Our Platform

Patient tissue is the foundation of our platform and an extraordinary gift from which we maximize scientific value. We collect and catalog normal and diseased tissues, and preserve each sample in multiple formats – enabling target identification, validation, and pharmacological testing.

By putting patient biology first, we can better understand expression, localization, and disease relevance – greatly improving the chances of success in the clinic.

Our target identification engine begins with tissue from thousands of tumors and every human organ system. We have built one of the most comprehensive healthy and tumor cellular atlases to date, with an emphasis on highly annotated, accurate, and curated data. To generate this high-quality data, we have developed proprietary methods that break apart complex tissues into individual cells without disrupting or damaging the underlying cellular biology.

We have developed a suite of analysis algorithms that assess targets across a multitude of factors. Our high-resolution atlases enable us to assess the off-tumor toxicity risk and on-tumor efficacy potential of every single antigen and antigen pair. This is integrated into a predictive toxicology index that quantifies the therapeutic potential of each target or pair in silico. This stringent analysis and filtering ensures that only the best targets and target pairs progress forward, and that the targets or target pairs that advance are precisely suited for the indication and therapeutic approach.

Our healthy and tumor cell atlases contain tens of millions of cells spanning 1,000s of cell types and more than 100 tissue types. From that proprietary base of biologic data, we layer proprietary AI methods with computational algorithms to compare every antigen in each malignant cell to each healthy cell, looking for optimal targets and target pairs aligned with a given therapeutic strategy. This bespoke analysis pipeline results in a rank-ordering of the best targets and target pairs in any indication of interest, which serves as the basis for downstream validation and therapeutic work.

We comprehensively validate the biology and expression profile of all targets of interest.

Ultra-specific targets demand ultra-specific reagents, and we have built a state-of-the-art workflow to rapidly discover high-quality diagnostic antibodies. In addition to traditional validation methods, our tissue sourcing pipeline enables comprehensive measurements of target biology on patient tissues, allowing us to build deep conviction in each target or target pair.

Our antibody engineering engine combines the best of in silico analysis and design with wet lab discovery, optimization, and validation. We integrate immunization techniques with novel sequencing and display technologies alongside machine learning methods to discover the best human antibody sequences. Our internal AI optimization engine, which is built off more than a dozen individual AI/ML models, is used to tune affinity, specificity, and viability in the clinic. The result is faster cycle times and superior therapeutic candidates.

Cartography’s pharmacology-driven discovery engine uniquely enables patient-informed antibody drug design by directly integrating single-cell human biology into every stage of molecular testing.

The deep insights into target biology from our ATLAS and SUMMIT platforms enable the selection and design of cell models that optimally recapitulate human biology. We map target expression, surface density, kinetics, and off-tumor impact across tens of millions of patient cells, allowing targets and target pairs to be selected with maximal therapeutic index, broad patient coverage and minimal toxicity. These insights are then translated into rational molecular design. The result is new therapeutics like CBI-1214, which was engineered to perform across the full spectrum of patient antigen expression while minimizing toxicity.