In the rapidly evolving landscape of drug discovery, the quest for high-resolution protein structures has long been the "holy grail." For years, cryo-electron microscopy (cryo-EM) has held the crown, offering near-atomic resolution that allows scientists to visualize the machinery of life. However, this precision comes at a steep price: months of painstaking sample preparation, grid optimization, and massive capital expenditure.
Enter Immuto Scientific, a biotechnology innovator co-founded by electrical engineers Faraz A. Choudhury, Ph.D., and Daniel Benjamin, Ph.D. By marrying the sensitivity of mass spectrometry with the predictive power of artificial intelligence, Immuto is carving out a niche that promises to accelerate structural biology from a months-long marathon into a high-throughput sprint.
The Bottleneck of Traditional Structural Biology
Cryo-EM is undeniably powerful, but it is not without its limitations. As Benjamin, the company’s CTO, points out, the workflow is notoriously iterative. Researchers must often screen thousands of grids, optimize buffer conditions, and navigate the complex physics of protein vitrification before a single usable structure is resolved. For many biotech and pharmaceutical companies, these timelines are simply too slow to keep pace with modern drug discovery cycles.
"Cryo-EM is always going to be a relevant tool, especially for proteins that haven’t been solved yet," Benjamin acknowledges. "But it requires a big investment, specialized equipment, and specialized expertise."
While cryo-EM provides a static, high-resolution "snapshot," it often struggles with proteins that are inherently flexible or disordered—the very proteins that are frequently the most attractive targets for therapeutic intervention.
The Immuto Approach: A Paradigm Shift in Throughput
Immuto Scientific is taking a fundamentally different tack. By leveraging AI-assisted mass spectrometry, the company can probe proteins in their native environments—including living cells—to gather structural data that is both rapid and highly relevant.
"We can get data on about 1,000 samples per week with our platform, so that roughly translates to something like 100 structures per week," Benjamin explains. This throughput represents a quantum leap over traditional methods. By moving away from the "grid-and-freeze" model, Immuto is able to study proteins in complex systems, including 3D cultures, patient-derived tumors, and organoids. This shift is critical: for target discovery, the goal is to capture the native biology and inherent heterogeneity of human disease, rather than relying solely on immortalized, artificial cell lines.
Chronology: From Engineering Roots to Clinical Ambition
The journey of Immuto Scientific began in 2018, born from the vision of Choudhury and Benjamin. Applying their background in electrical engineering to the biological sciences, the founders sought to solve the structural "data gap" that hampered early-stage drug development.
- 2018: Founding of Immuto Scientific. The team begins developing a platform to integrate radical labeling mass spectrometry with AI-driven computational modeling.
- Early Development: The platform is validated using standard human cell lines to demonstrate its capability in measuring residue-level information.
- Expansion: The technology evolves to handle increasingly complex biological matrices, including 3D organoids and actual patient-derived tissue resections.
- 2025: Immuto announces a strategic partnership with Daiichi Sankyo, focusing on a solid-tumor program that leverages the platform for novel target discovery and antibody development.
- 2026/2027: The company prepares to share its v1 antibody-antigen structural model at the PEGS conference, with plans to advance its internal oncology lead program into clinical trials by 2027.
Supporting Data: AI as a Catalyst, Not a Replacement
The rise of protein structure prediction tools—such as AlphaFold, Boltz, Chai, and ByteDance’s Protenix—has fundamentally reshaped the field. Yet, even these tools have limitations. They are exceptional at generating "plausible" structures, but when tasked with predicting the exact interface of an antibody-antigen complex, they often struggle with the "ranking" problem.
"If you were to output, let’s say, 1,000 different possible structures, the correct structure will be in there, but it won’t necessarily be the top-ranked structure," says Benjamin.
Immuto’s platform solves this by providing empirical, experimental constraints derived from mass spectrometry. These constraints act as a filter for the AI models, allowing the system to identify the correct structure with high confidence. According to Benjamin, the results are "almost dead on with what you would see with Cryo-EM," providing the validation needed to move from a computational prediction to a high-confidence biological truth.
Strategic Philosophy: The Epitope-First Strategy
Immuto’s internal pipeline reflects its technological strengths. Rather than starting with the traditional metric of "binding affinity," the company focuses on the "binding site" (the epitope).
"We intentionally start with medium- or low-affinity binders just to ensure they are binding to the exact right epitope," Benjamin notes. "Once we know it binds to the right site, we can engineer all the binding affinity we need."
This "epitope-first" strategy is a direct result of their structural platform’s ability to map interactions in complex biological environments. By ensuring the antibody hits the target exactly where it matters, Immuto reduces the risk of failure later in the drug development pipeline.
Official Responses and Industry Implications
The implications of Immuto’s platform extend far beyond the laboratory. By lowering the barrier to entry for structural information, the company is democratizing a process that was once the exclusive domain of major academic institutions and massive, well-funded R&D departments.
The industry is already taking note. The partnership with Daiichi Sankyo serves as a strong signal that major pharmaceutical players are hungry for more efficient ways to validate drug targets. As Immuto gears up for its clinical debut in 2027, the focus will be on proving that its platform can not only predict structures but can successfully guide the development of life-saving therapeutics.
The Future: Toward In Vivo Structural Biology
Perhaps the most exciting frontier for Immuto is its work with patient-derived models. By moving away from the limitations of purified, recombinant proteins and toward the study of proteins in their native, cellular contexts, Immuto is bringing "structural biology into the clinic."
The ability to look at tissue resections and capture the heterogeneity of a tumor is a game-changer. It allows researchers to see how a drug interacts with a target in the context of the entire cellular environment, including post-translational modifications and protein-protein interactions that are lost in standard structural biology workflows.
Conclusion
Immuto Scientific sits at the intersection of two of the most disruptive forces in modern science: high-throughput mass spectrometry and predictive AI. By addressing the "bottleneck" of structural biology, the company is not merely iterating on existing methods—it is defining a new workflow that prioritizes speed, biological relevance, and precision.
As the company prepares to present its v1 model at the PEGS conference, the scientific community will be watching closely. If Immuto can consistently turn raw mass spec data into accurate, high-resolution structures at the scale they claim, they may well become the primary bridge between the promise of AI-driven drug discovery and the reality of clinical success. For a field that has spent decades waiting for the next "step change," the future, it seems, is finally taking shape.
About the Author
