CAREER: Decoding Polymer-Drug Interactions with AI for Rational Carrier Design

About This Grant

NON-TECHNICAL SUMMARY: Many promising new medicines fail to reach patients because they do not dissolve well in water, making it hard for the body to absorb them. This poor solubility affects over 90% of drugs in development, limiting treatment options, increasing costs, and often forcing researchers to abandon otherwise effective compounds. Polymeric micelles are tiny, self-assembled carriers made from long-chain molecules that offer a powerful way to encapsulate and deliver these hydrophobic drugs safely and efficiently. However, current designs often achieve only low drug loading, requiring excessive carrier material that limits treatment effectiveness and clinical outcomes. This CAREER project tackles this challenge using artificial intelligence (AI) and machine learning to decode the fundamental molecular interactions between polymers and drugs. By systematically evaluating key interaction forces such as π-π stacking and hydrogen bonding, and integrating advanced quantum computational tools with precisely curated experimentation, the research establishes predictive principles for drug loading. These physics-based interaction energies and structures train machine learning models to predict drug loading and recommend improved carrier designs, linking quantum-level insight to data-driven prediction. This shifts drug carrier design from a tedious trial-and-error paradigm into rational, AI-guided engineering of high-loading carriers tailored to specific drugs, paving the way for a new generation of nanomedicine and more effective, affordable treatments. Broader impacts of this work include training students in AI-enabled materials and drug delivery, developing new undergraduate laboratory modules, and an elective course that integrates computation with formulation science, providing research experiences for K-12 and graduate students, and creating an educational game, "Code-a-Cure," that introduces learners to nanomedicine design and modern AI-driven discovery. TECHNICAL SUMMARY: Over 90% of active pharmaceutical ingredients in development exhibit poor aqueous solubility, severely limiting bioavailability and creating a major bottleneck in drug formulation. Polymeric micelles can encapsulate hydrophobic drugs through non-covalent interactions, yet drug loading is low for most compounds because current design approaches lack a predictive, mechanistic understanding of specific polymer-drug forces. This CAREER project establishes molecular principles that govern polymer-drug affinity and converts them into physics-informed machine learning tools for rational, high-loading carrier design. The central hypothesis is that systematic quantification of π-π interaction energetics and valency, hydrogen bonding strengths, and their interplay in mixed-functionality copolymers yields predictive descriptors of loading capacity and efficiency. Building upon this understanding, research is organized into three synergistic thrusts: (1) mapping π-π interactions by systematically varying electronics and pendant density to capture non-linear valency effects; (2) evaluating hydrogen bond donor/acceptor strengths and valency; and (3) investigating synergistic or antagonistic effects in mixed functionality copolymers. Experimental characterization of loading is integrated with density functional theory (DFT) calculations into supervised machine learning models with uncertainty estimation and an active-learning loop that prioritizes the most informative polymer-drug experiments. A generative forward design search strategy further recommends polymer structures and compositions predicted to improve loading. This proposed work aims to broaden participation while highlighting the importance of biomaterials, nanomedicine, and polymer science through integrated educational initiatives such as the development of a "Code-a-Cure" educational game, new undergraduate laboratory modules, an elective course incorporating AI and drug delivery concepts, and outreach providing hands-on experiences for K-12 and undergraduate students. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.