How AI and Automation Are Reshaping Protein Design
AI and automation are accelerating protein design, reshaping how candidates are screened, analyzed, and optimized.
The rapid rise of AI has begun to reshape how proteins are designed, optimized, and analyzed. From structure prediction to screening and downstream characterization, AI‑driven approaches are accelerating workflows that were once limited by experimental trial and error. At the same time, automation is becoming essential for handling the increasing volume of candidates and data generated by these models.
At analytica 2026, Technology Networks caught up with Dr. Simon Sumer, indirect sales channel manager at Bio‑Techne, to discuss how generative AI models and automated protein analysis workflows are working together to reduce development cycles, improve data consistency, and redefine the skills scientists need in modern laboratories.
In this interview, Sumer explores where AI is having the greatest impact on protein development and what capabilities will define protein analysis in the coming years.
AI’s impact on protein design and development
How have modern AI models changed the way proteins are designed and optimized?
According to Sumer, the most transformative impact of AI is occurring upstream, during protein development and candidate selection. Generative AI models—including diffusion‑based approaches—are now able to predict protein structures more reliably, enabling researchers to focus experimental effort on the most promising designs.
What are diffusion‑based AI models in protein design?
Diffusion‑based AI models generate new, functional protein structures by taking corrupted (noised) protein structures and iteratively refining it toward a biologically plausible configuration. By learning how proteins fold and stabilize, these models can explore vast design spaces efficiently, helping researchers identify novel, stable, and functional protein candidates before experimental testing.
“The biggest impact definitely comes from protein development,” Sumer explained. “These models help us shorten development cycles and select the best candidates for downstream analysis.”
Rather than relying solely on iterative wet‑lab experiments, researchers can now computationally explore modifications that influence critical real‑world properties such as heat stability, receptor affinity, and activity. Improving these characteristics not only enhances experimental consistency but can also reduce the amount of protein required in downstream assays.
Sumer emphasized that AI does not replace experimental validation. Instead, it optimizes the upstream decision‑making process, reducing wasted effort and unnecessary testing.
Key impacts of AI‑driven protein design:
- Faster identification of high‑value protein candidates
- Reduced experimental burden during early development
- Shortened time from concept to downstream analysis
Closing the loop: Integrating AI with automation
What does effective integration between AI and laboratory automation look like in practice?
One of the most powerful developments Sumer described is the emergence of closed‑loop workflows, where AI and automation continuously inform one another.
In these workflows, AI models are used to screen and prioritize candidates in silico. Only the most promising proteins move into automated experimental pipelines, where hands‑on time is minimized. Data generated from these experiments is then fed back into the AI models, refining predictions and restarting the cycle.
“The idea is to reduce the time scientists spend hands‑on,” Sumer noted. “You select the hottest candidates, put them into an automated flow, analyze the data, and feed it back.”
By combining AI‑guided decision‑making with automated protein analysis methods—such as immunoassays, Western blotting, and capillary‑based techniques—laboratories can accelerate development while maintaining consistent, trustworthy results.
Tech Spotlight: High-Throughput Screening 2026
Explore reliable high-throughput screening workflows for vaccine and antibody discovery, covering assay design, validation and data analysis to improve reproducibility, data quality and candidate prioritization.View Webinar
Benefits of closed‑loop AI and automation:
- Reduced hands‑on time and manual intervention
- Faster iteration between design, testing, and optimization
- More consistent and reproducible data
- Shortened overall development cycles
The impact of AI and automation on discovery pipelines
Have AI and automation delivered measurable improvements in development speed?
“We’re increasing the number of proteins released per month,” Sumer stated.
These gains are driven primarily by better screening and automation, rather than simply increasing staffing. Automation reduces human error and variability, creating results that teams can trust without repeated validation.
In essence, AI improves what goes into the pipeline, while automation ensures that what comes out is reliable and scalable.
Observed impacts on development timelines:
- Increased throughput without proportional increases in staffing
- Fewer failed candidates entering downstream analysis
- Greater confidence in automated experimental outputs
Skills for the AI-enabled laboratory
What skills are becoming essential as AI and automation become standard tools?
Cell and Gene Therapies: The Evolution of Advanced Biologics
This eBook takes a look at the challenges shaping advanced biologics, alongside practical strategies to improve consistency, scalability, and compliance across development and manufacturing.View eBook
Concerns are often raised about whether researchers need deep AI or programming expertise to benefit from these tools. Sumer argued that scientists are already well-positioned, “Researchers were the first ones to adopt these models,” he said. “They understand the value, but they also understand the limitations.”
Rather than replacing scientific judgment, AI has increased throughput expectations. Sumer explained that with easier access to advanced screening and prediction tools, researchers are now managing larger datasets and facing greater pressure to generate publishable results.
Sumer suggested that while not every scientist needs to build models from scratch, understanding how and where to apply AI is becoming increasingly important.
As datasets grow larger and more complex, Sumer highlighted data handling and bioinformatics literacy as essential skills for modern scientists. “You’re not looking at one target anymore,” he said. “You’re looking at hundreds.”
Tasks such as experimental design, control strategy, replication, and data normalization now routinely span large experimental matrices. Skills that were once confined to specialists are becoming expected across research teams.
Sumer also encouraged practical use of AI tools for programming support, such as code review and troubleshooting, while cautioning against outsourcing scientific thinking itself.
Skills gaining importance:
- Managing and interpreting high‑throughput datasets
- Designing robust controls and normalization strategies
- Using AI tools to support, not replace, analytical thinking
From optimization to de novo design
What emerging capabilities will define AI‑driven protein analysis in the future?
How Agentic AI Is Transforming the Modern Lab
This whitepaper examines how agentic AI is reshaping laboratory operations and what this shift means for the future of scientific discovery.View Whitepaper
Looking forward, Sumer sees the field moving beyond optimization of existing proteins toward designing entirely new proteins with functions not found in nature.
Advances in bioinformatics and protein sequencing tools are opening the door to de novo protein design—an area with profound implications for therapeutics, diagnostics, and biotechnology.
“When it comes to AI-driven protein design, we combine AI with the automation processes, and the result is what we call a highly assisted laboratory, instead of an automated laboratory,” said Sumner. “We can’t fully automate the protein design workflow, but we can provide scientists with the tools to be able to focus more on analysis rather than manual work.”
Future directions in protein design:
- De novo protein design beyond natural templates
- Greater support from AI‑assisted decision‑making
- Scientists spending more time interpreting results, less time generating them
As AI and automation continue to converge, protein design and analysis workflows are becoming faster, more selective, and more data‑driven—without diminishing the role of experimental science.
Key takeaways:
- Generative AI models are transforming protein development by improving candidate selection and shortening development cycles.
- Closed‑loop AI and automation workflows reduce hands‑on time while improving throughput and data consistency.
- Future protein research will rely on highly assisted laboratories, where scientists focus on analysis and insight rather than manual execution.
This content includes text that has been created with the assistance of generative AI and has undergone editorial review before publishing. Technology Networks' AI policy can be found here.
Add Technology Networks as a preferred Google source to see more of our trusted coverage.
