Skip to content

GeneSpect PRS Reporter

Configurable genome-wide risk estimation for common complex disease - computed from existing sequencing data and delivered as clinician-ready reports.

A cluster of common complex diseases the platform estimates risk for: breast cancer, colorectal cancer, type 2 diabetes, coronary artery disease and prostate cancer

About GeneSpect PRS Reporter

GeneSpect PRS Reporter computes validated polygenic risk scores directly from sequencing data. Genotyping, imputation, scoring, calibration and reporting run in one workflow. Each model is versioned and traceable to its source; each percentile is issued against a named reference population.

600+

scores available for deployment

1,150+ curated models in library

Versioned

source-linked models

ID, weights, version and provenance retained

FASTQ → report

one continuous workflow

no array or manual hand-off required

CORE SOURCE LAYER

  • PGS Catalog
  • GWAS Catalog
  • Genebass
  • Global Biobank Engine

EXTENSIBLE TO VALIDATED SOURCES

  • Pan-UKB
  • FinnGen
  • Biobank Japan
  • OpenGWAS / IEU
  • GWAS ATLAS
  • Consortium-level statistics

Why polygenic risk

Genetic risk extends beyond rare pathogenic variants

Gene-panel testing is designed to identify rare, high-penetrance variants. It does not capture common genetic variation that contributes to disease susceptibility across the broader population.

Rare variants

Rare, high-penetrance variants

explain only a subset of inherited susceptibility.

Top 5%

of a venous clot risk score

reached Factor V Leiden risk from common variants alone.

90%

of mapped causal variants

in complex immune-mediated disease were non-coding.

Sources: Farh K.K.-H. et al., Nature 2015; Klarin D. et al., Nature Genetics 2019.

How it works

One score that reflects the combinedeffect of genetic variants.

  • 01

    Association

    Genome-wide association studies identify common variants associated with a trait, each assigned an estimated effect size.

  • 02

    Weighting

    Each risk allele carried by the patient is weighted by its effect size; one and two copies contribute differently.

  • 03

    Reference

    The weighted sum is compared with a named population reference distribution.

  • 04

    Result

    The result is reported as a percentile and, where incidence data support it, as absolute risk.

The limits, stated

A polygenic score modifies estimated risk; it does not determine outcome or replace monogenic findings. Its interpretation depends on the population and reference cohort used.

The report includes the relevant confidence metrics and reference population; unreliable results are withheld and documented.

Workflow

From sequencing data to reportable results,in one validated workflow.

The workflow accepts whole-genome, whole-exome, low-pass and array-derived input. The report documents input type, recovered coverage and any reconstruction required before scoring.

  • Input and genotyping

    FASTQ, or variants already called upstream. Alignment to GRCh38 without alternative loci, calling with GATK HaplotypeCaller, quality control, and genotyping at every position the selected scores require.

    • FASTQ
    • gVCF
    • VCF
    • WGS
    • WES
    • array
    • low-pass

  • Phasing and imputation

    Applied where the input does not cover the model's variant positions - array, low-pass and exome data above all. Haplotype phases are estimated and missing genotypes imputed with Beagle, so a score's variant set is reconstructed rather than silently truncated.

    • Beagle
    • phasing
    • imputation
    • GRCh37 → GRCh38 harmonisation

  • Polygenic scoring

    The scoring engine evaluates phased or unphased SNVs against YAML-defined models, one per trait. A new score is added as a definition file, not a pipeline rewrite, and every model is versioned for reproducible re-runs.

    • YAML models
    • WDL
    • versioned
    • reproducible re-runs

  • Calibration and reporting

    Genetic ancestry is estimated from the sample and the score is compared with the appropriate reference distribution, which is named in the result. The score is assigned an evidence label and rendered as a report or returned as structured data.

    • ancestry-matched reference
    • evidence label
    • JSON / FHIR
    • HTML / PDF

Configured to your programme

Scope, analytical depth and audienceare configurable.

Two decisions define deployment: which traits are scored, and how the results are reported. Both are configured for the programme.

The trait list is configured for the programme

01 - What is scored

A library of 1,150+ published scores across 480+ traits, with further published models added as definition files. Indications below are examples, not a fixed menu.

  • Cardiology

  • Oncology

  • Pulmonology and allergy

  • Ophthalmology

  • Fitness and anthropometry

  • Metabolic and diabetes

  • Haematology

  • Neurology

  • Sleep and circadian

  • + any published model

Integrated reports can pair each indication with the relevant rare-variant gene set.

Two analytical depths, two audiences

02 · What the report contains

Analytical scope and audience are configured separately, enabling clinician-facing and individual-facing report formats from the same workflow.

Polygenic only

For clinicians

Detailed PRS report

percentile, absolute risk, plus technical appendix

For individuals

Preventive report

plain language, colour scale, lifestyle guidance

Polygenic + Rare Variants + Patient Phenotype

For clinicians

Integrated risk

condition-level view

Guidance text configured with your clinical team for each product

See how GeneSpect PRS Reporter can simplify analysis and reporting for your laboratory.

GeneSpect PRSReporter Resources

View all