antibody binders and performing downstream antibody candidate optimisation.

Following in the footsteps of tournaments such as the Critical Assessment of Structure Prediction (CASP), which have led to substantial breakthroughs in computational methods for biomolecular structure prediction, the AI-ntibody initiative seeks to establish a periodic benchmarking exercise for in silico antibody discovery/design methods. It should help to identify the most significant breakthroughs in the space and orient future methods’ development.

AI-ntibody not only reveals the relative performance of computational antibody methods in scenarios of varying difficulty, but pitches them against modern display library technologies (which have been subject to continual optimisation since the first methods were established in the mid-90’s), contextualising the performance of in silico methods against strategies currently employed in drug discovery campaigns.

The first AI-ntibody contest will be in the context of antibodies against the SARS-CoV-2 receptor binding domain and will assess the ability of methods to:
(a) identify higher-affinity candidates in NGS sequence data
(b) rank-order candidates by affinity
(c) generate novel antibody sequences (outside of the NGS data) that have higher affinity

Each antibody expressed will also be profiled with a panel of developability assays.

Participants can choose to take part in one or more challenges, noting that each expressed/assayed antibody will be subject to a (subsidised) fee since the tournament is not yet externally-funded.

The contest is open for registrants until 23rd January 2025. All information will soon be available at AIntibody.org.

Author