Enzymes are having a moment
These days, there’s a lot going on in the enzyme space. Machine learning (ML) is changing the way we think about the previously hard problems of modeling protein structures and predicting protein functions. Enzyme-driven industries like pharmaceutical manufacturing, diagnostics, and industrial green chemistry are looking forward to new and better enzymes made possible by ML.
At Ginkgo, we love technologies that make biology easier to engineer. We’ve built an ML-guided enzyme engineering stack that draws on our Foundry’s established strengths in assembling smart DNA libraries, engineering organisms, and screening in (ultra) high throughput. The result is the new Ginkgo Enzyme Services offering that enables our customers to discover new enzymes and optimize existing ones. We’re super excited to share the results being delivered.
So let’s take a moment to talk about enzymes
On December 15th, Ginkgo hosted the Enzyme Intelligence Virtual Event. Dr. Emily Wrenbeck, Head of Protein Engineering, presented to an audience of more than 450 about how Ginkgo’s Foundry combines ML with massive experimental datasets to enable enzyme engineering.
The presentation was structured around 4 case studies:
- Enzyme Discovery. An ML-guided search of a large metagenomic library led Ginkgo to an unexpected corner of sequence space, producing dozens of candidate enzymes and opening up IP options for the customer.
- Better Activity. A customer’s project-critical enzyme had a reputation for being recalcitrant to engineering. Iterative protein design cycles produced a 10-fold improvement in activity over wild type as measured by kcat/KM.
- Better Specificity. A naturally sourced enzyme made 4 undesired byproducts, with the desired product only 10% of the total. Ginkgo tested over 10,000 variants with ultra high throughput GC-MS to bring the product specificity to over 75%
- Better Biocatalysts. A customer’s enzyme suffered from a low S:H (synthesis:hydrolysis) ratio. In a single round of engineering, Ginkgo was able to improve both enzyme speed and specificity under industrially relevant process conditions.
A recurring theme in the event was the value added by integrating ML tools with the broader capabilities of Ginkgo’s foundry. An engineering cycle that brings together DNA assembly, assay development, high-throughput screening, computational tools, and human experts is required to consistently deliver improved enzyme performance.
Far from replacing data, ML makes data more valuable. The massive biological datasets Ginkgo can generate were key to the success of these enzyme engineering programs.
It’s time for end-to-end Enzyme Services
Following the technical presentation, Ginkgo’s Business Development team took the stage to answer questions about Ginkgo’s latest offering. Building on powerful ML-guided enzyme engineering capabilities, Gingko Enzyme Services offers end-to-end solutions for R&D leaders looking for better enzymes.
- Novel Enzyme Discovery. Ginkgo’s large in-house metagenomic libraries are combined with ultra high throughput screening to identify new enzymes with a desired activity. These enzymes are often unrelated to previously described candidates, opening up new IP options for customers.
- Function & Stability Optimization. ML-guided protein design is combined with a range of sequence-based, structure-based, and classical enzymology tools to improve a candidate enzyme’s activity, stability, specificity or expression level.
- Robust Host Strains. Ginkgo customers have access to proprietary bacterial and fungal host strains optimized for protein production, including methanol-free Pichia pastoris and low viscosity Aspergillus niger.
- Fermentation Process Development. When it comes to commercial-scale enzyme production, Ginkgo offers automated DOE (Design of Experiments) in Ambr® 250 bioreactors to optimize the production process.
- Manufacturing Scale-Up & Tech Transfer. Ginkgo can perform pilot-scale fermentation up to 3,000 L. Customers are supported in the development of downstream processing and purification methods. Ginkgo can transfer processes to a manufacturing partner, including for GMP production.
Make the most of this moment for enzymes
A key theme of the Enzyme Services portion of the event was the flexibility of Ginkgo’s offering. As a horizontal platform, Ginkgo supports customers through the entire enzyme R&D process from discovery to development and deployment. This versatility is essential to serve the diverse needs of customers across a range of enzyme-powered industries including: animal feed, biocatalysis, diagnostics, green chemistry, metabolic engineering, mRNA manufacturing, waste remediation, and more!
A recording of the Enzyme Intelligence Virtual Event is available here. Ready to engage Ginkgo’s Foundry for your project? Have questions? Contact [email protected]
What will you grow with Ginkgo?