A-Alpha Bio The company announced on Thursday that it will build a laboratory space in downtown Seattle and provide its machine learning team with $20 million in new venture capital for its business to identify therapeutic proteins. Seattle’s Madrona Venture Group, a technology company focused on life sciences, led the Series A financing.
This funding promotes the company’s $2.8 million seed funding from the University of Washington spin-off project to develop into a mature biotechnology that supports the development of potential treatments for COVID-19 and other diseases.
The more protein a company analyzes, the better its performance. A-Alpha Bio uses information from its data set to train its algorithms to predict favorable protein interactions.
“Over time, we are accumulating these huge protein interaction databases, and we can also use this data in conjunction with computational tools such as machine learning to better understand and design protein interactions,” said the CEO David Young In an interview with GeekWire.
Younger co-founded the company with CTO in 2017 Randolph Lopez Use the technology that their graduate researchers at the Institute of Protein Design at the University of Washington helped develop.
The drug development platform company combines computational tools with yeast experiments to determine potential therapeutic proteins. It works with pharmaceutical companies to help them find the best drugs to test under a range of conditions.
The technology relies on single-cell yeast pools designed to express various proteins or protein fragments. When two proteins interact, yeast fuse and count the interacting proteins. This method can recognize protein interactions, such as antibodies that adhere tightly to viral proteins.
Some existing technologies can more accurately measure the strength of protein interactions, but only at the level of a single protein. There are other methods that can detect a large number of protein interactions, but the resolution is very low.
Younger said that A-Alpha Bio has reached the best position for high precision and high throughput.
“Historically, this kind of very dramatic trade-off does exist, either to get high quality or to get a lot, and it is difficult to maximize both,” he said.
A-Alpha Bio has established partnerships with three major pharmaceutical companies and is seeking more cooperation. It also collaborates with the Bill and Melinda Gates Foundation, Twist Biopharma, and Seattle-based Lumen Biosciences to help develop projects such as COVID-19 antibodies. Lumen grows therapeutic proteins in the algae Spirulina and recently announced plans to build a new production plant in the former Seattle bakery.
Younger said that A-Alpha Bio’s high precision and throughput give it an advantage among competing platform technology companies. Other companies in this field include Adimab and Distributed Bio.
Younger and his colleagues recently A study Identify antibodies against variants of the COVID-19 virus. This antibody may be used as a therapy for the disease.
The researchers started with yeast expressing various antibodies and proteins from different virus variants. Younger said: “We put all of this in a jar, draw all the interaction diagrams, and then determine which antibodies will have the correct characteristics.” The paper appeared on the preprint server BioRxiv and has not yet been peer-reviewed.
In addition to recognizing antibodies, A-Alpha Bio also focuses on finding “molecular glue” targets. This “glue” connects the target protein with another type of protein and tells the cell to eliminate the target.
“It’s essentially a method of degrading specific proteins,” Younger said. It is also the basis for the chemotherapy drug Revlimid. Revlimid has brought billions of dollars in revenue to Celgene, a company of Bristol Myers Squibb, and has sparked a strong interest in the field of molecular “glue”.
A-Alpha Bio also uses calculation methods developed by IPD to predict how proteins fold. IPD’s tool, called Rose TTAFold, a similar tool developed by DeepMind recently shocked the life science community with its speed and accuracy in predicting protein folding.
The existing large number of known protein folding data sets make these tools possible, and the training set for predicting protein-protein interactions is still limited. But A-Alpha Bio believes that this situation is changing as it builds a database.
Younger said that A-Alpha Bio can measure millions of protein interactions in a single experiment. “So with this data, we can start training machine learning models, which can basically learn the rules of protein-protein binding. He added: “Our vision is to reach a point where computer antibody and antibody predictions can basically be made for a given unprecedented target.”
The combination of laboratory science and computational methods is an emerging investment area for the Managing Director of Madrona Matt McElvin Tell GeekWire before.
This decades-old venture capital company traditionally invests in technology companies, but increasingly supports companies at the “intersection of innovation.” These companies between biology, computer and data science also include Nautilus Biotechnology, TwinStrand Biosciences, Ozette and Terray Therapeutics.
“As machine learning, computing power, and wet laboratory automation advancements merge, we expect more companies to solve research problems in a paradigm shift,” McWayne and Madrona partners Chris Picardo Said in a blog post about the new funding.
A-Alpha Bio currently has 13 employees and is expected to expand to about 50 in the next two years, while establishing its machine learning team. Before the pandemic, “we strongly believed in keeping the team as local as possible,” said Younger.
But like many operations, A-Alpha Bio has learned to operate remotely and effectively, and is willing to accept remote recruitment. Younger said: “We recently learned that we don’t want to sacrifice talent under any circumstances.”
The company also transformed the existing office space near Belltown into its new laboratory for yeast experiments, which are organisms that do not require a lot of maintenance.
“One of the real benefits of using yeast as a model organism is that from an infrastructure perspective, our needs are very simple,” Younger said, “without having to fight other biotech companies for the 1% vacancy. In Seattle We can move into office space and fully convert it into a laboratory that meets all our needs.”
After the recent Icosavax IPO, the company is the latest IPD spin-off company that develops vaccines for COVID-19 and other diseases.
The previous seed funding was led by OS Fund, and the company also received funding from the Gates Foundation and the National Science Foundation. In addition to Madrona, participants in the Series A financing also include Perceptive Advisors and Lux Capital. McIlwain, who is also a member of the board of directors of the Fred Hutchinson Cancer Research Center, has joined the company’s advisory board.