GBE1 (GBED)

Crossing two of them

Born, and then lost

Lethal at birth, so the foal really is born: it gets a name, a record and a place in the family tree, and then dies over a few seconds with a chat line naming what took it.

That is deliberate. A pairing that silently produced nothing would teach a player nothing about the two horses they just bred, and the whole value of a recessive lethal is that it tells you something about both parents at once. The mechanics of the death are on the trait system page.

Its gene carrot

There is none. A disorder locus sets hasGeneCarrot() false — forcing one onto somebody’s line is hostile — and SpliceSafety keeps it out of the random splice as well, derived from the fact that it declares a non-informational Condition rather than from any hand-written list.

Gene key

horsegenetics.gbe1

Priority

100

Alleles

gbed N

Outcomes

wild, gbe1-carrier, gbed

Coat effect

none — every outcome is a wild type

Default allele

N

Wild population

2.0% of founders carry one copy; none carries two

Founder draws

1 × nextFloat()

Catalogue entries

1 — it never changes what a horse looks like

The combination table

The largest heart reduction in the mod

Which is the right shape for a disorder whose real form kills most affected foals before they are born at all.

That raises the obvious question of why it is not LETHAL_AT_CONCEPTION, which is the path for exactly that. The answer is that MET owns that path, and owning it alone is what makes it legible: one gene where a pairing simply yields nothing, and everything else born and lost where a player can watch it happen and learn something about both parents. A second conception-lethal would make "no foal appeared" ambiguous, which is the one thing that outcome cannot afford to be.

What the enzyme does

GBE1 is glucan (1,4-alpha-), branching enzyme 1. It encodes glycogen branching enzyme, which puts the α-1,6 branch points into an otherwise α-1,4-linked glycogen chain. The branching is the whole point of glycogen: it makes the molecule compact and soluble, and it creates a great many free ends from which glucose can be released at once. That matters most in the tissues with the highest and most variable energy demand — skeletal muscle, heart, liver and brain.

With no functional enzyme, cells build a poorly branched, amylopectin-like polyglucosan instead. It is a bad energy reserve and it accumulates. Affected foals show PAS-positive globular or crystalline intracellular inclusions in muscle, heart and liver, with branching-enzyme activity essentially absent. OMIA 000420

The equine disorder goes by several names — GBED, equine glycogen storage disease type IV, amylopectinosis. It is broadly homologous to human GBE1-associated GSD IV, which does not license assuming the clinical course of a different variant in a different species is the same.

The variant

Stopping at residue 34 is not a subtle activity-altering change. It is expected to abolish the enzyme outright, which is exactly what the biochemistry in affected foals shows. UC Davis VGL

Inheritance in a real population

The last two rows are hypothetical in the real world — no affected horse survives to breed — and they are impossible in the mod for the same reason. Not sex-linked: colts and fillies are affected at equal rates. In testing nomenclature N/N is clear, N/G carrier, G/G affected; a carrier passes the allele to about half its foals and is not itself a diagnosis of muscle disease.

What happens to an affected foal

Reported outcomes span late-term abortion, stillbirth, and live birth followed by profound weakness and poor muscle tone — persistent recumbency, hypothermia, flexural limb deformities, seizures, respiratory failure, cardiac failure, and death or euthanasia in early life. Laboratory findings can include leukopenia, intermittent hypoglycaemia, and raised CK, AST and GGT. The early course can look like ordinary neonatal septicaemia or pneumonia, which is why the DNA test is what makes the diagnosis.

The prognosis is grave and there is no treatment that touches the enzymatic defect. UC Davis reports that every known affected case died or was euthanised by five months, and that foals surviving to term typically die or are euthanised by 18 weeks on weakness alone.

Carriers are clinically normal

The best-supported conclusion is that N/G horses are unaffected. Early family studies found branching-enzyme activity in blood cells of dams and relatives at roughly 50% of control values — biologically consistent with carriage, and not associated with any recognised clinical phenotype.

Three things follow, and the third is the one that gets forgotten: a carrier is not an affected horse; carrier status does not mean the horse has PSSM1, RER, HYPP or any other muscle disorder; and a symptomatic adult carrier should not have its signs pinned on GBED just because the test came back N/G. The mod agrees — the carrier row costs nothing at all.

Not a coat, build or performance gene

There is no credible evidence that GBE1 controls or modifies base colour, dilution, white patterning, roan, grey, appaloosa patterning, mane and tail character, or coat texture and shedding. It sits on chromosome 26, away from the familiar pigment loci; where GBED tracks with a colour or a line, that is population history.

Nor is there a validated association between one carrier allele and mature height, bone length, frame, muscling, body mass, conformation or growth rate. For a G/G foal, “build” is affected the way grave illness affects build — weakness, poor tone, transient flexural deformity — which is disease, not evidence that GBE1 is a conformation gene.

Breeds and frequency

Established primarily in American Quarter Horses, American Paint Horses, and horses with relevant Quarter Horse or Paint ancestry. It was first characterised in Quarter Horse families and the same variant was later documented in Paints.

UC Davis estimates carrier frequency around 8.3% in Quarter Horses and 7.1% in Paints, and gives roughly 8–11% for the US Quarter Horse population overall, higher again in Western Pleasure and Cutting subgroups. AQHA requires the test for breeding stallions and includes it in the five-panel framework. These are population estimates; an individual programme can sit well away from them.

The mod’s WILD_CARRIER_PERCENT is 2.0 — lower, because it describes every founder horse in the world rather than one breed with a bottleneck. A per-breed carrier weighting is the thing that would let this page’s 8% and the mod’s 2% be the same number seen from two distances.

What is settled, and what is not

The nuance worth keeping: “GBE1 is metabolically important” is true and “p.Y34* is a lethal recessive” is true, and neither of them implies that ordinary variation at GBE1 explains ordinary variation in adult body type or athletic ability. That third claim would need population-scale genotype–phenotype work, and it does not have it.