Dyno Therapeutics Uses Machine Learning to Solve Gene Therapy’s Challenges

By Jane Cook
July 29, 2021

Dyno Therapeutics

We are living in the middle of the gene therapy revolution – and Dyno Therapeutics is capitalizing on the need for improved delivery methods for these new medicines.

Genomic Analysis

Improved genomic analysis methods combined with the breakthrough of gene editing has allowed scientists to propose novel therapeutics that would have been unimaginable even 10 years ago.

But what’s the hold up? Why have so few gene therapies hit the market so far? To date, a mere 22 cell and gene therapy treatments have been approved by the FDA despite the overwhelming promise of gene therapy.

Answer: delivery. And that’s the cool challenge being targeted by Dyno’s research team. Gene therapies are frequently delivered to target cells by adeno associated virus (AAV) vectors, viruses that can have their genetic information replaced with a gene therapy and then transfer this information to the appropriate cells.

However, AAVs circulate in nature and thus can be readily targeted by the immune system, which destroys the genetic material before it can ever reach its target cell. Researchers have spent years tweaking the sequences of AAVs to assemble new vectors that will effectively target cells, but the task has been slow and laborious.

Machine Learning

Dyno and their collaborators at Google Research and Harvard have tackled this task using machine learning. Their machine learning model published in Nature Biotechnology successfully identified AAV2 virus capsids that were distinct enough from wild viruses to evade immune detection and were structurally viable to deliver DNA to target cells.

The huge advantage of applying machine learning to this task is scale. A machine learning model like Dyno’s can assess the viability of millions of sequence variations in AAV capsids to narrow down a small set for researchers to synthesize and investigate further in the lab.

Dyno’s approach holds great promise for overcoming this barrier in gene therapy delivery and demonstrates the utility of applying machine learning to challenges in biology.

Jane Cook, Journalist & Content Writer, Bridge Informatics

Jane is a Content Writer at Bridge Informatics, a professional services firm that helps biotech customers implement advanced techniques in management and analysis of genomic data. Bridge Informatics focuses on data mining, machine learning, and various bioinformatic techniques to discover biomarkers and companion diagnostics. If you’re interested in reaching out, please email daniel.dacey@old.bridgeinformatics.com or dan.ryder@old.bridgeinformatics.com.





Recent Posts