FREQUENTLY ASKED QUESTIONS
The Intelligent Design™ platform is Feromics’ proprietary function-first discovery workflow. It begins by
physically isolating immune cells based on directly observed functional behavior — what cells are actively doing
in the context of a specific disease — before any molecular data is collected. These functionally defined cell
populations are then profiled molecularly, producing datasets where every data point carries a verified functional
label. The result is biological data purpose-built for AI model training, target discovery, and clinical prediction.
Most AI models in immunology train on bulk genomic data — thousands of cells measured together and
averaged into a single number. In that averaging, disease-specific functional signals get flattened into noise.
Feromics solves this at the source. Our data is function-labeled before the AI ever sees it, so models train on
verified signal rather than a mixture of signal and noise. The result: faster training, fewer samples required, lower
cost, and more reliable outputs that translate into actionable discoveries.
Far fewer than bulk approaches require. Because we start with cells already selected for functional relevance, the
data contains significantly less noise. High-quality, function-labeled data means fewer samples are required to
generate high-confidence results — reducing cost and accelerating timelines for partners.
Feromics generates functional single-cell immunology datasets that describe immune cell activity in disease,
functional immune cell subpopulations, signaling pathways driving immune response, and interactions between
immune cells and disease biology. Every data point carries a verified functional label — directly observed, not
inferred. These datasets are suitable for target discovery, biomarker identification, and AI model training.
Our patented microwell array technology enables high-throughput functional single-cell analysis at scale —
specifically covering measurement of serial killing and cell persistence, the two most clinically valuable functional
readouts in cell therapy development. The platform is designed to generate large, AI-ready datasets across
multiple disease areas and partnership programs simultaneously.
Standard single-cell platforms — including RNA sequencing-based tools — measure what immune cells contain:
their genes, their proteins, their molecular state at a moment in time. They produce a static snapshot. Feromics
measures what immune cells are actively doing in the context of a disease — directly observed functional
behavior in real time. That functional behavior is what determines therapeutic response. Our data cannot be
collected by any other platform, cannot be approximated from public atlases, and does not exist anywhere else.
Until recently, isolating and analyzing immune cell function at single-cell resolution at scale was not technically
feasible. Advances in microfluidics and functional assays have now made it possible. Feromics was built
specifically to leverage these advances — and our patented microwell array is the only technology that measures
serial killing and cell persistence at single-cell resolution at scale. That is why the datasets Feromics produces did
not exist before.
Feromics is fundamentally a functional immunomics platform company. The Intelligent Design™ platform is the
engine — it generates the functional, disease-specific data that makes drug discovery, diagnostic development,
and AI model training possible at a level no other platform can match. When the platform identifies a promising
target or pathway, Feromics has the full in-house capability to advance it into a therapeutic candidate or
diagnostic application. The platform creates the discoveries. We have the capability to translate them.
Feromics partners across multiple phases — from functional data generation and collaborative research programs to translating platform discoveries into viable pipeline assets. The right partners are pharmaceutical companies, biotech organizations, AI and technology companies, cell therapy manufacturers, clinical developers, and academic medical centers and research institutions advancing work at the frontier of immunology.
Partnerships may include collaborative discovery, functional dataset access, or end-to-end translation of functional biology into pipeline assets.
Three independent validations: Published science — a preprint demonstrating target discovery through
functional single-cell analysis, identifying the KLRG1 target via the Intelligent Design™ platform, a discovery
traditional bulk datasets could not produce. Patented technology — a patent on our proprietary microwell array
covering measurement of serial killing and cell persistence at scale. Government validation — a $4.1 million
grant from ARPA-H, the U.S. Advanced Research Projects Agency for Health, under Contract No.
75N91024C00036, providing independent external validation of the platform and approach.
Feromics sits at the intersection of three large and growing markets: AI in drug discovery, immunotherapy
development, and precision immunology. The platform can be applied across target discovery, biomarker
identification, AI model training, cell therapy manufacturing, companion diagnostics, and therapeutic
development. As more functional immunomics datasets are generated through partnerships, the platform
advantage compounds automatically — more data improves model accuracy, better models attract more
partners, more partners generate more data.
Most AI models trained on genomic data learn that certain gene expression patterns tend to correlate with
certain outcomes. Correlation is not causation — and in noisy bulk data, many correlations are unreliable.
Because Feromics data is function-labeled from directly observed outcomes, the AI learns that specific molecular
programs caused specific functional behaviors — physically verified at single-cell resolution. The model learns
from verified ground truth, not statistical association. That is why Feromics-trained models produce outputs that
translate into actionable clinical predictions.
Donor Intelligence is Feromics’ application of functional immunomics to cell therapy manufacturing. Rather than
selecting donors based on standard expansion metrics — which do not predict clinical performance — Feromics
identifies and selects high-performance donors based on verified functional killing data: directly observed
cytotoxic activity at the single-cell level. The cells going into manufacturing are the cells most likely to produce
durable clinical outcomes. This reduces manufacturing risk and improves the probability that cell therapy
products will perform in patients.
The moat operates at three levels. First, patented technology: our microwell array is the only system that
measures serial killing and cell persistence at single-cell resolution at scale — and it is exclusive to Feromics.
Second, biology superiority: function-selected cells produce molecular data that is cleaner, more specific, and
more meaningful than data collected without functional context. Even a competitor who decided tomorrow to
lead with biology would need our patented technology to replicate it. Third, the data network effect: with every
experiment run and every partnership engaged, the platform generates more functional immunomics data —
improving model accuracy, attracting more partners, and widening the advantage automatically.