Automation and machine learning: informatic tools to engineer biofunctional polymers

Synthetic polymers have played a key role in medicine and drug delivery for the last 50 years. Their ability to be uniquely tailored for very specific biomedical needs makes them highly powerful tools. Here, we talk about how these material parameters can be tuned with incredible precision to achieve highly diverse characteristics. Recent advances in oxygen tolerant controlled/living radical polymer chemistry have also enabled a new and exciting ability to make polymer libraries using laboratory automation. Coupling this custom polymer automation with machine learning now allows us to sort through a very large range of characteristics to identify highly valuable compositions. This enables a transition from ‘screening’ experiments to intelligent profiling of quantitative structure-activity relationships (QSAR). In effect, we believe this unique set of tools may significantly enable the emerging field of polymer informatics, particularly for biomedical applications.

Automated and machine learning driven process for oxygen tolerant controlled/living radical polymer chemistry. Using this process, one can effectively take an informatics approach to new material designs.