This hands-on mini-course introduces practical AI workflows for drug discovery and development, beginning with legacy pitfalls - poor translation, docking bias, single-target limits - and why AI can solve them. Students clean and curate public datasets, map data types to model families, train a graph-CNN and LLM with low/no-code tools, design lab-in-the-loop experiments, and evaluate hits for toxicity, and commercialization potential. Guest speakers from pharma and venture capital share real-world AI wins. No deep programming required; ideal for experimental and computational scientists seeking to learn applied AI applications.
1 units · Medical Satisfactory/No Credit
This hands-on mini-course introduces practical AI workflows for drug discovery and development, beginning with legacy pitfalls - poor translation, docking bias, single-target limits - and why AI can solve them. Students clean and curate public datasets, map data types to model families, train a graph-CNN and LLM with low/no-code tools, design lab-in-the-loop experiments, and evaluate hits for toxicity, and commercialization potential. Guest speakers from pharma and venture capital share real-world AI wins. No deep programming required; ideal for experimental and computational scientists seeking to learn applied AI applications.
Offered in Winter 2027 at Stanford University.