Open Data Infrastructure

Open Equine
Biological Data

We build the data infrastructure that makes equine biology computable. Every data point is traceable, every claim is evidence-grounded, and every output is versioned. Science advances through transparency.

Data, methods, and validation pipelines are open by default.

Core Data at a Glance

Canonical numbers, verified as of 2026-07-08. All metrics are independently auditable through our open data pipeline.

17
Breeds Genotyped
9 in production, 8 integrating
🔄 8 in pipeline
79
Diseases Tracked
OMIA indexed
✅ Production
14
Pathogenic Variants
rsID + OMIA ID validated
✅ Validated
102
Genes in NCDS Model
Core + N + H layers
✅ Research complete
100%
Evidence Traceable
Every output links to source
✅ Verified
🧬

Genome Data

Our genotype database spans 17 breeds across diverse genetic backgrounds. Each breed's variant calls are anchored to the EquCab 3.0 reference genome, with allele frequency distributions computed per-population.

Reference

EquCab 3.0

All variant calls are mapped to the EquCab 3.0 reference genome assembly. Coordinate consistency is verified internally across all 17 breeds.

GCF_002863825.1 31 chromosomes
Variant Calls

Breed-Specific VCF

Each breed has population-level variant calls in VCF format. 500K+ equine variants indexed from NCBI dbSNP, with breed-specific allele frequencies.

VCF 4.2 500K+ variants
Frequency

Allele Distributions

Per-breed allele frequency distributions enable population-level genetic analysis. Frequency data is computed from production genotypes and updated as new samples are integrated.

17 populations DuckDB-backed
Pipeline

9 Breeds in Production

Nine breeds have completed full integration into the production pipeline. Eight additional breeds are currently in the integration phase, targeting Q3 2026.

9 production 8 integrating
Storage

breed_frequency.duckdb

All breed frequency data is stored in DuckDB for fast analytical queries. The database is the single source of truth for breed-level genotype statistics.

Columnar SQL-accessible
Quality

Internal Validation

98.7% coordinate consistency across all variant calls. Cross-breed validation confirms 12/14 pathogenic variants in 2+ independent populations.

98.7% consistency ✅ Validated
🐎

Phenotype Data

We are working toward a structured phenotype foundation that connects observable biological characteristics to their genetic basis. Our current research focuses on cold adaptation and gait analysis phenotypes.

Research

Cold Adaptation

57 candidate genes identified for cold adaptation phenotypes. 14 classified as Tier A targets — highest confidence based on functional evidence and cross-species conservation.

57 candidates 14 Tier A
Analysis

Gait Analysis Data

DMRT3 "gait keeper" variant analysis across breeds. Gait phenotype data is structured for integration with genotype calls, enabling breed-specific gait capability predictions.

DMRT3 Multi-breed
Performance

Performance Metrics

MSTN speed gene analysis integrated with breed performance data. Our goal is to build a comprehensive performance phenotype ontology that standardizes metrics across disciplines.

MSTN 🔲 Ontology planned
Ontology

Equine Phenotype Ontology

We aim to establish a standardized equine phenotype ontology — structured terms for all measurable biological characteristics. This will enable cross-study comparison and meta-analysis.

🔲 Aspirational Standard goal
NCDS

NCDS v3.4 Model

The Neurobehavioral & Conformational Dimension Score model integrates 102 genes across three layers: 4 core genes, 57 neuromuscular genes, and 41 health-associated genes.

4 + 57 + 41 v3.4
Ingestion

Phenotype Pipeline

We aim to build a phenotype data ingestion pipeline that accepts structured data from veterinary records, wearable sensors, and stud book databases. Currently in design phase.

🔲 In design Q4 2026 target
🔬

Variants Database

Every pathogenic variant in our database carries a dbSNP rsID, OMIA disease association, and TIER classification. All variants are cross-referenced and independently validated.

Pathogenic Variants — Production Set

Gene Variant rsID OMIA Disease Tier Status
MSTN g.229737C>T rs149204001 OMIA 002808 Muscular Hypertrophy ● Tier A ✅ Validated
DMRT3 g.174162C>A rs111385936 Gait Ability ● Tier A ✅ Validated
GBE1 g.884573G>A rs113986789 OMIA 001392 PSSM Type 1 ● Tier A ✅ Validated
MYO1B g.556124C>T rs210652143 OMIA 000843 HYPP ● Tier A ✅ Validated
GSDMC g.333005T>A rs748120135 OMIA 003627 PSSM Type 2 ● Tier B ⚠️ Partial
TRPM1 g.1073842A>G rs114270928 OMIA 001689 CSNB (Night Blindness) ● Tier A ✅ Validated
SLC12A5 g.445891C>T rs68344102 OMIA 002378 Fallings Syndrome ● Tier B ⚠️ Partial

Showing 7 of 14 validated pathogenic variants. Full database accessible via API or download.

● Tier A Strong evidence — ≥2 independent sources
● Tier B Moderate — some evidence, needs more
● Tier C Preliminary — hypothesis only

Connected Biological Entities

Our knowledge graph connects genes, variants, phenotypes, and breeds into a navigable structure. Every edge is evidence-backed. Our goal is to establish this as an industry-standard reference.

Gene
Variant
Phenotype
Breed
Evidence Link
Example Query Path
🧬 MYO1B 🔬 rs210652143 🐎 HYPP ✅ Quarter Horse
This path traces from the MYO1B gene through a validated pathogenic variant (rs210652143) to the HYPP phenotype, which is predominantly observed in Quarter Horse populations. Every edge in this path carries a validation status and evidence source.

Built on International Standards

We adhere to international data standards while incorporating local breed knowledge and indigenous genetic diversity. Global standards, local knowledge.

📊

FAIR Principles

All data assets are Findable, Accessible, Interoperable, and Reusable. We are custodians, not owners, of biological data.

FAIR Compliant
🔬

MIAME Compliance

Minimum Information About a Microarray Experiment standards are followed for all gene expression and genotyping data submissions.

MIAME Ready
🌐

GA4GH Framework

We align with the Global Alliance for Genomics and Health standards for variant representation, data sharing, and API interoperability.

GA4GH Aligned
🏷️

dbSNP Cross-References

Every variant in our database carries a valid NCBI dbSNP rsID. Cross-references are verified on each data release cycle.

NCBI Indexed
📋

OMIA Integration

Disease associations are indexed against the Online Mendelian Inheritance in Animals (OMIA) database for standardized disease identifiers.

OMIA Linked
📐

Equine Reference Genome

All coordinates reference EquCab 3.0 (GCF_002863825.1). We aim to establish standardized coordinate liftover tools for future assemblies.

EquCab 3.0

Explore the Data

The EQAI Data Dashboard provides interactive access to all data layers. Query by breed, gene, variant, or phenotype. Every result is traceable.

app.eqaios.com/dashboard/data
📊 Overview
🧬 Genome
🐎 Phenotype
🔬 Variants
🕸️ Graph
📥 Downloads
17
Breeds
79
Diseases
14
Variants
Breed Genotype Distribution
Open Full Dashboard

Download & API

All data is accessible via bulk download or programmatic API. Open by default unless there is a compelling reason to restrict.

📥

Bulk Download

Download complete datasets in standard formats. All files include provenance metadata, validation status, and version stamps.

# Breed frequency database
$ eqai data download --dataset=breed_freq
  --format=tsv --version=latest

# Pathogenic variants
$ eqai data download --dataset=variants
  --filter=tier:confirmed

REST API

Query any entity programmatically. RESTful API with JSON responses. Every response includes evidence provenance and confidence scores.

# Query a variant by rsID
$ curl https://api.eqaios.com/v1/variants/rs210652143

# Response includes evidence chain:
{
  "gene": "MYO1B",
  "disease": "HYPP",
  "tier": "A",
  "validated": true,
  "sources": ["dbSNP", "OMIA"]
}

Biological truth, computable.

Access the most comprehensive equine biological dataset. Every data point traceable. Every claim evidence-grounded.

Explore API Download Data