Coupling machine learning and in vivo screening for identifying novel therapeutics for Spinal Muscular Atrophy

Therapeutic development is a long and laborious process, very expensive and risky. The high costs are linked to the high risk of failure and there is therefore a need to develop new tools and new approaches that can accelerate and de-risk these processes.
The development of digital technologies brings enormous potential to traditional therapeutic development processes. In particular, deep learning opens up an unprecedented possibility for quickly and efficiently identifying compounds with high therapeutic potential. Combined with high throughput preclinical in vivo testing, the potential of these digital approaches can be harnessed and thus offer a new way to identify therapeutic candidates from an unlimited repertoire of molecules.
By combining the expertise of two emerging Quebec companies, Valence Discovery and Modelis, this project will identify small candidate molecules against a rare neuromuscular disease: spinal muscular atrophy. This proof of concept could be extended
to the entire portfolio of each company and thus further increase Quebec’s competitiveness in the field of health sciences.

Faculty Supervisor:

Éric Samarut

Student:

Partner:

Valence Discovery Inc;Modelis inc.

Discipline:

Life Sciences

Sector:

Professional, scientific and technical services

University:

Université de Montréal

Program: