Protein design using artificial intelligence has the potential to transform various fields from drug discovery to industrial material development. However, design alone has its limits. All inventions must also be tested in the real world. Enter Adaptyv Bio, a Swiss start-up. Adaptyv Bio hopes its technology will solve a potential blockage in the innovation pipeline.
Julian Englert, CEO and co-founder of Adaptyv Bio, said, “Proteins are at the heart of the biorevolution in the form of new drugs, better enzymes for research and industrial applications, or materials with new properties. And advances in AI solutions such as DeepMind’s AlphaFold are revolutionary in this regard, allowing researchers to automate the design of new proteins.
The problem, says Englert, is that each new design requires trial and error. “We have become very accustomed to AI tools that generate content such as text, but with such output we can quickly see if it works or if it is gibberish. ,” he explains. “You can’t tell by looking at an AI-designed protein whether it will do what you want it to do. It needs to be tested in the lab.”
That’s where Adaptyv Bio comes in. The company’s lab, his Englert, describes the facility as a “full-stack protein engineering foundry,” allowing designers to test new proteins quickly and at scale. They can submit hundreds or thousands of designs to Adaptyv Bio. Adaptyv Bio uses robotics, microfluidics, synthetic biology and other technologies to test in labs to identify which ones work best. You can repeat this process over and over again until the design is exactly what you need. The data generated with each new test can be used to fine-tune the design and improve the performance of subsequent AI-powered protein designs.
Englert believes Foundry can provide breakthrough moments for scientists and democratize the protein engineering industry. Scientists who do not currently have access to their own labs (primarily the reserves of the largest pharmaceutical companies and industrial conglomerates) are forced to wait months to get test results for each new design, at considerable cost. .
Englert compares his company’s value proposition to the software development department. “Imagine if every time we used Github Copilot to generate code, he had to wait 10 weeks for the code to run or notify us that there was a bug,” he said. increase. “And imagine each execution costs him $1,000. That’s the situation with protein designers today.”
Adaptyv Bio was launched about two years ago. This is because the company has worked with many clients in the biotech sector to fine-tune the technology at its Lausanne-based foundry. “We are now ready to start talking about it because the technology is working so well in an automated state,” he adds Englert.
The company sees the drug discovery sector as its primary target market, at least initially. In particular, smaller biotech companies, including “virtual biotechs” that have raised research funding but do not have their own facilities, can use Adaptyv Bio’s foundry to test the real-world effectiveness of their protein designs. increase.
However, protein design has applications in multiple fields, and scientists are constantly working to develop new materials with specific properties. Industrial applications in the environmental sector are just one of the potential areas of interest.
“Let’s look at the proteins that make up the incredibly powerful molecular machinery in every cell of our body,” says Englert. “What technological advances could humanity make if we could start designing new proteins for industrial applications such as personalized medicines, new enzymes, or better, more sustainable materials? Please try to imagine.”
Englert and his co-founders met at the Swiss Federal Institute of Technology in Lausanne to found Adaptyv Bio and grow the business through the Y Combinator accelerator program. He then raised $2.5 million in funding from Wingman Ventures.
Follow me please twitter.
