TeselaGen Biotechnology announced today a collaboration with the Agile BioFoundry (ABF) to lay the groundwork for deploying machine learning models on community-generated data to accelerate biomanufacturing and commercialization of bioproducts.
Project aims to develop an industry-wide interoperability standard for the easy exchange of fermentation process data to accelerate the commercialization of bioproducts.
TeselaGen Biotechnology announced today a collaboration with the Agile BioFoundry (ABF) to lay the groundwork for deploying machine learning models on community-generated data to accelerate biomanufacturing and commercialization of bioproducts. Under terms of the collaborative research and development agreement, funded by the U.S. Department of Energy (DOE), TeselaGen will assist in the creation of a set of interoperable, data acquisition and sharing tools to enable machine learning in the fermentation process. The development of Phycus Biotechnology’s sustainable bioproduct alternatives to petroleum-based chemicals used in cosmetics will be used as a real-world test case.
“We’re excited to work with the Advanced Biofuels and Bioproducts Process Development Unit (ABPDU) of Lawrence Berkeley National Laboratory, and Phycus Biotechnology, as part of this Agile BioFoundry project to develop a cloud-based commercial fermentation operating system that will modernize data and workflow management and incorporate machine learning to optimize production,” said TeselaGen CEO Eduardo Abeliuk, PhD. “As an extension of our partnership with the ABPDU, this collaboration will also further help to advance the technological readiness of our software and machine learning solutions for accelerating and optimizing fermentation processes.”
The development of advanced bioinformatics tools and machine learning techniques have enabled the rapid expansion of new, high-quality bioproduct candidates, but downstream fermentation studies and strain optimization workflows have become a bottleneck in the development cycle. High-throughput bioreactor services and larger scale-up services alleviate some of the strain and incorporate various software and hardware to manage the data generated. Despite this, data from disparate sources does not easily integrate and is not machine learning ready, increasing the time it takes to develop and commercialize products.
“Use of machine learning-enabled strain engineering & bioprocess development holds promise to rapidly advance bioengineering, but this won’t happen unless data are captured and integrated in a systematic and standardized fashion,” said James Gardner, program manager in the Agile BioFoundry. “This project offers a unique window into how a modern, scalable infrastructure can aid early-stage companies, and simultaneously demonstrate its utility with real-world data sets. As such, this collaboration has the potential to promote widespread adoption of interoperability standards in industry.”
This is one of six projects totaling over $5 million announced last month by DOE to conduct research and development needed to accelerate the U.S. biomanufacturing sector. The DOE Bioenergy Technologies Office (BETO) Agile BioFoundry is a consortium of seven national laboratories and more than a dozen university and industry partners. Through the Agile BioFoundry, these projects will leverage national laboratory capabilities to address early-stage challenges in biomanufacturing.
For more such updates and perspectives around Digital Innovation, IoT, Data Infrastructure, AI & Cybsercurity, go to AI-Techpark.com.
