Artificial intelligence is transforming the pharmaceutical industry by compressing the lengthy timelines traditionally associated with drug discovery. While developing biologic medicines—therapies derived from engineered proteins—has historically been a costly and failure-prone endeavor, machine learning models are now being used to navigate the vast landscape of potential molecular combinations.

AstraZeneca is among the companies integrating these computational tools into their research and development infrastructure. According to Puja Sapra, senior vice president and head of R&D biologics engineering and oncology targeted discovery at AstraZeneca, the company employs a “build-measure-learn” loop. In this process, AI models prioritize candidate molecules, allowing scientists to focus laboratory resources on the designs most likely to succeed. This approach aims to reduce dead ends and enable the pursuit of disease targets previously considered untreatable.

Beyond accelerating existing workflows, AI is facilitating the creation of more complex, multi-specific biologics that can hit multiple disease pathways or deliver payloads to specific cells. To support these advancements, AstraZeneca is developing a “lab of the future” in Cambridge, Massachusetts. This facility is designed to function as a closed-loop system where AI predictions, robotic experimentation, and automated data generation feed directly back into the models to refine future iterations.

Advertisement

The ultimate goal for many in the field is “de novo” design, where AI generates entirely new protein sequences from scratch. Achieving this requires high-quality, standardized training data and robust safety prediction capabilities. To address safety, researchers are utilizing virtual clinical trials, which pair advanced cell systems and micro-scale organ models with AI to evaluate potential molecules before they reach human testing.

Despite the shift toward autonomous, agentic AI systems, human oversight remains a critical component of the process. Scientists continue to provide the strategic direction and judgment necessary to ensure that AI-driven outputs are explainable and ethically sound. As engineering teams work to build systems that act as “thinking partners,” the collaboration between machine learning experts and biologists is expected to remain central to the development of next-generation therapies.

Source: MIT Technology Review