Development of a machine learning-based in-process quality monitoring technology for plants during vaccine manufacturing

Challenge: Biological systems are often difficult to implement and maintain in an industrial setting. There is a large number of parameters to optimize and the complexity is great. Artificial intelligence (AI) can assist in uncovering patterns and identifying features among parameters that can improve the effectiveness of a process as well as the requirement for a sustained high-quality production. Medicago R&D is a leading clinical-stage biotechnology company that uses proprietary plant-based technologies to rapidly develop and produce novel vaccines and antibodies. In this project, the company will benefit from AI approaches to improve its production platform.

Solution: The project is led by an interdisciplinary team combining expertise in big data analytics and metabolomics, machine and deep learning as well as vaccine production. The team aims to enhance the productivity of the Quebec vaccine manufacturer Medicago R&D. In this project, high throughput metabolomic analyses coupled to machine and deep learning algorithms will be used to identify molecular signatures of productive plants and define the molecular boundaries of parameters that impact the productivity of the pharmaceutical production process. More specifically, the project aims at developing a mass spectrometry-based system that can model the biological processes throughout the growth and incubation period of the plants during virus-like particle production.

Achievements/Impact: This project will lead to the creation of a database of responses that could be mined to better understand the effects of multiple parameters on the production of vaccines and other biotherapeutics leading to an increase in yield and quality. The data will be of great value to improve the process of vaccine production of Medicago R&D, that must ensure that its manufacturing technology is kept optimal and its productivity remains predictable. A successful project will provide Medicago R&D with a tool set that will enhance control and precision early in the manufacturing process, greatly improving its competitivity in the space.







































Principal Investigator:

Jacques Corbeil
CHU de Québec-
Université Laval


Marc-André D’Aoust
Medicago R&D


Ongoing Project
$ 1,137,500/ 3 years


Supported by CQDM through:

And by co-funding partner(s):
• Medicago R&D