Gene editing has a design problem. You can build a CRISPR construct, an AAV vector, a base editor — but choosing the optimal configuration for a given genetic target still involves enormous amounts of trial and error. For diseases driven by repeat expansions — where stretches of DNA abnormally replicate themselves, causing conditions like ALS, Huntington’s, and myotonic dystrophy — the design space is even more complex. The editing tool needs to find and cut at two precise locations flanking the expansion, and the parameters that determine success or failure are poorly understood.
Specific Biologics, a Toronto-based precision genome editing company, just received approximately CAD$1.8 million from Genome Canada and Ontario Genomics to build a machine learning-enabled prediction platform for the design and optimization of therapeutics built on its proprietary Dualase® technology. The collaboration with Western University — where the Dualase technology was originally discovered in Dr. David Edgell’s lab — is designed to turn what is currently an artisanal process into a computationally optimized one.
The Biotech Voyager
Early-stage biotech signals, personalized.
The signals that matter to you — contextualized and written directly to you — so you cut through the noise and immediately understand why it matters.
Dualase is an interesting platform in its own right — it’s an in vivo genome editing system that uses a two-site DNA editing mechanism, meaning it cuts at two locations simultaneously to excise the pathogenic repeat expansion. The lead preclinical program targets C9ORF72-associated ALS, the most common genetic cause of amyotrophic lateral sclerosis. The ML layer is intended to accelerate candidate design by predicting which Dualase configurations will be most effective for a given repeat expansion target — reducing the experimental cycles needed to optimize each new therapeutic.
The company recently added Dr. Richard Glickman to its board — the biotech veteran who founded Aspreva Pharmaceuticals and led it through a $915 million acquisition. That kind of board-level addition at a preclinical company signals preparation for the next stage of development.
This is a quiet signal, but it’s the kind of infrastructure investment that compounds over time. If the ML platform works, it doesn’t just optimize one program — it accelerates every future Dualase program against any repeat expansion disease. ALS today, Huntington’s and myotonic dystrophy tomorrow. The Canadian government is betting that the intersection of ML and gene editing design is worth seeding early. Given the complexity of the design problem and the severity of the diseases involved, that seems like a smart bet.
