The pharmaceutical industry is undergoing a significant shift as machine learning models become central to the research and development of biologic medicines. By integrating computational design into the traditional drug discovery process, companies aim to reduce the time and high failure rates typically associated with creating new therapies, particularly those derived from engineered proteins.
AstraZeneca is among the organizations adopting a "build-measure-learn" framework to streamline this process. According to Puja Sapra, senior vice president and head of R&D biologics engineering and oncology targeted discovery, the company uses AI to prioritize molecular candidates computationally. This allows scientists to focus laboratory resources on the designs most likely to succeed, effectively narrowing the vast number of potential molecular combinations to a manageable set.
The application of these technologies extends beyond simple acceleration. Researchers are now exploring the design of multi-specific biologics capable of hitting multiple disease pathways or delivering therapeutic payloads to specific cells. This requires balancing complex variables such as potency, stability, and manufacturability. To support these efforts, AstraZeneca is developing a "lab of the future" in Kendall Square, Cambridge, Massachusetts. This facility aims to create a closed-loop system where AI models generate predictions, and robotic systems execute the corresponding experiments, feeding data back into the models to refine future iterations.
A primary goal for the field is "de novo" design, where AI generates entirely new protein sequences from scratch. Achieving this requires high-quality, standardized training data and the ability to predict the safety of generated molecules within the human body. To address safety, researchers are utilizing virtual clinical trials, which pair AI with advanced cell systems and micro-scale organ models to simulate biological responses.
Despite the move toward autonomous systems, human expertise remains a critical component of the workflow. Sapra emphasizes that scientists and engineers are essential for providing oversight, strategic direction, and ensuring that AI outputs are explainable and ethical. The collaboration between human talent and machine learning is viewed as the key to unlocking treatments for diseases that were previously considered untreatable.
Source: MIT Technology Review
No comments yet. Be the first to share your thoughts.