Combining the capabilities of a leading antibody-discovery organization with cutting-edge machine learning unlocks transformative opportunities in biotherapeutics discovery. At AbbVie, we are embedding predictive models directly into scientific workflows, addressing all aspects of a target product profile. This innovation enables programs to progress faster through stage gates, while also making better medicines by exploring previously inaccessible sequence space. We will share the principles that guide our building of this powerful predictive engine, which relies on a robust data platform and is powered by cross-functional teams shaping the future of biotherapeutics discovery.

NextGenPLM: a new paradigm in multimodal foundational models that fuses sequence, 3D structure, and interaction data in a single efficient transformer—enabling high-throughput, repertoire-scale antibody–antigen screening. On a diverse benchmark of antibody–antigen complexes, matches Chai-1 and Boltz-1x on contact-map and epitope accuracy at a fraction of the compute cost. In an internal affinity-maturation campaign, ranking mutants by predicted contact probabilities and masked-language-modeling (MLM) log-likelihoods helped achieve up to 17× affinity improvements—demonstrating its potential for rapid, data-driven biologics discovery.

Operationalizing machine learning and generative AI in biologics discovery means more than building models—it involves embedding them into scientific workflows to run analyses alongside experimental data. This talk will cover strategies for streamlining data usage for model training, integrating in silico-generated data with laboratory results, and overcoming implementation challenges.

Miniproteins are a powerful yet underutilized therapeutic modality with a structure that enables binding with high affinity and specificity to their targets and achieves remarkable stability using only canonical amino acids. By combining de novo design with synthetic biology and laboratory automation, we accelerate the discovery and optimization of protein binders and create a vast toolbox of modular miniprotein domains, each with ideal drug-like properties and developability profiles.

The EDA Toolkit is an open-source Python library designed to standardize and accelerate exploratory data analysis across domains. It combines automated data profiling, visualization, and preprocessing in a single, reproducible framework. This presentation demonstrates how the toolkit simplifies data strategy, enhances collaboration between teams, and promotes trust in AI model development through transparency and reproducibility.