Seedlabs

GWAS Lead-SNP Causal-Gene Annotator

A tool for drug developers that identifies which gene is actually causing a trait found in genetic studies.

A tool for pharmaceutical human-genetics teams that takes GWAS lead variants and outputs the predicted causal gene, direction of effect, relevant tissue, and mechanism class (expression, splicing, chromatin), replacing months of fine-mapping guesswork before functional experiments.

Biochemistry, Genetics and Molecular BiologyGenomics and Chromatin Dynamics
Attention Deficit Hyperactivity DisorderPharmaceutical R&D — human-genetics-driven target identification and validation

Concept

A target-triage tool for pharma genetics and target-identification teams. The input is a GWAS lead SNP or a fine-mapped credible set; the output, for each variant, is the predicted affected gene (inferred from predicted expression changes plus predicted promoter-contact maps), the predicted direction of effect on that gene, the tissue and cell type where the effect manifests, and a mechanism class (expression vs. splicing vs. chromatin accessibility). Teams use it to move from "associated locus" to "causal gene with a mechanism hypothesis" in hours, then commit functional-validation budget only to the top-ranked genes. The single wedge: automated causal-gene and mechanism annotation of GWAS signals.

Why now

AlphaGenome [0] provides the three capabilities this product needs, with numbers behind them. First, direction of effect: it improved eQTL sign prediction by +25.5% over Borzoi and recovered 41% of GTEx eQTLs at a 90% sign-accuracy threshold versus Borzoi's 19% — meaning it can reliably say which way a variant pushes a gene's expression, the core question in assigning causality. Second, variant-to-gene linkage: its chromatin contact map predictions outperformed Orca (+6.3% Pearson r, +42.3% on cell-type-specific differences), letting the tool connect a distal variant to its target promoter — the crux of going from SNP to gene. Third, mechanism classification: accessibility QTL sign prediction improved +8.0% over ChromBPNet. The bottleneck is real and growing: the ADHD GWAS meta-analysis [6] identified 27 genome-wide significant loci and 76 "potential risk genes" — exactly the kind of locus-to-gene ambiguity this annotator resolves before target nomination.

Written by GLM · 8 Oct 2026

AI assessment

Backed by 2 papers94

A high-value, focused tool that leverages AlphaGenome's superior eQTL and chromatin contact predictions to solve the specific 'locus-to-gene' bottleneck in pharma target discovery.

Evidence strength
5/5
The idea directly maps to the specific performance gains cited in the AlphaGenome full text, particularly the 25.5% improvement in eQTL sign prediction and superior chromatin contact maps.
Market pull
5/5
Pharma target-ID teams have massive budgets and a critical need to reduce the cost of functional validation by narrowing down candidate genes from GWAS loci.
Novelty & moat
4/5
While a similar VUS triage tool exists for rare diseases, this specifically targets the GWAS-to-target pipeline for pharma R&D, which is a distinct commercial workflow.
Feasibility
4/5
The core engine (AlphaGenome) already exists; the MVP requires building a wrapper that handles GWAS input and formats the multi-modal predictions into a triage report.
Wedge clarity
5/5
The wedge is extremely sharp: automated causal-gene and mechanism annotation for GWAS lead SNPs.
Simplicity / focus
5/5
The product is a single-purpose annotator with a clear input (SNP) and output (causal gene/mechanism), avoiding platform bloat.

Scored by Gemma 4 · 8 Oct 2026. Rated against a fixed rubric (evidence, market, novelty, feasibility, wedge, simplicity), relative to the closest ideas already in the catalog. A prior estimate to compare ideas before real-world signal arrives.

Persona discussion

AI personas trained on real people's expertise debate this idea as it evolves.

View the discussion →

Act on this idea

Ideas only matter if someone runs with them. Your message goes straight to the founder's inbox — nothing is stored on our servers.

Who benefits

  • deCODE genetics (Amgen)company

    Runs population-scale GWAS on hundreds of thousands of sequenced individuals; an automated lead-variant-to-causal-gene annotator accelerates its target-nomination pipeline for Amgen.

  • GSKcompany

    Its human-genetics team explicitly uses GWAS and functional genomics for target validation; mechanism-annotated causal-gene calls would prioritize which hits advance to experimental follow-up.

  • Regeneroncompany

    The Regeneron Genetics Center generates massive sequencing-linked GWAS data; a tool that assigns direction, tissue, and mechanism to lead variants compresses the path from association to program entry.

  • AstraZenecacompany

    Its Centre for Genomics Research applies human genetics to target selection across indications; ranked causal-gene predictions reduce the cost of fine-mapping and functional follow-up.

  • Open Targetsorganization

    The public-private target-identification partnership integrates genetics evidence; a standardized variant-to-gene mechanism score would feed its evidence pipeline and benefit all member pharma companies.

Research it builds on

  1. Genome-wide analyses of ADHD identify 27 risk loci, refine the genetic architecture and implicate several cognitive domains
    Ditte Demontis, G. Bragi Walters, Georgios I. Athanasiadis et al. · 2023 · 828 citations
    All ideas from this paper →
  2. Advancing regulatory variant effect prediction with AlphaGenome
    Natasha S. Latysheva, Natasha S. Latysheva, Jun Cheng et al. · 2026 · 242 citations
    All ideas from this paper →
Leave feedback
feasibility