commvita
Connected care platform
Genomics

Genomics, and the family behind the result

How a genomic referral flows from indication to cascade, how genetic data is classified and gated, how the panel, the pedigree and the result fit together, and how the whole thing is shaped to the NHS Genomic Medicine Service — with what is not built stated as plainly as what is.

Live vs demonstrated: Live — real, API-backed platform logic (wired end-to-end today) Demonstrated — representative control surface with seeded data / illustrative UI mock-up

1What this is, in one paragraph

commvita is not a sequencing platform and not a variant-interpretation engine. It is the clinical layer around the laboratory: the place a clinician decides a test is indicated, records the consent that makes it lawful, chooses the right national panel, routes it to the right laboratory, draws the family, receives the classified result back onto the person’s record, and offers testing to the relatives that result implicates. The genome is sequenced by a Genomic Laboratory Hub. Everything on either side of that is what this module does.

The design premise. A genomic result is the one clinical finding that is never only about the patient. A pathogenic variant identifies a family — people who never consented, were never seen, and may not know the test happened. Every design decision below follows from that: consent is a gate rather than a checkbox, access denies by default, refusals are audited as carefully as disclosures, and a result opens a cascade rather than closing a case.

2How the flow works

Six states, one gate and two returns. The states are the platform’s own vocabulary, not a diagram drawn for this document.

CONSENT GATENothing is sent to a laboratory without an explicit, recorded consentDraftIndication, HPO terms,panel chosendraftAcceptedEligibility met,GLH routedacceptedSubmittedSample takenand sentsubmittedIn labSequencing andinterpretationin_labResultVariant classifiedACMG 1–5results_availableActionedReported, filed,cascade offeredactionedThe person’s record — Genomics tabRendered only where the reader holds the domain.A class 4 or 5 finding tints the tab.Cascade — a relative becomes a new referral, not a note on this one
Consent is a gate, not a field. A referral can be drafted and its eligibility assessed before consent exists — that is the conversation in which consent is sought. What cannot happen is a sample reaching a laboratory without it. And the pathway does not end at a report: a class 4 or 5 finding opens a new referral for each relative offered testing, because a relative is a patient in their own right with their own consent, not a row on somebody else’s record.

3How the data is classified — and who may read it

Genetic data is special category data under UK GDPR Article 9(1). In commvita that is not a banner: it is an RBAC domain that every genomics route is gated on.

Roles that hold the genomics domainRead is a disclosure here, so a read grade is a real permission — there is no “view only is harmless” for a genome.superadminFPlatform administrationadminFPlatform administration, including RBAC itselfgp_partnerRWOrders testingclinicianRWOrders testingcaldicott_guardianROversight of the disclosure itselfgp_registrarRPharmacogenomics affects prescribingclinical_pharmacistRPharmacogenomics affects prescribingmental_health_practitionerRPharmacogenomics affects prescribingEveryone else — DENIED BY DEFAULTexecutive · care_coordinator · social_worker · community_nurse · district_nurse · anp · first_contact_physio · hcaig_manager · pcn_admin · group_admin · ods_admin · social_prescribing_link_worker · read_only — a role added tomorrow with no genomics grade inherits refusal, not access.
Default deny is the whole point. A role with no explicit genomics grade gets none — so a role added next year cannot acquire access to genomes by being forgotten. Executive is denied deliberately: it is a senior oversight role, and seniority is not a clinical reason to read somebody’s genome. Care coordinators are denied under minimum-necessary — arranging an appointment does not require the variant. Both are grantable by an administrator on the RBAC page; neither is granted by default.

Refusals are audited

Not just successes

Every access decision is written to the audit log whether it was granted or refused. A successful-access log tells an IG reviewer who read a genome; a refusal log tells them who tried, which is the question an access review is actually asking.

The tab’s absence is the control

Not rendered, then refused

On the person’s record the Genomics tab is appended only where the reader holds the domain. A tab that appears and then 403s has already disclosed that this person has a genomic record — and would write a denial every time any clinician opened any patient, until the log stopped meaning anything.

Your own grade is not secret

One ungated question

GET /genomics/access is deliberately not behind the gate: your own permission level is not special-category data. The record asks once and either offers the tab or does not, rather than provoking a 403 to find out.

4Consent, and what withdrawal actually does

One test, six separable consents. Bundling them into a single signature is how a person ends up in a biobank they never agreed to.

Clinical testing

The test itself

The only element that is required for the test to proceed. Recorded with who obtained it, who witnessed it, the date, the method and the consent form version.

Incidental findings

What you did not go looking for

Whether the person wishes to be told about a finding unrelated to the reason for testing. Answered before sequencing, because afterwards is too late to ask honestly.

Cascade authorisation

Telling the family

Whether relatives may be approached on the strength of this result. Held separately because consenting to a test is not consenting to a family conversation.

Research

Secondary use

Separate from care. A person can be tested and decline research entirely, and the register records that as a first-class answer rather than an omission.

Biobank

The physical sample

Storing the sample is a different question from analysing it, and is asked as one.

Long-term storage

The data, over time

How long the sequence itself is retained — the element most likely to matter decades after the consultation everyone has forgotten.

Withdrawal is shown, never hidden. A withdrawn consent (UK GDPR Art.7(3)) leaves the record visible with its withdrawal on it, rather than the row disappearing. A clinician must be able to see that testing was consented and no longer is — a vanished row reads identically to a test that never happened, which is a different clinical fact entirely.

5Panels — ordering the right test

Referrals are placed against R-coded panels, the units the National Genomic Test Directory is written in, each carrying its genes, its eligibility criteria, its sample requirement and its turnaround target. A clinician chooses a panel, not a gene list.

CodePanelGenesTest typeTarget
R208Hereditary breast and ovarian cancerBRCA1 · BRCA2 · PALB2 · CHEK2 · ATMTargeted panel42 days
R30Hereditary colorectal cancer (Lynch)MLH1 · MSH2 · MSH6 · PMS2 · EPCAMTargeted panel42 days
R27Heritable thoracic aortic aneurysmFBN1 · TGFBR1/2 · SMAD3 · ACTA2 · COL3A1Targeted panel56 days
R67Inherited cardiac conditions — arrhythmiaSCN5A · KCNQ1 · KCNH2 · RYR2 · CALM1–3Targeted panel42 days
R116Familial hypercholesterolaemiaLDLR · APOB · PCSK9 · LDLRAP1Targeted panel21 days
R304PharmacogenomicsCYP2D6 · CYP2C19 · CYP2C9 · HLA-B · DPYD · TPMT · NUDT15Pharmacogenomics28 days
R59Rare and undiagnosed diseaseWhole genome — trioWGS trio180 days
R240Solid tumourWhole genome — tumour vs germlineWGS somatic28 days
Eligibility travels with the panel. Each entry carries the criterion it is ordered against — NICE NG151 family history for R208, Simon Broome or a Dutch Lipid Clinic score of 6 or more for R116, Amsterdam II or Bethesda for R30 — so the question “is this person eligible?” is answered where the test is chosen rather than at the laboratory’s rejection letter. Referrals also carry HPO terms, the standard phenotype vocabulary, so the clinical picture reaches the laboratory in a form it can act on.
Verification ceiling — read this before quoting the table. The panel content above is what commvita holds. england.nhs.uk is not reachable from the build environment these documents are produced in, so the R-codes, gene lists and turnaround targets could not be checked against the live National Genomic Test Directory. The directory is versioned and changes; confirm against the current publication before relying on any specific code, and treat the shipped set as a configuration to be updated, not a national reference.

6The pedigree

A family history written as prose cannot be reasoned about. Drawn, it answers in one glance the question the whole referral turns on: is this pattern inherited, and who else is at risk?

Generation 1Maternal grandmotherMaternal grandfatherGeneration 2MotherFatherGeneration 3PProbandPathogenicSisterRelative — cascadePendingReading the pedigreeCircle = female · square = maleFilled = affected by the conditionDiagonal line = deceasedViolet outline = genetically testedPP = proband, the person who brought the family to attentionInheritance pattern: autosomal dominantWhich is the whole reason a result is never only about one person:each first-degree relative carries a one-in-two prior.
Every symbol is a stored field, not a drawing. Sex, affection, whether the person has died, whether they have been tested and what the result was are all recorded per member, with the inheritance pattern held on the pedigree itself — so the diagram is a rendering of the record rather than a picture beside it. Partners are drawn to complete each union, as pedigree convention requires; only the members the record holds carry a status. The cascade relative appears here as pending the moment their referral is raised, which is what makes the pedigree the working document of a family clinic rather than a historical sketch.

7The result

Results come back classified on the ACMG/AMP five-tier scale. The classification is the laboratory’s; commvita’s job is to carry it faithfully and make the class 3 problem visible rather than tidy it away.

Class 5 — PathogenicACMG/AMP 2015 criteriaActed on. Management changes; cascade testing offered to relatives.Class 4 — Likely pathogenicACMG/AMP 2015 criteriaActed on. Treated as clinically actionable; cascade offered.Class 3 — Uncertain significance (VUS)ACMG/AMP 2015 criteriaNOT acted on. Not a diagnosis and not a clean bill of health — may be reclassified as evidence accumulates.Class 2 — Likely benignACMG/AMP 2015 criteriaNot acted on.Class 1 — BenignACMG/AMP 2015 criteriaNot acted on.commvita records the classification the laboratory issued. It does not classify variants, and it does not re-classify them — that is the laboratory’s accountable act.
Class 3 is the honest difficulty. A variant of uncertain significance is neither a diagnosis nor reassurance, and the commonest harm in genomic medicine is a VUS being read as one or the other. It is displayed as its own tier with its own meaning, and it is excluded from the actionable count — the tile that tints a person’s record counts class 4 and 5 only.

What a class 4 or 5 changes

Actionable finding

It surfaces on the person’s record without being hunted for, it makes cascade testing a live offer to first-degree relatives. In the pharmacogenomic panels it ought to reach the prescriber — a CYP2D6 poor metaboliser is a reason not to prescribe codeine, and that is a fact about a prescription rather than a fact filed in a genetics letter. It does not today: see section 9.

What commvita does not do

Deliberately out of scope

It does not call variants, classify them, or re-classify them when evidence moves. Those are the laboratory’s accountable acts under its own accreditation. A platform that quietly re-scored a variant would be practising genomics without a licence.

8The person, and the portal

Three lanes, and the third is the one worth reading carefully.

Live today — the portalFamily history, self-declaredExplicitly NON-genomicFeeds earlier screening/portal/family-historyA patient can tell the service that breast cancerruns in the family and be offered screening earlier.No genomic data is involved, and that separation isdeliberate.Live today — the recordGenomics tab on the person’s recordRendered only where the domain is heldClass 4/5 finding tints the tab/genomics/patient/{id}The tab’s absence IS the access control. It isnot rendered and then refused — a tab thatappears and 403s has already disclosed that thisperson has a genomic record.Designed, NOT builtPatient-signed genomic consentWithdrawal from the portalNotification to the laboratoryNo /portal/genomics endpointThe consent record models a portal signature andcommvita Sign is the ceremony that would carry it.The screen does not exist. Said here so nobody buysit.
Family history and genomic data are kept apart on purpose. The portal’s family-history declaration is a non-genomic statement — “my mother had breast cancer” — and its lawful basis is preventive medicine and direct care. It can therefore open earlier screening for a great many people without any of them being sequenced, consented to Article 9 processing, or entered into a genomics register. Collapsing the two would drag the heaviest consent regime in the platform onto a screening reminder.

9Pharmacogenomics — the same test, a different question

Diagnostic genomics asks why is this person ill? Pharmacogenomics asks something the whole population has a stake in: how will this person handle the drug I am about to prescribe? The panel is a buccal swab, taken once, and the genotype does not change — so a single test speaks to every future prescription of an affected medicine.

That changes the economics of testing. A diagnostic panel answers one question for one family. A pharmacogenomic panel answers a question that recurs at every prescribing decision for the rest of somebody’s life — which is why it is ordered as R304 through exactly the same referral, consent and routing machinery as every other panel, rather than as a separate service.

One test, several prescribing decisions — and none of them reaches a prescriptionGENEDIPLOTYPEPHENOTYPEWHAT IT IS RELEVANT TOCYP2D6*4/*4Poor metaboliserAffects medicines that require conversion to an active form.CYP2C19*1/*17Rapid metaboliserAffects exposure to several commonly prescribed medicines.CYP2C9 + VKORC1*1/*2 · -1639A/AAltered sensitivityCombined genotype relevant to anticoagulant dosing.DPYD*1/*1Normal metaboliserRelevant to a class of chemotherapy agents.TPMT*1/*3AIntermediate metaboliserRelevant to thiopurine tolerance.SLCO1B1521T>CDecreased functionRelevant to statin exposure.UGT1A1*1/*28Intermediate metaboliserRelevant to one chemotherapy agent.Two gaps, and the second is the one that matters clinicallyNO DOSING GUIDANCE IS SHOWN HERE, and none should be read into the table. commvita ships nodrug–gene knowledge base: the published pharmacogenomic sources are unreachable from the environmentthese documents are produced in, so nothing has been read from source. A recommendation typed from memory isindistinguishable from a curated one at the point where it would change a prescription.THE LIBRARY NOW EXISTS, AND IT IS THE ORGANISATION’S. A governed pharmacogenomic library ships at/population-genomics: an interaction cannot be saved without its own evidence source and version, andan EMPTY library is reported as empty — never as safe, because a prescriber reads a clear screen asreassurance and the silence would be commvita’s, not the biology’s.WHAT IS STILL NOT THERE: nothing carries a phenotype into the prescribing path by itself. The library ischecked when somebody opens it; commvita never changes a prescription, and an alert records what aCLINICIAN decided. This tab remains illustrative and calls no API.
What is missing from this table is the point. An earlier version of it carried evidence grades and dose guidance, reproduced from the demonstration data on the module’s own screen — and that data has since been found to carry citations nobody had read, attached to a specific drug dose. They were removed from the product and they are absent here. A plausible citation is indistinguishable from a checked one at the point where it would change a prescription, which is the worst place for the difference to stop mattering.

Consent is the same regime

Article 9, not a lighter touch

A pharmacogenomic panel is genetic data and carries the same six separable consents, the same access control and the same withdrawal rules as any other genomic test. It is tempting to treat it as routine because the result is a dosing note; the data is not routine, and a CYP2D6 diplotype identifies a family exactly as any other variant does.

What has to be built

Named, not implied

One of the two has arrived. A persisted phenotype and a governed library now live at /population-genomics, where every interaction carries the evidence source and version the deploying organisation configured it from. What is still missing is the rule that fires where the prescription is written. Until that exists, a pharmacogenomic result is a document somebody has to remember to open.

10How this is built to fit the NHS Genomic Medicine Service

The service is centralised by design: seven Genomic Laboratory Hubs serve the whole of England, ordering is against one national test directory, and the clinical decisions sit with the referring service. commvita is shaped to that model rather than around it.

Ordering is by national code

Not a local test menu

Referrals carry an R-code, its eligibility criterion and its turnaround target — the same unit the directory is published in. A local catalogue would have to be re-mapped at every boundary; a national code does not.

Routing is to a hub

Seven GLHs

Every referral records the GLH it went to, and the default is resolved from the jurisdiction rather than hardcoded — a Crown Dependency referring into the service under a memorandum of understanding routes differently from an English trust, and neither is a code change.

The phenotype travels

HPO, not free text

Clinical features are recorded as Human Phenotype Ontology terms, which is what makes a referral interpretable by the laboratory and re-analysable later when a gene panel is updated.

Consent follows the national forms

Form A, Form B, versioned

The consent record names the form and version used, and holds the national elements separately — including data sharing with the national programme, which is its own answer rather than an implication of being tested.

Beyond England

Configured, not forked

A jurisdiction with its own national genomic programme configures its own regulator, panels and routing through the Jurisdiction Profile. ACMG/AMP classification and HPO are international and do not vary; the programme around them does.

What is NOT connected

Stated plainly

There is no message exchange with a GLH or with the national service. No test order leaves the platform, no report arrives electronically, and commvita is not on any national genomics interface. A referral is recorded and progressed by people.

11Where each part lives

CapabilityWhereHowStatus
Referral register — create, route, progress/genomicsGET/POST/PUT /genomics/referralsLive
Panel directory — R-codes, genes, eligibility, turnaround/genomicsGET /genomics/panelsLive
Consent register — six separable elements, withdrawal/genomicsGET/POST/PUT /genomics/consentsLive
Pedigree — members, affection, testing, inheritance/genomicsGET/POST/PUT /genomics/pedigreeLive
Result & ACMG classification/genomicson the referral recordLive
Analytics — by status, by panel, consent rate, pathogenic count/genomicsGET /genomics/analyticsLive
The Article 9 gate — default deny, refusals auditedevery route aboverequire_genomics()Live
Genomics tab on the person’s record/patients/:id/whole-personGET /genomics/access + /genomics/patient/{id}Live
Pharmacogenomics — diplotype and phenotype, illustrative only/genomicsseeded client-side, calls no APIDemonstrated
NHS GMS / GLH message exchangenot builtDemonstrated
Patient-signed genomic consent in the portalnot builtDemonstrated
FHIR R4 / openEHR projection of a genomic reportmapping defined, no endpointDemonstrated

12The honest edges

The register was held in memory. It is now a database, and that is worth saying out loud. Until recently the genomic referrals, consents and pedigrees were module-level Python lists rather than tables, so a record created through the interface was lost when the service restarted and, with the platform’s default of two API replicas, invisible to the other one. A consent withdrawal that evaporates on a deploy is a consent nobody can evidence. The three registers are now real tables keyed to the person, the demo set is folded in once and no longer changes, and a record a clinician actually created is never re-pointed at somebody else to tidy a demonstration. A reference that matches nobody is stored unlinked and reported rather than attached to a convenient patient.

A write could be made by a role that may not read. Three endpoints were gated on seniority rather than on the genomics permission itself, which let the executive role — deliberately denied every genomic read — edit a genomic record. That is closed, and it is named here for the same reason as the three claims below: the fix is only half the work if nobody is told what was wrong.

Nothing is written to an openEHR clinical data repository. The intended record shape is CLUSTER.genetic_findings with a FHIR R4 DiagnosticReport + Observation/variant projection. commvita ships no archetype definitions and no AQL, and the default openEHR target in the shipped configuration is a mock, so no composition is persisted. The openEHR rating for this module is Partial, and the identifiers that appear on seeded referrals are seed values, not evidence of a write.

The pharmacogenomics tab is illustrative, and it carried citations nobody had read. The diplotypes and phenotypes shown are demonstration data. The tab also asserted per-alert provenance naming published pharmacogenomic guidelines, a regulator safety update and a research knowledge base, and stated a specific drug dose — none of it read from source, because those sites are unreachable from this build environment. All of it has been removed from the product and none of it appears in this document. The clinical argument — that a metaboliser status belongs at the point of prescribing rather than in a letter — is sound; the knowledge base and the wiring are both absent, and each organisation would have to govern and source its own.

There is no patient-facing genomic consent screen. The consent model anticipates one and commvita Sign is the signature ceremony that would carry it. Until it is built, consent is recorded by the clinician who obtained it.

Three claims in the module were wrong and were corrected while this document was written. It named an RBAC permission that has never existed, asserted that genomic data was held in a CDR, and stated that all genomic compositions were queryable over FHIR. None of that was true. They are named here rather than quietly fixed, because a document that inherits a product’s false claims launders them.

Non-SaMD. This module records clinical decisions and presents recorded data. It does not diagnose, does not classify variants, and does not decide who is tested. The clinical judgement is the clinician’s and the classification is the laboratory’s.

UK GDPR Article 9 — special category (genetic data)ACMG/AMP 5-tier variant classificationHuman Phenotype OntologyNHS Genomic Medicine Service — 7 Genomic Laboratory HubsNon-SaMD — records decisions, does not make them
© 2026 Commvita Digital Health Solutions Ltd. All rights reserved.