As genetic medicine progresses under the leadership of more and more biotechnology companies, drug discovery and development are advancing by leaps and bounds. According to Brendan Frey, the founder and CEO of Deep Genomics, the 20th century medicine revolves around small molecule chemistry, and the 21st century medicine will be based on information: the sequence of letters in the genetic code that determines health and wellness. disease.
Deep Genomics uses artificial intelligence to analyze genetic data. The Toronto-based company’s technology is providing a pipeline for internal and collaborative drug candidates that are advancing clinical testing. Frey is now setting ambitious goals, even if Deep Genomics becomes a global leader in the production and sale of genetic drugs.
“Drugs are information now,” Frey said. “How do we turn complex biology into a method of designing new drugs? In order to dominate the entire field, we need a way to transform this complex biology into this information medicine.”
On Wednesday, the company announced $180 million in financing Bring it closer to its goal. The C round of financing was led by SoftBank Group, and its Vision Fund 2 investment.
Deep Genomics focuses on RNA, the molecule that carries the instructions to make proteins. Frey continues to compare biology with information, comparing the process of developing Deep Genomics drugs to writing patches to fix software bugs. The company’s proprietary technology is called AI Workbench, which applies artificial intelligence and machine learning technology to discover the targets of its drugs, and then design therapies that can solve these targets.
Frey said you can think of Deep Genomics drugs as genetic software patches. Drugs are oligonucleotides, which are short chains of nucleic acids. These gene drugs are steric blockers. In other words, they bind to RNA and prevent this part from being independent of the cell’s protein manufacturing mechanism. Frey says that doing so will change the way the cell reads RNA, which in turn will correct the mutation. For haploid deficiency characterized by insufficient levels of important proteins, the patch can be programmed to increase protein yield by RNA. If the mutation results in a bad protein, genetic patches can reduce protein production or ensure that the correct protein is produced. An example of a commercial steric blocking oligonucleotide is Spinraza, a Biogen drug approved for the treatment of spinal muscular atrophy.
2019, deep genomics Unveil Its first drug discovered by AI, an experimental drug for treatment Wilson disease. This rare genetic disease is caused by a genetic mutation that prevents the liver from excreting excess copper. As a result, copper accumulates in the liver and other parts of the body, leading to complications and in some cases even death.
Since then, the Deep Genomics pipeline has been expanded to nine additional programs. Frey’s goal is to bring four into the clinic by 2023-all of which were discovered and developed by AI Workbench. Some drugs are for rare diseases, but Frey said the technology can also be applied to more common diseases; the company’s projects include refractory gout and Parkinson’s disease. Frey said that with new financing, Deep Genomics’ goal is to quickly expand to 30 projects.
Deep Genomics is also seeking partners with large biopharmaceutical companies.This startup currently has a partnership, a Alliance with BioMarin Pharmaceutical That started in November last year. The four projects in the partnership are addressing undisclosed goals, but Frey stated that the alliance includes disease goals for Deep Genomics brought by the San Rafael, California-based company and the startup’s pipeline. project.
Frey said the reason for finding a partner is that the productivity of AI Workbench is so high that it generates more drug opportunities than Deep Genomics itself can pursue. The partners bring capabilities that Deep Genomics does not have, such as expertise in specific therapeutic areas or experience in running international clinical trials. Artificial intelligence also facilitates cooperative decision-making. Analysis will not eliminate all risks, but it will tell you where these risks are. For example, some projects may require longer clinical trials to prove efficacy, while other projects may show their value in early studies. Frey said analyzing these risks helps Deep Genomics make reasonable decisions about who is best to develop a particular program.
Deep Genomics has a lot of competition in the field of gene medicine, but Frey pointed out a feature that distinguishes his company’s drugs from gene therapy and gene editing drugs. Unlike therapies that cause permanent changes, Deep Genomics’ genetic patches can be removed. In this way, if a serious problem occurs, treatment can be stopped to reverse the adverse reaction.
Another difference is the value created by the Deep Genomics platform. Frey predicts that the CRISPR patent is now 10 years old, and the technology will eventually become a commodity, thereby reducing its value. Because Deep Genomics technology learns and improves every time it is used, he said its value will increase over time. However, CRISPR editing drugs and gene therapy are not the only competitors of Deep Genomics.Other startups RNA platforms that provide functions for Deep Genomics technology include Strand Therapeutics and CAMP4 therapy.
Before announcing the financing on Wednesday, Deep Genomics’ most recent financing was in 2020. US$40 million in Series B financing Led by Future Venture Capital Corporation. The company’s latest financing includes earlier investors Fidelity Management & Research Company, Canadian Pension Plan Investment Commission, True Ventures, Amplitude Ventures, Khosla Ventures and Magnetic Ventures. Frey said that Deep Genomics is looking for investors with long-term investment prospects. He added that “when the time is right,” an initial public offering may take place. This time depends in part on the progress of the Deep Genomics pipeline. Frey said that the company will not go public until there are four clinical programs that demonstrate the capabilities of AI Workbench.
Photo courtesy of Business Wire
