Artificial intelligence approach diversifies AAV vectors for gene therapy
Adeno-associated viruses (AAVs) are widely regarded as a promising means of delivering gene therapies to defective tissues in the human body, yet there is still a way to go before they can meet their full potential as effective gene delivery products. To this end, researchers at Dyno Therapeutics have collaborated with Google Research, Harvard’s Wyss Institute and Harvard Medical School (all MA, USA) to employ a machine-learning approach that generates a diverse range of AAV capsids.
Building on this work, researchers have applied a computational deep-learning approach to design capsid variants from the AAV2 serotype across DNA sequences encoding a protein segment that plays a role in immune recognition as well as targeting specific tissues.
The team used multiple machine-learning tools to design and produce a vast library of distinct AAV capsid variants, more than 57,000 AAV of which exhibited much higher diversity than naturally occurring AAV serotypes.
“Our approach achieves the highest functional diversity of any capsid library thus far. It unlocks vast areas of functional but previously unreachable sequence space, with many potential applications for generating improved viral vectors, like AAVs with much reduced immunogenicity and much improved target tissue selectivity, and also for highly efficient gene therapies,” explained last-author Eric Kelsic (Dyno Therapeutics).
This work is set to advance the sophistication and efficacy of AAVs in delivering products such as gene therapies, and gives a glimpse into the future of artificial intelligence use in the design of new drugs as well as drug-delivery approaches.
“The more we change the AAV vector from how it looks naturally, the more likely we are to overcome the problem of pre-existing immunity,” commented Sam Sinai, Dyno co-founder and Machine Learning Team Lead. “Key to solving this problem, however, is also ensuring that capsid variants remain viable for packaging the DNA payload. With conventional methods, this diversification is time- and resource-intensive, and results in a very low yield of viable capsids. In contrast, our approach allows us to rapidly unlock the full potential diversity of AAV capsids to develop improved gene therapies for a much larger number of patients.”
Link: https://www.regmednet.com/artificial-intelligence-approach-diversifies-aav-vectors-for-gene-therapy/
ارسال به دوستان