Science
Mechanism-guided antibody design
Why giving antibodies new actions changes everything
Extending the reach of antibody medicines across disease
95% of today’s therapeutic antibodies work by blocking biology, leaving an enormous opportunity for biologics that can do more. By expanding the mechanisms antibodies can use, Metaphore is expanding the kinds of diseases antibody medicines may be able to treat.
Some diseases require pathways to be activated. Others require signals to be tuned, multiple targets to be engaged, or several functions to work together in one simple medicine. These are therapeutic challenges that blocking alone cannot solve.
Metaphore’s platform makes these antibody actions designable. This unlocks a wider therapeutic landscape, enabling new mechanisms, new targets, and new treatment possibilities for millions of patients.
Where living biology directs antibody design
Metaphore’s platform combines high throughput live-cell experimentation with machine learning to design biologics that go beyond blocking.
Instead of relying on static protein structures, Metaphore starts with functional data from living systems. We measure how proteins behave in their native environment to understand what drives biological activity, including pathway activation, signaling bias, selectivity, multi-target engagement, and other therapeutic actions.
These data are interpreted through AI/ML models that identify the features responsible for specific functional outcomes. Those insights become the foundation for designing antibody medicines with new capabilities.
With each design cycle, the platform learns from new functional data, improving its ability to create antibodies that do more than blocking biology. This approach allows Metaphore to design molecules with precise therapeutic actions in weeks instead of years, expanding what antibodies can be built to do across disease.
How Metaphore’s platform works
Capture functional data from living systems through large-scale, live-cell measurements.
Map the activity landscape to reveal how proteins behave and which features drive functional outcomes.
Use these functional insights to design new antibody candidates with intended biological actions.
Refine and optimize properties through machine learning to achieve precise control of therapeutic function.
Applications
Unlock intractable targets
Mechanism-informed functional data allows us to address targets that have been inaccessible to structure-based methods.
Design for action
We design antibodies to drive specific therapeutic actions, including agonism, specificity, signaling bias, and multi receptor activity from the outset.
Define specificity
Functional mapping enables precise tuning to engage multiple targets of interest and to minimize off-target activation, all while driving the desired biological response.
Choose the ideal modality
Therapeutic actions can be designed across several antibody-based formats to match the needs of each disease.