Enveda Biosciences Announces Microsoft Collaboration and Reveals Foundation Model to “Read and Translate” Chemistry
BOULDER, Colo. – Enveda Biosciences (“Enveda”), a biotechnology company using AI to translate nature into new medicines, announced today a new collaboration with Microsoft and released details on their work creating their Foundation Model, called PRISM, trained on the world’s largest collection of experimental mass spectra. The announced initiatives support Enveda’s continued leadership in natural product drug discovery and its evolution into a clinical-stage company.
“By combining Microsoft Azure infrastructure with Enveda’s data, expertise, and vision, an entirely new realm of chemistry can be explored to identify new medicines.”
The collaboration with Microsoft includes access to compute resources needed to power Enveda’s novel drug discovery platform. Enveda will use Microsoft Azure to scale its generative AI models, including the newly revealed model, PRISM.
Built with Azure, PRISM (Pretrained Representations Informed by Spectral Masking) represents a step-change in the ability to use AI to interpret mass spectrometry data. Mass spectrometry is the main method used to detect small molecules in the natural world. By training PRISM on over a billion unlabeled mass spectra, the model has learned fundamentals about the language and grammar of this data type, essentially enabling the model to “read” life’s chemistry. PRISM can then be used to power downstream predictions, such as de novo chemical structure predictions by “translating” mass spectra into representations of chemical structure or properties, underlying Enveda’s ability to source new medicines from nature.
“Given their leadership in advancing and implementing AI across industries, we are thrilled to collaborate with Microsoft,” says Viswa Colluru, Ph.D., CEO and Founder of Enveda. “We believe this relationship and our new foundation model will further accelerate our ability to find powerful new medicines from natural sources.”
“Large foundation models trained on massive sets of raw data are the basis for modern AI applications. By training PRISM on the largest collection of mass spectra ever assembled, we’ve created a tool that can help us interpret the language of chemistry and to translate it into new, impactful medicines,” says David Healey, Ph.D., Vice President of Data Science at Enveda.
“The intersection of AI, science, and drug discovery presents a frontier of possibilities,” said Christopher Young, EVP of Business Development, Strategy and Ventures at Microsoft. “By combining Microsoft Azure infrastructure with Enveda’s data, expertise, and vision, an entirely new realm of chemistry can be explored to identify new medicines.”
About Enveda Biosciences
Founded in 2019, Enveda is a biotechnology company aiming to understand and utilize the chemistry that forms the foundation of life. Using their AI-powered tools to identify and characterize the wide range of molecules that are produced by living organisms – the vast majority of which have never been explored by science – Enveda is creating a new database of chemical biodiversity: the library of life. By growing, organizing, translating, and searching this unique library, Enveda learns from life’s evolved solutions to address today’s pressing clinical needs. For more information on Enveda, visit enveda.com.
