Automated Protein Expression

Pt.2 Predictions to Proteins In-Hand in just 48hrs - Rapidly Express AlphaFold Predictions


The integration of AlphaFold AI within the eProtein Discovery framework brings forth a multitude of advantages, facilitating structure guided protein construct design with unprecedented precision.


Sign up:

How AlphaFold2 fits within Nuclera - cultivated protein design

As proteins make up 95% of drug targets, the need to acquire functional proteins to support drug discovery pipeline is crucial. Many of the easy targets have been worked on and protein scientists are often challenged with expressing difficult targets. Having access to accurate prediction of protein structure can aid in rational construct design to accelerate production of soluble, active proteins.

Nuclera is collaborating with Google Cloud, merging eProtein Discovery with Google DeepMind’s pioneering protein structure prediction tool, AlphaFold2 served on Google Cloud’s Vertex AI machine learning platform. The 3D structural insights provided by AlphaFold2 will enable Nuclera and its customers to optimize their protein variation synthesis process and gain deeper insights into the interactions between residues and the 3D folding protein structure.

Learn about how:
  • AlphaFold2 and Nuclera is creating a scalable Application Programming Interface (API) service that accesses an execution of AlphaFold2 in Google Cloud
  • An analytics dashboard which allows users to visually and quantitatively compare predicted 3D structures for protein variants
  • The protein of interest (POI) recommendation feature of AlphaFold2 proposes possible synthetic protein variants (isoforms, truncations, mutations, orthologs or fusions) using intelligent selection algorithms, taking into account various constraints such as computationally generated scores or conserved domains


Nuclera’s eProtein Discovery™ benchtop system speeds up protein expression and purification workflows in research labs, automating construct screening to inform protein scale-up in under 48 hours. Partnering with Google Cloud, Nuclera has integrated AlphaFold2 into its software, which makes it easier to access structural predictions. Direct access to AlphaFold2 results in enhanced construct quality with an in silico filter during experiment design, increasing the likelihood of identifying optimal target proteins for discovery programs.

Full-length protein expression often results in inclusion bodies, non-expression, or aggregation. To maximize yield and solubility, exploring variants of different lengths is effective. Alphafold2 can aid in the rational design of protein constructs by identifying and eliminating flexible, disordered, or aggregation-prone regions in the predicted structure.

This feature will be available to our existing customers in Q3 2024 and available publicly by Q4 2024

The length to fold a structure depends on the size of the protein ie. the number of residues it has. As an indication, if the size is about 1000 residues, you can obtain the predicted folded protein structure in about 3 hours. In the eProtein Cloud Software this prediction happens in the background and alerts you when the structure is ready for you to work with.

The eProtein Discovery system was designed to allow scientists to screen multiple options (construct variations, solubility tags and cell-free expression conditions) simulaneously to increase the chance of finding a viable option for their target on the same cartridge. Solubility tags are recognized for their ability to improve protein solubility and stability. These tags vary in size and properties, offering a range of options for researchers. In addition to solubility tags, additives in cell-free blends can also be explored. Strategies such as incorporating chaperones, creating oxidizing environments, or supplementing expression mixes with cofactors and metal ions are viable approaches to increase both yield and the proportion of soluble proteins.

Another aspect of our screening involves solubility tags. We offer a selection of seven different solubility tags alongside your target protein, which helps address challenging proteins. For instance, when folding these structures with your target protein or its variant fused with a solubility tag, such as a classic fusion protein, we observe that the solubility tags fold effectively on their own due to their well-characterized structures available in databases like PDB. This separation is facilitated by a linker that prevents significant interaction between the tag and the target protein. Our implementation of AlphaFold2 has shown that it predicts these structures reliably.

Our technology enables highly efficient reaction miniaturization, significantly reducing the amount of DNA required. This is the first benefit. As part of our product line-up, once you've designed your DNA constructs using our software, we offer a convenient service allowing you to place orders directly with our preferred DNA vendor. Currently, we utilize linear DNA for our expression templates and have successfully tested our technology with plasmids.

The innovation brought by predictive models like AlphaFold2 is transformative for downstream processes in biotechnology and pharmaceuticals. By accurately predicting protein structures, AlphaFold2 streamlines protein engineering, drug discovery, and bioprocessing. Researchers can design proteins with specific functions or optimize existing ones more efficiently. This precision reduces trial and error in protein production, enhances protein stability and solubility, and enables the design of novel therapeutics with better efficacy and fewer side effects. Ultimately, AlphaFold2 accelerates the development of new drugs and biotechnological products, making the entire process more cost-effective and sustainable.

About eProtein Discovery

Nuclera's eProtein Discovery platform combines digital microfluidic droplet automation with cell-free protein synthesis technologies, it empowers protein scientists to identify the best conditions for expressing and purifying proteins of interest within 24 hours and informs protein scale-up in under 48 hours - all on a single consumable cartridge.

The system offers a significant advantage over traditional protein expression methods, allowing researchers to save time and resources by simplifying and automating the process. Its ability to handle multiple genes and customizable cell-free blends makes it a valuable tool for protein scientists in academia and the biopharma industry.