BOSTON, March 16, 2026–Manifold Bio, a platform therapeutics company focused on direct-to-vivo drug discovery, has announced a collaboration with NVIDIA to validate the latter’s generative model for protein binder design, known as Proteina-Complexa. This joint study involved testing 1 million binder designs against 127 targets in a single multiplexed experiment, measuring over 100 million protein-protein interactions. The results indicated that specific binders were successfully identified for 68% of the targets tested, showcasing Proteina-Complexa as a competitive model in the field of protein binder design.
Dr. Pierce Ogden, Co-founder and Chief Technology Officer of Manifold Bio, emphasized the significance of the platform in enabling such an extensive study. “Manifold Bio’s platform uniquely enabled this massively multiplexed study, which establishes Proteina-Complexa as competitive with state-of-the-art methods,” he stated. He noted that the molecular synthesis and measurement technologies developed by Manifold aim to unlock the potential of generative AI in drug discovery, reinforcing the idea that more designs lead to more successful outcomes.
Anthony Costa, Director of Digital Biology at NVIDIA, highlighted the innovative architecture behind Proteina-Complexa. “Proteina-Complexa was built to generate protein binders at the speed and scale that drug discovery demands, powered by a novel architecture that redefines generative design,” he explained. Costa added that the model benefits from test-time scaling, which allows it to refine its logic before producing a final output. The successful testing of such a large number of designs against multiple targets is a testament to the scalability of the approach.
This experimental methodology was initially reported by Manifold in October 2025 when the company open-sourced mBER, an AI model designed for creating epitope-specific antibodies. The previous study utilized mBER to design and test over 1.1 million VHH antibodies against 145 diverse targets, further illustrating the capabilities of Manifold’s platform. The integration of multiplexed in vivo studies adds a crucial layer of physiologically relevant data for the AI-designed binders, enhancing the overall effectiveness of the drug discovery process.
The combination of de novo generative design and massively multiplexed measurement in the latest study produced thousands of new annotated structures, laying the groundwork for future training. This achievement sets a new benchmark for aligning the throughput of generative AI models with experimental validation, ensuring that advancements in AI can be matched by rigorous scientific testing.
Founded to revolutionize the landscape of drug discovery, Manifold Bio is committed to building the first AI-guided direct-to-vivo discovery platform. The company’s mDesign engine integrates AI-guided protein design with multiplexed in vivo data generation, fostering a virtuous cycle of design and testing. Alongside its internal pipeline of novel therapeutics, Manifold collaborates with leading global pharmaceutical partners to translate organism-scale biology into tangible clinical benefits.
This collaboration between Manifold Bio and NVIDIA not only underscores the growing importance of AI in drug discovery but also highlights the potential of generative models to accelerate the development of effective therapeutics. As the industry evolves, the implications of such high-throughput studies may redefine how drug candidates are identified and validated, paving the way for innovative treatments in the future.
For more information, visit Manifold Bio.
View the source version on businesswire.com: Business Wire.
Contacts
Media contact: [email protected]
Sam Altman Praises ChatGPT for Improved Em Dash Handling
AI Country Song Fails to Top Billboard Chart Amid Viral Buzz
GPT-5.1 and Claude 4.5 Sonnet Personality Showdown: A Comprehensive Test
Rethink Your Presentations with OnlyOffice: A Free PowerPoint Alternative
OpenAI Enhances ChatGPT with Em-Dash Personalization Feature
