The Reker lab tightly integrates biomedical data science and wet-lab experiments for the analysis and design of therapeutic opportunities. Automated experimentation can be guided by active machine learning to generate knowledge-rich datasets. A key aspect of our research is improving our understanding of the most effective active machine learning workflows to enable the broad deployment of adaptive machine learning and automated experimentation.
We focus our adaptive model development on critical drug properties such as efficacy, biodistribution, metabolism, toxicity, and side-effects. Prospective applications of these predictions enable us to better understand limitations of currently approved medications as well as design new drug candidates, nanoparticles, and pharmaceutical formulations. By integrating clinical data analysis, we can rapidly validate the translational relevance of our predictions and conceive big data-driven protocols for precision medicine and personalized drug delivery.