- Docente: Cosmin Stefan Butnarasu
This MOOC explores the impact of artificial intelligence on the discovery and development of drugs. Starting with the definition of a drug and the main stages of the pharmaceutical development process, it explains why AI is increasingly being used to manage complexity, large datasets, high failure rates, and costly setbacks in later stages. Participants explore key AI concepts, such as machine learning, deep learning, generative AI, and complex language models. They also examine how these approaches are applied to target identification, protein structure prediction, chemical space exploration, virtual screening, preclinical development, clinical trials, pharmacovigilance, and responsible AI practices.
The course is organized into four blocks and fourteen modules.
Block 0 introduces the objectives and structure of the course.
Block 1 presents the definition of a drug, the drug discovery and development process, and the reasons for adopting AI in pharmaceutical research and development.
Block 2 introduces the conceptual foundations of AI, including machine learning, deep learning, generative AI, model validation, and the limitations of data quality.
Block 3 focuses on AI applications throughout the entire drug development process, including target identification, exploration of chemical space, ligand-based and structure-based virtual screening, preclinical development, clinical trials, pharmacovigilance, and the limitations, threats, and opportunities of AI.
The course includes short quizzes for each module, a final quiz, and interactive H5P exercises for self-assessment and knowledge consolidation.
By the end of the course, participants will be able to:
-
Define what a drug is and distinguish between an active ingredient, a medicine/pharmaceutical product, and a treatment.
-
Describe the main stages of the drug discovery and development process and identify key decision points and risks of failure.
-
Explain why AI is increasingly used in pharmaceutical research and development, particularly for prioritization, prediction, and early signal detection.
-
Distinguish between artificial intelligence, machine learning, deep learning, and generative AI.
-
Interpret the basic logic of model training, validation, testing, overfitting, bias, and data quality limitations.
-
Explain how AI can support target identification, data analysis, target structure prediction, chemical space exploration, and virtual screening.
-
Compare ligand-based and structure-based virtual screening approaches and describe their main requirements and limitations.
-
Describe examples of AI applications in preclinical development, clinical trials, and pharmacovigilance.
-
Adopt a responsible AI mindset by recognizing the importance of human oversight, clear context of use, validation, governance, and lifecycle monitoring.
Quiz completion.
No formal prerequisites are required. A basic knowledge of biology, chemistry, biotechnology, pharmacy, medicine, or biomedical sciences is recommended. No programming experience is required.
Cosmin Stefan Butnarasu
cosminstefan.butnarasu@unito.it
For technical issues:
edvancedeh@unito.it