Graduate Student Seminar Series – Nijash Sooriyakumaran

Graduate Student Seminar Series
Please ensure you invite your Principal Investigator by adding their email via the ‘Add Guest’ button and they will also be notified of your presentation.
Location: 2nd Floor Auditorium (TRI/KITE) – 550 University Ave
Presentation Title: Development of Hybrid Machine Learning Algorithms Combining Model-Based and Model-Free Methods for The Discovery of Cell Culture Medium Formulations Amenable to Industrial Translation
Abstract: Chemically-defined serum-free cell culture medium is important due to replicability, safety, standardization of quality during experimentation, and many other reasons. Cell culture medium is comprised of hundreds of compounds, consisting of required growth factors, essential vitamins and minerals, amino acids, amongst others. As a result, the search space for optimizing formulations becomes exceedingly large (for instance, conducting a full screen for a medium composed of 100 factors at 6 possible dosage levels would require testing 6100 formulations!). In this research, a novel algorithm is developed to address the issues that arise as more components are added to media. Alongside development, testing is done on various in-silico problems to test robustness and behaviour before in-vitro testing.
Supervisor Name: Julie Audet
Year of Study: 2
Program of Study: MASc
Reschedule Reason: Update to title and abstract
Powered by Calendly.com