Science & Enterprise subscription

Follow us on Twitter

  • Results from a clinical trial show a combination of two synthetic antibodies reduces deaths in hospitalized Covid-1… https://t.co/AtztCHfZLl
    about 9 hours ago
  • New post on Science and Enterprise: Antibodies Help Some Hospitalized Covid-19 Patients https://t.co/lJhns0fZzJ #Science #Business
    about 9 hours ago
  • https://t.co/BUeUy0ajO3 ... Crocodile on the Tarcoles River in Costa Rica, for #NaturePhotographyDay ... February 2020.
    about 1 day ago
  • https://t.co/mtU98giel5 Oystercatcher for #NaturePhotographyDay ... Cape May, New Jersey. June 2019.
    about 1 day ago
  • Two biotechnology companies are designing therapies to deliver edited genes to treat Friedreich’s ataxia and amyotr… https://t.co/7tbDhyf5lv
    about 1 day ago

Please share Science & Enterprise

AI Gene Therapy Platform Gains $100M in Early Funds

Adeno-associated virus

Adeno-associated virus (Jazzlw, Wikimedia Commons)

6 May 2021. A company using machine learning to create engineered viruses to deliver gene therapies is raising $100 million in its first venture funding round. Dyno Therapeutics Inc., a three year-old company in Cambridge, Massachusetts, is spun-off from the lab of Harvard University geneticist and serial entrepreneur George Church.

Dyno Therapeutics aims to to produce better techniques for delivering gene therapies that often rely on viruses. The adeno-associated virus or AAV is the most commonly used of these viruses, a benign and naturally occurring microbe that infects cells, but does not integrate with the cell’s genome or cause disease, other than at most mild reactions. As a result, AAVs have become a workhorse for delivering modified genes to treat inherited diseases, including therapies approved by FDA. In their natural state, however, AAVs are imperfect and inefficient delivery vehicles, which company seeks to improve.

Research by Church and colleagues, reported by Science & Enterprise in November 2019, revealed techniques using machine learning to systematically understand the structure of a key part of AAVs, their outer shell called the capsid. The study, published in the journal Science, found 735 amino acids making up the capsid’s proteins. Eric Kelsic, a postdoctoral researcher in Church’s lab at the time, led the team, which genetically sequenced and uniquely identified some 200,000 variations of those protein components, and uncovered a previously unknown protein that helps bind AAVs to their target cells.

Kelsic, Church, and others founded Dyno Therapeutics in May 2018, where Kelsic is now CEO. The company extends the academic lab’s work in a technology called CapsidMap that uses machine learning algorithms to design optimized AAV capsids. The algorithms find millions of optimal combinations of targeting ability, payload size, immune evasion, and manufacturing capability, then give each variation a unique DNA identifier. With machine learning, optimized capsids are assembled to meet specific therapeutic needs, with each design adding to and refining the algorithms’ experience.

Three licensing and collaboration deals

“Dyno was the first to combine machine learning with data from high-throughput in vivo experiments,” says Kelsic in a company statement, “to optimize and accelerate the design of improved capsids for gene therapy. Our CapsidMap platform brings unprecedented scale and technical sophistication to solving in vivo delivery, the key challenge for gene therapy, making therapies more effective, safe, manufacturable, and capable of benefiting more patients.”

Since its founding, Dyno Therapeutics landed three collaboration and licensing deals for its CapsidMap technology with pharmaceutical and biotechnology companies. In October 2020, Dyno agreed to license its process to drug maker Roche and its subsidiary Spark Therapeutics to design new AAV capsids that improve gene therapy delivery for treatments affecting the central nervous system and liver. In May 2020, Sarepta Therapeutics gained an option to license Dyno’s technology for rare inherited muscle diseases. Also in May, Dyno agreed to collaborate with drug maker Novartis on using CapsidMap to discover and design synthetic viruses to deliver gene therapies for eye diseases.

Dyno Therapeutics is raising $100 million in its first venture funding round. The financing is led by technology venture investor Andreessen Horowitz in New York. Taking part in the round are new investors Casdin Capital, GV (formerly Google Ventures), Obvious Ventures, and Lux Capital, with current investors Polaris Partners, CRV, and KdT Ventures. Dyno plans to apply the new round’s proceeds to expand CapsidMap into treatments for lung, heart and kidney diseases, as well as expand its staff. The company raised $9 million in seed financing in May 2020.

More from Science & Enterprise:

*     *     *

1 comment to AI Gene Therapy Platform Gains $100M in Early Funds