Cube Biotech and IBA Lifesciences join forces - Read the full press release here
Go to homepage
Show all categories Back

AI-driven drug discovery

Artificial intelligence (AI) has become a valuable tool in drug discovery. It supports various stages of the process, from initial target identification to drug design, prediction of molecule properties, as well as validation and optimization. 

The ability to integrate huge amounts of data and learn from previous results vastly accelerates the whole process, reducing costs and improving accuracy. The speed at which novel compounds are predicted and screened potentially leads to a higher chance of successful clinical implementation. 

One approach to effectively discover and optimize novel drugs is a combination of machine learning (ML), a subset of AI, and synthetic biology, referred to as generative biology. In this approach, AI designed molecules are expressed and characterized in wet lab experiments. The results are subsequently used to train and validate the machine learning models. Via these design-make-test-learn (DMTL) cycles, potential candidates can be discovered and iteratively improved to finally fulfill all necessary characteristics. A prerequisite for this and other AI-based approaches are high quality input data derived from wet lab experiments during the “make & test” cycle phases to ensure that outcomes are as accurate as possible.


For effectively combining AI with synthetic biology, a system is needed that can keep up with the amount of predictions made by machine learning models and is suitable for automated procedures. 

One system that fulfills these requirements is the Strep-tag® technology. This technology offers features that can greatly contribute to quickly generate high quality data required for AI/ML-driven drug discovery approaches and streamline the whole workflow: 

  • Simple & fast one step purification suitable for automated systems
  • Outstanding target purity of >95%
  • Specific binding & specific elution 
  • Superior affinity (pM) between Twin-Strep-tag® (affinity tag) and Strep-Tactin®XT (ligand)

Due to the outstanding target purity after automated one-step purification, a direct transition to downstream applications is possible. This enables the screening of >2000 candidates per week, significantly speeding up the drug discovery process.

The Strep-tag® in AI-driven drug discovery

High-throughput screening for AI-based drug discovery

"Make" - Expression & purification

Screening AI-predicted molecules for the desired characteristics is key for identifying the most promising drug candidates.

Ideally, all required data should be generated in a short time, while its quality remains unaffected. A first step in this screening process is the expression and purification of predicted proteins, for example antibodies or antibody fragments. A cell-free expression system can be advantageous here, due to its ease of use and rapid protein production. Subsequently, a robust automated purification system is needed that reliably delivers highly pure proteins. In combination with a suitable affinity tag, magnetic beads are a good choice in this case, since they can be used in robotic systems, are suitable for small scale purifications and generate consistent results.

Especially MagStrep® Strep-Tactin®XT beads, which specifically bind the Twin-Strep-tag®, are an ideal tool for candidate screening in AI-driven drug discovery due to the following features: 

  • Specific binding and elution
  • Fast binding kinetics
  • Fast processing of only 39 minutes
  • Reliable one-step purification with outstanding purity
  • Compatible with almost all magnetic systems
  • Compatible with a vast array of additives

Automated protein purification with MagStrep® Strep-Tactin®XT beads - consistently high recovery & purity

Figure 1. Automation and parallel protein purification with MagStrep® Strep-Tactin®XT beads were tested using a Tecan Fluent® (A). αCD45 nanobody fused to a Twin-Strep-tag® was purified in parallel from 88 samples. Consistently high recovery and purity was achieved across all samples.

"Test" - Analysis

After successful purification of drug candidates, further analysis steps are required to determine the specific properties of a protein. The use of the same affinity tag as used for purification can be advantageous, since the amount of required tags is reduced, keeping the impact on the target protein as minimal as possible. Due to the picomolar affinity of the Twin-Strep-tag® to its ligand Strep-Tactin®XT, this system not only enables efficient purification with high yield and purity, but also further downstream analysis such as ELISA  or western blot. Due to the effective immobilization of tagged protein, even kinetic measurements via e.g. SPR or BLI are possible. This way, purified proteins are easily transitioned to further analysis steps and no change in affinity tag system is required.

SPR measurement using a Twin-Strep-tag® protein bound to Strep-Tactin®XT – effective immobilization for affinity measurements

Figure 2: SPR chips coated with Strep-Tactin®XT were used for immobilizing a CD45 nanobody tagged with Twin-Strep-tag®. The results show that with Strep-Tactin®XT the nanobody is bound stably over a long-period of time. These data demonstrate that Strep-Tactin®XT can be used to analyze high-affinity interactions with slow dissociation rates.

This website uses cookies to ensure the best experience possible. More information...