Soma Nova Limited
Log in Request a pilot
Platform · The Digital Twin

One person. Eleven layers. One lifelong asset.

A continuously updated, computable representation of an individual. The genome is fixed; its meaning is re-read as knowledge, phenotype, nutrition and environment evolve.

Request a pilotPlatform overview
Ownership

The twin belongs to the platform.

Not to any one product. One identity, one consent record, one contiguous audit trail — which is the only way an interpretation written today remains readable and defensible at the point of care in ten years.

OmnigeniX writesMolecular interpretation, classified deterministically and linked to the evidence snapshot that produced it.
CytoMind simulates and surfacesRuns trajectories against the twin; the Personal Health Cloud is the surface through which a person sees and governs it.
TheraMind readsAt the point of care, at the moment of prescription, and after the dose.
MSI CORE composesAgents assemble twin state into the shape of the task a person is doing.
The eleven layers
01
GenomeVariants and annotations — a stable genetic baseline, reinterpretable for life.
02
Multi-omicTranscript, protein and metabolite state — the current operating state.
03
ClinicalDiagnoses, history and medications — context for genotype–phenotype reasoning.
04
PhenotypeHPO-coded signs, driving phenotype-aware prioritisation.
05
LaboratoryLabs and biomarkers — dynamic physiological state.
06
RadiologyStructured imaging findings correlated with molecular signal.
07
NutritionIntake, composition and timing — the largest modifiable input in chronic disease.
08
LifestyleBehaviour, activity and sleep — modifiable determinants of long-run risk.
09
WearableContinuous device signals for real-time monitoring and trend detection.
10
EnvironmentalExposures, geography and context — gene–environment interaction.
11
PopulationCohort context for comparative and public-health inference.
A trusted twin

Every twin carries its own provenance.

A twin is not a data lake. Each layer declares where it came from, how fresh it is, and under what consent it may be used.

Data sources & provenance

Every observation records its origin, method, version and quality flags.

Consent status

Purpose-bound and revocable; revocation propagates to derived artefacts.

Twin & model versions

Which version of the twin, and which model versions, produced any given output.

Confidence & uncertainty

Calibrated intervals, separating missing-data uncertainty from model uncertainty.

Simulation boundaries

Where the model declines to predict, and why, stated rather than hidden.

Audit history

Every access and assertion logged immutably, available to the data custodian.

Bi-temporality

What was true, and what we knew.

Every observation carries two independent time axes: when the fact was true of the person, and when the system came to know it. That is what makes "what did the system know when this was signed?" a query rather than an investigation — and it is why data arriving late, which it always does, does not corrupt the record.

Get started

Build a twin on your population.

Anchor sites commit a defined cohort and named reviewers.

Request a pilotCytoMind — the twin’s interface