From Bench to Bytes: When Tissue Engineering Leads to Artificial Intelligence
Seemingly unrelated career transitions can create unique competitive advantages in drug development, particularly when diverse scientific backgrounds converge in unexpected ways. A journey from tissue engineering research on tumor microenvironments at Cornell through cardiovascular immunology and spatial transcriptomics to AI-driven pharmaceutical sciences illustrates this principle. Spatial thinking developed through scaffold engineering enhanced vascular biology research, while microscopy expertise naturally evolved into computational image analysis capabilities. Understanding biological complexity at multiple scales, from cellular microenvironments to tissue architecture to physiology, now directly informs machine learning approaches for predicting drug behavior. Specialized skill sets often contain latent value that enables transitions across disparate fields, yet traditional linear career planning overlooks these opportunities at the intersection of converging disciplines. In modern scientific research, interdisciplinary mobility and diverse expertise increasingly drive innovation in pharmaceutical development, rewarding those who recognize connections between seemingly unrelated domains.
Bio: Michael G. McCoy III, Ph.D. ’18, is a senior scientist at Takeda Pharmaceuticals leading computational ADME sciences for small molecule, oligonucleotide, and digital twin initiatives. After completing his Ph.D. in tissue engineering at Cornell (2018), he pursued postdoctoral research in cardiovascular immunology at Cleveland Clinic, followed by a postdoctoral research appointment in noncoding RNA biology at Harvard Medical School where he investigated microRNA and lncRNA regulation in cardiometabolic disease. In 2022, he joined Takeda’s Drug Metabolism and Pharmacokinetics division to improve pipeline bioanalysis and support for AAV, cell, and oligonucleotide therapies before transitioning to lead AI/ML-driven ADME prediction across all therapeutic modalities. He currently leads enterprise-wide DMPK in silico strategy, developing multi-omics computational frameworks that integrate transcriptomic, proteomic, and metabolomic data with traditional ADME endpoints for advancing pharmacokinetic/pharmacodynamic predictions across all therapeutic modalities.