Skip to content
PRECISION GENOMICS

Turning genomic complexity into precision medicine.

Genoprism brings phenotype, variant evidence, scientific literature, disease mechanisms and genomic interpretation into one structured investigation workspace.

GRCh38WES / WGSHPO-AWAREEVIDENCE PROVENANCE
INPUTPhenotype and candidatesHPO-mapped case terms alongside filtered WES or WGS candidates.
RETRIEVALEvidence across categoriesPopulation, clinical, functional, literature and protein context.
STRUCTUREComparable evidence profilesThe same dimensions for every candidate in the case.
OUTPUTPrioritised, traceable candidatesRanked by coverage and case fit, with sources attached.
01 — THE PROBLEM

Variant interpretation shouldn't require twenty open tabs.

After filtering, an exome may still leave 10–30 plausible candidates. Each one sends the analyst across population databases, clinical assertions, phenotype ontologies, prediction tools, protein resources, functional studies and primary literature — then back again to reconcile what was found.

The evidence exists. The workflow is fragmented.

CLINClinical databases
POPPopulation frequency
GENEGene–disease validity
HPOPhenotype ontologies
PREDPrediction tools
LITPrimary literature
PROTProtein resources
FUNCFunctional studies
MODELModel organisms
GUIDEInterpretation guidance
STEP 01Bring the case inFiltered candidates and case phenotype enter the workspace. Free-text terms are mapped to structured concepts and unmapped terms stay visible.
STEP 02Retrieve across dimensionsEach candidate is populated with evidence from every category, with the source, its version and the retrieval time recorded.
STEP 03Compare on equal termsCandidates are shown against the same dimensions, so gaps in evidence are as visible as findings.
STEP 04Decide and recordThe analyst reviews each item, accepts or rejects proposals, and the reasoning is written into the case record.
02 — THE WORKSPACE

One workspace. Every evidence dimension.

Candidates are held in a single case view, each carrying the evidence that has been retrieved for it and the evidence that is still missing. Select a row to open its investigation summary.

CASEGP-000127PHENOTYPESeizures · Developmental delay · AtaxiaCANDIDATES24GRCh38 · WES · trio incomplete
HP:0001250 SeizureHP:0001263 Global developmental delayHP:0001251 Ataxia+2 unmapped terms
GENE / VARIANTPHMEPOCLFUEVIDENCE STATUS
SCN1Ac.3968G>A · p.Arg1323His · HET
PH
ME
PO
CL
FU
Strong evidence
PHENOTYPE OVERLAP
Seizures, developmental delay and ataxia overlap the reported phenotypic spectrum for this gene–disease relationship.
MECHANISM
Established mechanism includes loss of function; missense variants in this domain have been reported with functional support.
OUTSTANDING
Segregation not yet available. Parental testing pending.
CONCEPTUAL INTERFACE · ORDERING REFLECTS EVIDENCE COVERAGE AND CASE FIT, NOT A CLASSIFICATION
03 — VARIANT INVESTIGATION

Open a candidate. See every dimension, and where it came from.

GENE
SCN1A
HGVS
NM_001165963.4:c.3968G>A
PROTEIN
p.Arg1323His
ZYGOSITY
HET
ASSEMBLY
GRCh38
EVIDENCE
Allele frequency is below the disease-informed threshold configured for this gene–disease relationship, with no homozygotes observed.
SOURCE
gnomAD
VERSION
v4.1
RETRIEVED
2026-08-21 09:14 UTC
REVIEW
analyst confirmed
Evidence status is descriptive. Genoprism records what each source states and when it was retrieved; the interpretation remains the analyst's.
04 — TWO QUESTIONS

Pathogenicity is only part of the question.

How a variant is classified and whether it explains a specific patient's phenotype are related but distinct questions. A variant may remain a VUS and still be the strongest candidate in a case.

VARIANT EVIDENCE
Population frequencyFunctional studiesComputational predictionsClinical assertions
CASE FIT
Phenotype overlapInheritance patternDisease mechanismSegregation
05 — MECHANISM-AWARE

Context changes interpretation.

A damaging-looking variant is not automatically disease-causing. If the established mechanism for a gene–disease relationship is haploinsufficiency, a variant whose predicted molecular consequence does not produce loss of function warrants different weight.

GENEGeneThe gene and the conditions curated against it, each with its own validity assessment.
DISEASEDiseaseA gene may associate with several conditions that differ in mechanism and presentation.
MECHANISMMechanismThe molecular mechanism established for that specific gene–disease relationship.
VARIANTCompatibilityWhether the predicted consequence of the variant is compatible with that mechanism.
Loss of functionGain of functionDominant negativeHaploinsufficiency
06 — PROVENANCE

Every conclusion should have a trail.

Genoprism is being designed so analysts can inspect where each piece of evidence came from — the source, its version, the retrieval time and the reviewer — rather than receiving unexplained conclusions.

CLAIM
Population frequency supports rarity
scope: this variant · case GP-000127
EVIDENCE
Allele frequency below configured threshold
AF 0.0000041 · 0 homozygotes
SOURCE
gnomAD
dataset: exomes + genomes
VERSION
v4.1
retrieved 2026-08-21 09:14 UTC
RECORD
Publication and database references retained
PMID and accession stored with the claim
ANALYST REVIEW
Confirmed by reviewer with rationale
analyst_id 0412 · 2026-08-21 09:31 UTC
07 — HUMAN IN THE LOOP

Built to support judgment, not replace it.

Genoprism surfaces evidence and proposes where it may apply. The analyst accepts, modifies or rejects each suggestion, and the decision — with its rationale — is written to the case record.

GENOPRISM SUGGESTS
Population frequency for this variant is below the disease-informed threshold used for this gene–disease relationship.
gnomAD v4.1 · 0 hom · AF 0.0000041
threshold set at case configuration
AUDIT · awaiting analyst decision · every action is recorded with reviewer, timestamp and rationale
08 — EVIDENCE LAYER

The categories an investigation draws on.

Genoprism is being built to structure evidence across the resource categories analysts already work with, including public resources such as ClinVar, ClinGen, HPO, gnomAD and PubMed. No partnership or endorsement by any resource is implied.

01Clinical databasesSubmitted assertions with submitter and review status.
02Population databasesFrequency data with dataset version and ancestry breakdown.
03Phenotype ontologiesStructured terms mapped from free-text case notes.
04Gene–disease resourcesCurated validity classifications and curation dates.
05Scientific literatureGene-, residue- and variant-level reports kept distinct.
06Prediction modelsIndividual outputs reported without composite scoring.
07Protein resourcesDomains, conservation and structural context.
08Model organismsAnimal model phenotypes relevant to the relationship.
09Interpretation guidelinesCriteria frameworks applied as proposals for review.
DATA HANDLINGCase data stays scopedGenoprism is being designed so case data remains inside the organisation that owns it, with access scoped per case and per role.
AUDITActions are recordedRetrieval, proposals and analyst decisions are written to an audit trail with reviewer, timestamp and rationale.
VERSIONINGSources are pinnedEvidence carries the version of the resource it came from, so a reanalysis can be compared against what was known before.
09 — VISION

Built beyond a single analysis.

The long-term aim is an evidence infrastructure that connects genomic data to interpretable, patient-specific molecular insight — across WES, WGS, rare disease genomics, reanalysis and clinical genomics workflows.

01SequenceWES and WGS output entering the workspace.
02VariantsFiltered candidates carried as case objects.
03EvidenceRetrieval across every category, with provenance.
04InterpretationAnalyst-led reasoning over structured evidence.
05ValidationCandidates prepared for confirmatory work.
06Precision medicineMolecular insight tied to a specific patient.
10 — TEAM

Genomics, engineering and clinical practice in one room.

Genoprism is being built by people who have worked inside variant interpretation rather than adjacent to it. The team is forming; roles below describe the shape of the founding group.

CS
Clinical genomics lead
Variant interpretation practice, ACMG/AMP application and case review workflows.
FOUNDING ROLE
BI
Bioinformatics lead
WES and WGS pipelines, annotation, variant representation and normalisation.
FOUNDING ROLE
EC
Evidence curation lead
Source modelling, versioning and provenance across public genomic resources.
FOUNDING ROLE
PE
Product and platform engineering
The analyst workspace, review tooling and audit infrastructure.
FOUNDING ROLE
SA
Scientific advisors
Clinical geneticists and rare-disease researchers reviewing the workflow as it is built.
ADVISORY
PLACEHOLDER PROFILES · REPLACE WITH NAMED TEAM MEMBERS AND CREDENTIALS BEFORE LAUNCH
11 — WHO IT'S FOR

Genomic Analysts

Investigate candidates without rebuilding the same evidence picture in a dozen tabs.

Clinical Genetics Teams

Review the reasoning behind a candidate, not only its conclusion.

Diagnostic Laboratories

Structured records, version-pinned sources and an auditable review trail.

Rare Disease Researchers

Reanalyse cases as evidence changes, with prior review preserved.

Precision Medicine Programs

Connect genomic findings to patient-specific molecular interpretation.

12 — QUESTIONS

What Genoprism does, and what it does not.

Does Genoprism classify variants?
No. It aggregates and structures evidence and shows how that evidence may map to interpretation criteria. Classification stays with the analyst and the laboratory.
Does it turn a VUS into a pathogenic call?
No. A variant may remain a VUS and still be the strongest candidate for a case. Genoprism keeps those two questions separate.
Where does the evidence come from?
Public genomic resources and literature, recorded with the source, its version and the time of retrieval. Every item can be traced back.
Is it a replacement for a clinical geneticist?
No. It is decision support. Proposals are reviewed, modified or rejected by a person, and the decision is recorded.
What is available today?
Genoprism is in development. Early access is opening in stages with analysts, geneticists and rare-disease researchers.

Help build the next generation of genomic interpretation.

Genoprism is being developed with genomic analysts, geneticists and rare-disease researchers. Early access is opening in stages.

PREFER EMAIL? hello@genoprism.com · NO MAILING LISTS