BIOLOGICS

PROPHET-Ab: A Data Platform for Antibody Developability

Optimized and Automated Antibody Developability Assays: Now Available from Ginkgo Datapoints

Authors: Ammar Arsiwala, Rebecca Bhatt, Yaoyu Yang, Porfi Quintero Cadena, KC Anderson, Xiang Ao, Lood van Niekerk, Adam Rosenbaum, Aanal Bhatt, Alex Smith, Lucia Grippo, Xing Cao, Rich Cohen, Jay Patel, Olga Allen, Ali Faraj, Anisha Nandy, Jason Hocking, Berk Tural, Sara Salvador, Joe Jacobowitz, Kristin Schaven, Mark Sherman, Sanjiv Shah, Peter M. Tessier, David Borhani

Beyond binding and function: that's where the bigger challenge lies in antibody drug development. High expressibility, long half-life, low aggregation, high solubility—these are the keys to getting your drug from the lab to the patients who need it. But let's be honest, predicting developability is tough. We're building better tools to tackle this challenge, and it starts with data. Lots of it. Curious? Let's dive in...

Our latest preprint describes PROPHET-Ab, a high-throughput antibody developability assessment platform that enables data-generation at the scale needed to develop improved ML models to predict antibody developability. We run 246 therapeutic IgGs through 10 assays and deliver an ML-ready developability dataset.

A high-throughput platform for biophysical antibody developability assessment to enable AI/ML model training

Successful antibody drugs exhibit important “developability” properties, beyond tight and specific binding to their target, including high expressibility, high stability and solubility, low aggregation propensity, low viscosity, low polyreactivity, and long in vivo half-life. Collectively, developability properties predict favorable manufacturing, storage, administration, and safety, and deficiencies in these properties increase risk for clinical failure. Despite progress in developing machine learning models to predict structure and binding, antibody developability models lag behind, largely due to a lack of sufficiently large training datasets. We have built a high-throughput platform, PROPHET-Ab, that enables data generation at the scale needed to train improved AI/ML models to predict antibody developability.

What is PROPHET-Ab useful for?

Antibody drug discovery  is increasingly benefiting from powerful AI tools to predict binding. Making use of large, high quality protein structure datasets, they learn the key structural features that govern antibody-target interaction. But other critical properties - aggregation, viscosity, solubility, and thermal stability - have lagged behind. PROPHET-Ab was built to precisely fill this gap - generate high-quality antibody developability datapoints at the scale required by AI.

The PROPHET-Ab platform adapts and scales up classical assays like AC-SINS (self-association), HIC (hydrophobicity), and polyreactivity, to process thousands of antibodies per week. The results are consistent with low-throughput legacy methods, but come in ML-ready format.

The long-term vision of PROPHET-Ab is to make antibody therapeutics easier to engineer. More data leads to better models, smarter designs and more effective drugs. As multi-specifics and complex formats become the norm, scalable developability platforms like PROPHET-Ab are the key infrastructure supporting the next wave of therapeutic innovation.

Want to see what PROPHET-Ab can do for your antibody therapeutic?

Get in touch with the Ginkgo Datapoints team to start designing your developability data package today.