Basecamp Research, a global leader in AI-driven protein and biological system design, has partnered with the Ferruz Laboratory at the Institute of Molecular Biology of Barcelona to unveil ZymCTRL ("enzyme control"). Modeled after large language models (LLMs) like ChatGPT, ZymCTRL allows users to generate new enzyme sequences simply by inputting an enzyme identification code, specifying the desired activity.

Traditional protein design models have required extensive training and conditioning on known protein starter sequences. In contrast, ZymCTRL represents a significant leap forward, offering an end-to-end solution that bypasses the need for a seed sequence, thus providing users with unparalleled control over the design process. This next-generation tool is not only rapid and cost-effective but also capable of creating enzyme sequences with only 30% resemblance to those in its training set, broadening the horizons for novel enzyme development.

"With ZymCtrl, generating highly specific enzymes is as easy as interacting with a chatbot," said Noelia Ferruz, whose lab has been collaborating with Basecamp Research for over two years. The Ferruz Lab is renowned for its pioneering work in AI for protein design, having previously developed ProtGPT2, a deep unsupervised language model for protein design.

"Even before the release of ChatGPT,  we began working on large language models with Noelia because we think these models represent the future of biological research and protein design," said Dr. Philipp Lorenz, CTO of Basecamp Research. "We're deeply excited by these results and ZymCTRL's ability to create functional enzymes that can solve some of today's biggest challenges, from finding new ways to treat devastating diseases to building greener and more sustainable catalytic processes in bioindustry."

The open-source ZymCTRL model has undergone independent review by academics and has been featured in leading scientific journals, including Structural Biology and ChemBioChem. Researchers at The Institute of Biochemistry at Austria's Graz University of Technology praised ZymCTRL's efficiency and user-friendliness. "ZymCTRL designs putative enzyme variants on consumer GPUs within seconds and, remarkably, it creates these sequences with only an EC number as input," wrote Horst Lechner, principal investigator at the institute.

Basecamp Research is making ZymCTRL available as an open source to researchers worldwide, envisioning a wide range of applications, from disease treatment and diagnostics to biofuel production and sustainable agriculture innovations. While initially trained on publicly available datasets, ZymCTRL can be further optimized with additional data, including Basecamp Research's proprietary BaseGraph database, enhancing its sequence output capabilities.

The release of ZymCTRL marks a significant milestone in the field of AI-driven enzyme design, promising to accelerate advancements in biotechnology and offering innovative solutions for global challenges.

‍