Day One

• Chai successfully designs antibodies against a range of therapeutically relevant, challenging targets, achieving high binding hit rates with at most 96 designs per target, where the large majority also meet developability profiles on par with clinical-stage therapeutic antibodies
• Experimentally determined structures of Chai-designed antibodies match their in silico predictions with sub-Angstrom accuracy while differentiating in sequence and structure from existing binders, demonstrating that de novo design can now deliver atomically accurate models of novel molecules
• Case studies including functional antibodies that mediate GPCR agonism, and highly specific antibodies that selectively discriminate cancer neoepitopes, demonstrate how de novo design is opening up creative, targeted therapeutic strategies beyond accelerating drug discovery

This structured networking session provides the perfect opportunity to connect with industry frontrunners and key opinion leaders working at the interface of AI/ML and computational drug design for biologics. Establish meaningful connections to build upon at the rest of the conference.

• Optimized AI generation achieves higher hit rates, nearly tripling computational design success and drastically slashing early-stage trial-and-error costs
• Wet-lab results confirm that our AI-designed sequences successfully fold and exhibit real catalytic activity, providing validated functional enzymes
• Sequence optimization significantly boosts antibody expression, thus improving production yields and directly derisking downstream manufacturing bottlenecks

• Developing an end-to-end AI-MPRO platform to design miniprotein therapeutics against challenging targets, including ion channels and complex multi-protein interfaces that are difficult to address with conventional modalities
• Combining de novo protein design with rapid in silico experimental iteration, enables efficient progression from computational discovery to in vitro and in vivo preclinical proof-of-concept
• Generating compact, highly optimized miniprotein candidates with the potential to deliver first-in-class or best-in-class therapies for targets traditionally considered intractable

• Shifting from screening-based discovery towards biologics by design through AI momentum and applications
• Exploring how real impact today is accelerating design-make-test cycles and reducing experimental iteration
• Recognizing that success requires integrated AI platforms that combine generative models, predictive models, and experimental feedback

• Exploring how a de novo protein was designed to target a heterodimeric receptor in the absence of a solved structure
• Achieving biased agonism with a de novo protein
• Emphasizing the importance of a human-in-the-loop AI/ML pipeline for biologics design

AI/ML adoption in biologics discovery is often constrained not by model capability, but by fragmented data, inconsistent experimental protocols, and limited interoperability across discovery workflows. This workshop focuses on the data, workflow, and operational data foundations required to deploy AI meaningfully and at scale across protein‑based discovery campaigns.

Participants will explore:

• Strategies for generating, capturing, and standardizing diverse experimental datasets to support AI‑ready biologics discovery, including how to leverage and derisk historical datasets
• How to benchmark biologics data maturity against small molecule discovery – which lessons translate, and which do not?
• What it takes to digitalize discovery workflows, including: NGS data alignment to antibody CDRs, epitope discovery and sequence diversification, library analysis and assay‑spanning data integration
• Implementing harmonized experimental protocols and metadata standards to enable robust AI/ML training and validation
• Integrating in silico tools with conventional computational methods and wet‑lab validation through human‑in‑the‑loop QA/QC
• Best practices for scaling AI/ML workflows across the enterprise, including: MLOps and dataset versioning, bias detection and mitigation, AI/ML model generalizability across discovery campaigns
• Emerging federated learning and consortium‑led approaches (e.g. FAITE) to address data scarcity, heterogeneity, and inter‑lab/inter-protocol variability