🧬  Partnership Story: Helping Equine Breeders do Affordable Predictive Risk Modeling for Naked Foal Syndrome

Customer Profile:

Experienced Horse Breeder wanting to do predictive analysis to determine the risk of a foal having a recessive hereditary disorder, based on known genetic data, blood typing data, and the horse’s registry data.  In this scenario, a long-time Akhal-Teke breeder is expanding the purview of a recessive genetic disorder called Naked Foal Syndrome, (NFS). While NFS is rarely seen in Akhal-Tekes, as it is a recessive genetic disorder, it could be lurking in the background of untested horses. Most reputable breeders do know the NFS status of their breeding horses, but some do not, for whatever reason. Tracking down the possibility of an untested horse being a carrier, if parents haven’t been tested or verified clear, is the goal.

⚠️ The Challenge 

Even with a known genetic marker for NFS, breeders face steep barriers to proactive risk analysis:

• High cost of integration: Merging genetic, blood, and registry data for a single analysis can cost $20,000+.
• Specialist bottlenecks: Each question requires a data scientist to manually wrangle and interpret the data.
• No structural comparison: Existing tools can’t quantify how close an untested horse is to a known NFS-positive or -negative profile.
• Limited scalability: Exploring broader correlations (e.g., lineage, phenotype, geography) is prohibitively expensive and narrowly scoped.

💡 The Lens Solution

Using the GIPS powered Lens service we’ve created a dynamically aware model that:

• Ingests and unifies 3 different data sources containing genetic, bloodtyping, and biographical data.
• Constructs a geometric “Horse Model” for each animal, preserving structural relationships across data types.
• Filters and aligns tested vs. untested horses.
• Measures proximity of untested horses to known NFS-positive and -negative profiles—quantifying risk as a geometric distance.
• Emergent patterns surface in blood/genetic markers that correlate with NFS risk, even in the absence of direct testing.

🧠 GIPS vs Traditional Tools

Capability	Manual Analysis	LLMs / ML Models	GIPS
Data integration across domains	⚠️ Time-intensive	⚠️ Limited	✅ Seamless, multi-source ingestion
Structural similarity measurement	❌ Not supported	⚠️ Approximate	✅ Geometric, constraint-aware
Explainability	✅ Human-driven	⚠️ Black-box	✅ Transparent, auditable logic
Cost per analysis	❌ $20K+	⚠️ Variable	✅ Scalable, reusable models
Pattern discovery beyond known genes	❌ Manual only	⚠️ Data-hungry	✅ Emergent, interpretable patterns

✅ Outcome

We are working with researchers from the several Universities who have collected large quantities of genetic and blood typing data represented in diverse formats.  We received data registry information from the national association containing over 5000 Akhal-Tekes, both living and deceased, also in a different format.  The Lens ingested all the diverse data from these sources. Working with a subsection of this data, we showed that the Lens identified high-risk individuals who had not been NFS tested and measured how closely these individuals whose genetic and blood profiles closely matched known NFS-positive cases—enabling potentially targeted testing and informed breeding decisions at a fraction of the traditional cost.

➡️ Next Actions

Continue expanding tools in the field of bioinformation. From disease detection to structural risk depending on complex modeling, the Lens empowers experts to see and work with their various data systems to act with foresight—not just hindsight.

Contact us today and see how we can work together to make solving your issues faster, easier and cheaper.