Integrating Mass Spectrometry to Correlate Multi Parametric Physiochemical Profiles to Develop AI/ML Models for the Rational Design of Hybrid Gold Nanoparticles as Drug Delivery Devices

Pranav Illendula

Co-Presenters: Individual Presentation

College: Hennings College of Science Mathematics and Technology

Major: MS.BS/SC/TEC/BTEC

Faculty Research Mentor: Swinton, Derrick  

Abstract:

Gold nanoparticles (AuNPs) possess properties that make them promising drug delivery platforms, including high stability, low cytotoxicity, biocompatibility, and tunable size, shape, and surface chemistry. Their use in biomedical applications, particularly anticancer therapy, continues to grow; however, challenges such as controlled release and target specificity remain, largely due to unwanted biomolecular adsorption. In biological environments, AuNPs rapidly acquire a protein corona that governs nanoparticle behavior, and this corona is strongly influenced by surface chemistry and morphology. Understanding these structure–function relationships is essential for designing hybrid core–shell nanoparticles. To address these challenges, we integrate mass spectrometry with RTCA and PCR to generate multi-parametric datasets for developing AI/ML models that guide the design of hybrid gold nanoparticles.We employ an integrated, multimodal analytical workflow that combines high-resolution mass spectrometry, real-time cell analysis (RTCA), and polymerase chain reaction (PCR) to generate a multiparametric dataset for rational design of hybrid gold nanoparticles (AuNPs). Mass spectrometry characterizes nanoparticle surface chemistry and protein corona composition, thereby defining key physicochemical interactions in biological environments. RTCA adds dynamic measurements of cellular responses, including adhesion, proliferation, and cytotoxicity, as cells interact with nanoparticles. PCR quantifies nanoparticle-induced gene expression changes, revealing mechanistic pathways related to apoptosis. Together, these complementary techniques provide a multidimensional dataset integrating physicochemical, cellular, and molecular readouts. The dataset is used to train advanced AI/ML models to predict cytotoxicity and link nanoparticle features to apoptosis-associated outcomes, with transfer learning strategies improving model accuracy and guiding data-driven optimization of hybrid AuNP drug delivery systems.We have developed a novel, integrative framework that combines physicochemical characterization, real-time phenotypic monitoring, and molecular profiling with advanced artificial intelligence and machine learning to guide the rational design of hybrid AuNPs. Unlike conventional nanoparticle development approaches that rely on isolated assays or empirical optimization, this platform generates synchronized, multi-scale datasets linking surface chemistry, cellular behavior, and gene regulatory responses, and ultimately streamlines this process through automation.Keywords: Nanotechnology, Drug Delivery Devices, Artificial Intelligence, Machine Learning, Omics, Real-Time Cell Analysis