We combine the discovery power of machine learning (ML) methods with enhanced high-throughput screening techniques, to screen protein libraries for hits against targets of interest (specified by the client). Our team has broad experience with cancer and infectious disease targets of various levels of complexity. Utilizing the appropriate combination of data-generation and ML methods, we minimize the length and cost of the hit-finding process for biologics pipelines, while maximizing coverage. We rapidly mine hundreds of millions of lead peptides and antibodies against specific targets. Our platform surveys the virtually infinite space of sequences for biosuperiors with therapeutic and developable profiles produced with competitive affinity and efficacy.

Our screening platform (using an improved design of surface plasmon resonance (SPR) nano sensors) provides binding data on the target of interest, for thousands of proteins in a library. We use this data to train ML networks that learn the relation between proteins’ chemical structure and their binding to the target. The trained network is deployed to screen much larger libraries of millions of proteins, maximally leveraging the experimental data, delivering quality hits at an unprecedented rate.