e University of Washington’s Institute for Protein Design is on a warm trajectory. The study hub’s rapid development of new research, businesses, and commands is being fueled by advancements in artificial intelligence.
The most recent endeavor based on IPD systems was announced on Tuesday: Xaira, a business in the San Francisco area co-founded by IPD head David Baker, with backing from investors including Arch Venture Partners and Foresite Capital.
RoseTTAFold All-Atom, the IPD’s next-generation protein framework forecast and layout application, was just four days prior to that event on the cover of the journal Science.
And on Thursday, the IPD published an independent study in Science on designing specialized peptides, modern ultra-small proteins that form the basis for Vilya, one of ten IPD spinouts.
IoT-driven protein pattern is used to produce novel therapeutics, vaccines, biosensors, materials, and more. The industry is moving quickly, and IPD analysis is behind many of the improvements.
The rate of IPD study took off in 2021 with the launch of RoseTTAFold, the university’s receptive-access rival to DeepMind’s protein structure prediction tool AlphaFold. Both won Science Magazine’s Breakthrough of the Year award.
Since then, the IPD has made dozens of publications, as well as an open-access library of tools designed to create proteins of all shapes and sizes, from nanopores to antibody targets.
Xaira, which has offices in Seattle and already employs several people there, is being built using the IPD’s generative AI tools, such as RFdiffusion and RFantibody. The business hopes to use machine learning and large-scale data generation to create new models and treatments.
In addition to fostering new startups and producing a large number of proteins with an array of shapes and capabilities, the following are a list of other important IPD studies from the last ten months is a list.
- Introducing RFantibody, used to create designs for antibodies, powerful molecules that can recognize specific therapeutic targets. Released March 18 as a preprint, which has yet to be peer-reviewed.
- Using a new toolkit to simplify the creation of biomaterials. Published March 13 in Nature.
- Creating hydrogels to combine proteins in three-dimensional scaffolds inside or outside of cells. Published Jan. 30 in the Proceedings of the National Academy of Sciences.
- Designing proteins that bind efficiently to biomarkers including human hormones, by deploying RFdiffusion and another IPD tool, ProteinMPNN. Published Dec. 18 in Nature.
- Producing carbon-rich minerals that have the potential to serve as the carbon storage foundation through engineered organisms. Published Dec. 14 in Nature Communications.
- Creating protein crystals, setting the stage for further development that might lead to the development of new optical tools, chemical separation techniques, and other applications. Published Oct. 16 in Nature Materials.
- Designing protein fibers akin to those in silk, wool, and spider webs, opening the door to new textiles and bioengineering applications. Published Sept. 4 in Nature Chemistry.
- Developing switch-like proteins that can be used as environmental sensors or creative therapeutics when they change between two shapes. Published Aug. 17 in Science.
- Introducing RFdiffusion, which begins with an image of “pure static” that coalesces into an image of a protein. Published July 11 in Nature.
These studies were all released last year and utilize recent 2022 and 2023 research. This earlier research includes designs for custom enzymes, proteins that can slip through cell membranes, and a study showing how reinforcement learning, used in board games like chess, can support the design of new molecules. For more, see posts by the IPD and the Baker Lab.
