The science behind CohortLayer
CohortLayer is built on a well-established scientific finding: human genetic evidence is the strongest early predictor that a drug target will succeed, and that predictive power is growing as more of the genome is studied. We build on the published evidence the field relies on — ClinVar, the GWAS Catalog, and the peer-reviewed literature — and bind it continuously to population-scale human cohorts. We do not generate the underlying genetic research; we make published findings actionable against your specific target, across more than one cohort, and keep them current as interpretations change. Our validation approach is being developed with established cohort-genetics researchers, and peer-reviewed validation is in progress.
What scientific finding is CohortLayer based on?
That targets supported by human genetic evidence are substantially more likely to succeed in the clinic — roughly twice as likely to be approved, with the most recent analysis refining the advantage to 2.6-fold from Phase 1 to approval (Nelson et al., Nature Genetics 2015; Minikel et al., Nature 2024). Critically, that advantage is not saturating as the field matures — it is growing, because every new genetic finding adds evidence that was not there before (Minikel et al., Nature 2024). This is the scientific basis for both validating targets against human genetics and for doing so continuously rather than once.
What data sources does CohortLayer use?
CohortLayer builds on the open evidence sources the human-genetics field relies on — ClinVar (clinical variant interpretations), the GWAS Catalog (published genome–phenotype associations), and the peer-reviewed literature — and binds that evidence to large managed-access human cohorts for replication. The public sources tell us what has been found; the cohorts let us test whether it holds for a specific target, across populations.
How is CohortLayer's method validated?
Our approach is being developed and validated in collaboration with established cohort-genetics researchers, against the same public evidence sources and population-scale cohorts described above. Peer-reviewed validation is in progress.
Is CohortLayer a medical or diagnostic service?
No. CohortLayer is a research and analytics service that returns aggregate, population-level evidence to organizations — biotech, techbio, and research teams. It does not analyze or return any individual's medical or diagnostic information. The managed-access cohorts it works with cannot, by design and by governance, be used to produce individual-level or consumer results.
How does CohortLayer handle genetic data and privacy?
Under EU/GDPR-native principles, aggregate-only. Genetic data is the most sensitive category of personal data, so the design boundary is strict: individual records never leave the cohort's secure environment, nothing is re-identified or moved, and only aggregate statistics are returned. This is both the legal basis for working with managed-access cohorts and a deliberate trust commitment.
Sources
- Nelson et al., Nature Genetics 2015 — The support of human genetic evidence for approved drug indications.
- Minikel et al., Nature 2024 — Refining the impact of genetic evidence on clinical success.