AI and Automation Revolutionize Protein Engineering in Next-Generation Biologics

The pharmaceutical industry is witnessing a paradigm shift in protein engineering, driven by artificial intelligence (AI) and advanced automation technologies. As bispecific antibodies, antibody-drug conjugates (ADCs), and novel protein formats move into clinical stages, drug discovery teams are facing new challenges in engineering and producing increasingly complex molecules.

The Rise of Complex Biologics

The biologics landscape has evolved significantly beyond traditional monoclonal antibodies. With more than 200 ADCs in clinical stages and bispecific antibodies generating over $12 billion in sales in 2024, the industry is embracing a new era of structural diversity. The FDA's increasing support for novel modalities in 2025 has further catalyzed this trend.

However, this progress comes with its own set of challenges. Paul Wan, vice president of early discovery and business development at Viva Biotech, notes, "The biggest problem with these biologics is really the heterogeneity in actually generating these molecules. You're getting mispaired chains in bispecifics or trispecifics and also different conjugation variants when you're looking at ADCs."

To address these issues, the field has developed platform-level engineering approaches. These include controlled Fab-arm exchange, knobs-into-holes techniques, and crossMabs for bispecifics, as well as precise conjugation methods for improved drug-to-antibody ratio (DAR) control in ADCs.

AI-Driven Design and Manufacturability

Artificial intelligence has become an integral part of the protein engineering cycle, uniting sequence, structure, and manufacturability predictions. The evolution of AI models from conventional machine learning to deep neural networks and transformer-based architectures has significantly improved the accuracy of protein structure prediction.

Yue Qian, executive director of computational chemistry and artificial intelligence platform at Viva Biotech, explains, "Conventionally, there are several different ways to extract protein features and to present these proteins. These are mostly sequence-based. But nowadays, since we have protein large language models, they provide much richer approaches to better describe these proteins."

Viva Biotech's AIDD platform, comprising V-Scepter, V-Orb, and V-Mantle, exemplifies the integration of AI in the protein engineering workflow. These tools help teams move from concept to manufacturable candidates with greater confidence, addressing challenges in early parameterization, physics-based modeling, and generative AI design.

Automation and High-Throughput Screening

Parallel multistep purification and automation have dramatically increased the scale and speed of protein production and screening. Some laboratories now express up to 2,500 antibodies simultaneously, rapidly screening for optimal formats including bispecifics and tetraspecifics.

Jerry Zhang, director of biology at Viva Biotech, highlights the value of this approach for structural biology: "High-throughput protein production can allow us to try different variations, and potentially it can help us to find the best construct that we are looking for that we can use for protein production or some structural biology or even some assays."

However, as predictive tools improve, the fundamental approach to high-throughput screening may shift. Paul Wan predicts, "AI will be able to predict which construct to use, which buffers to use automatically. So, how will the evolution of high-throughput screening really develop as AI becomes more prominent in predicting one or two constructs or one or two protocols or expression systems rather than actually screening these huge numbers? That will be interesting to see in the future."