B4GALT7 (Friesian dwarfism)

Crossing two of them

What is drawn, and what is not

The real defect hits the long bones and the chest and leaves the skull almost alone. The mod cannot draw that yet: Attributes.SCALE is one number for the whole entity, and per-part scaling would mean owning the horse model rather than borrowing the vanilla one. So it renders as an overall three-quarter-size horse, with the ribcage weakness paid for in hearts. The per-part version is logged on the roadmap.

Its gene carrot

Gene key

horsegenetics.b4galt7

Priority

87

Alleles

d N

Outcomes

wild, b4galt7-carrier, friesian-dwarfism

Coat effect

none — every outcome is a wild type

Default allele

N

Wild population

4.0% 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

Not lethal — which is the interesting part

Affected foals survive. That makes this the one place in the model where a player ends up with a live animal in front of them and a decision to make about a carrier line, rather than a corpse and a chat message. Every other disorder here resolves itself; this one asks a question.

The locus

This locus is a good argument against reading a variant off its amino-acid change. Arginine to lysine is chemically conservative — both are basic — and the prediction software did not flag it hard. But the base that changed is also part of the splice-donor junction, where it matters for making a correct transcript at all. Leegwater et al., PMC5084406

What the enzyme does, and what the variant costs

B4GALT7 encodes an enzyme in the assembly of proteoglycans, which are a large part of the extracellular matrix — the material around cells that organises cartilage, tendon, ligament, bone and skin. A proteoglycan is a core protein, long glycosaminoglycan chains such as chondroitin or dermatan sulfate, and a four-sugar linker joining the two. B4GALT7 performs an early step in that linker: it adds a galactose after xylose has been placed on the core protein.

Get the linker wrong and many proteoglycans come out abnormal at once, which is why the syndrome is not confined to one bone or one tendon. The class of disorder has a name — a linkeropathy, a defect in linker synthesis — and in the horse it shows up hardest in growth plates, in tendons and ligaments, and in the rib cage.

The mechanism is defective pre-mRNA splicing, not simply a slightly wrong protein. In fibroblasts, RNA from the mutant allele was predominantly mis-spliced while the normal allele in a carrier spliced correctly; homozygous affected fibroblasts held about 2% of normal B4GALT7 mRNA by quantitative PCR. Some abnormal transcripts kept intronic sequence or picked up cryptic exons, and some are likely cleared by nonsense-mediated decay. So the defensible model is a severe reduction in usable expression — a strong partial loss of function — rather than proof that p.Arg17Lys is itself the damaging event. The authors said as much: conservation at that position may be a splice-site requirement rather than an indispensable arginine.

Inheritance in a real population

A simple autosomal recessive. Not sex-linked, so colts and fillies carry and are affected at equal rates, and it takes two copies to show the syndrome.

The 25% is a probability per foal, not a promise about any four of them. Two carriers can produce none affected, or several, by chance alone — which is the same thing the mod’s own draw does, and the reason a carrier pairing is a risk rather than a schedule.

Breeds and frequency

The variant is established in Friesians and Friesian crosses and nowhere else. Validation was unusually clean for a breed-associated recessive: all 29 affected Friesians with DNA available were homozygous, all eight obligate carriers were heterozygous, and 22 of 177 Friesians in a DNA-bank sample carried — about 12.4%, consistent with the ~12% figure UC Davis quotes. That is a discovery cohort, not a census; it should not be read as a current universal rate for every Friesian registry, country or crossbred group.

Carrier-to-clear breeding avoids affected foals while keeping the lines. Culling every carrier at once would narrow a breed that has already been through bottlenecks — which is exactly the decision this gene exists to put in front of a player.

Coat, build, and what “dwarfism” actually means here

There is no evidence that the allele touches coat colour, white markings, black pigment, mane length, feathering or hair texture. The Friesian’s black coat and heavy feather come from elsewhere. Nor is it established as a coat-quality gene: human B4GALT7 variants can affect skin, but Friesian dwarfs have not shown the loose skin, atrophic scarring or delayed wound healing described in the human syndromes.

The build is disproportionate dwarfism, not a uniformly smaller horse:

• Markedly shortened limbs.
• Shortened ribs and an abnormal rib cage.
• A head that reaches roughly normal size for the breed.
• A back of roughly normal length, so the body reads long against the legs.
• A broad chest that narrows and turns inward at the costochondral junctions — described as thickened, S-shaped junctions with inward protrusion around thoracic vertebrae 10–16, closer to a pectus excavatum than to a small horse’s chest.
• A weak, rounded abdomen and poorly developed musculature in mature affected horses.

The short stature comes from abnormal physeal — growth-plate — development, not from a growth-hormone deficiency. Affected foals showed growth retardation in limb and rib growth plates, disorganised and thickened chondrocyte columns, an irregular cartilage-to-bone transition, and dysplastic metaphyses, particularly at the distal metacarpus and metatarsus. A separate study of three dwarf foals found the hypothalamic-pituitary growth axis normal. This is an extracellular-matrix and skeletal-development disorder, and that is what explains the disproportion: long bones and ribs depend on highly ordered growth-plate matrix in a way the skull does not.

Speed, jumping and the carrier question

No published trial has measured jump height, gallop speed, stride kinetics or force-plate data in affected Friesians, so a numerical penalty has no source behind it. What the anatomy supports is a strong clinical inference: hyperextended fetlocks give unstable support and poor energy transfer, flexor-tendon laxity removes passive support of the distal limb, outward rotation through the carpi and hocks spoils straight tracking and take-off alignment, and shortened limbs change the leverage of bascule and landing. An affected horse should not be expected to be sound for jumping, sprinting, collection or high-speed turning.

The mod agrees on all three rows — the carrier costs nothing, which is what makes it hide. The affected horse’s −0.020 speed and −0.10 jump are the mod’s own numbers standing in for “not sound for athletic work”, and they are not derived from anything measured, because nothing has been.

What is settled, and what is not

Well established. The c.50G>A variant is strongly associated with the dwarfism-with-joint-laxity syndrome; it maps to ECA14 and segregates as an autosomal recessive; affected animals were homozygous and obligate carriers heterozygous throughout the validation cohort; the allele compromises the exon-1 splice donor and collapses transcript abundance in affected fibroblasts; the phenotype is disproportionate limb and rib restriction with a normal-looking head, fetlock hyperextension, flexor-tendon laxity, abnormal gait and rib-cage deformity; and the growth axis is endocrinologically normal.

Strong inference, incomplete chain. That the collapse in expression disrupts proteoglycan production in growth-plate and connective-tissue matrix; that matrix dysfunction drives the disorganised chondrocyte columns and the skeletal disproportion; that abnormal proteoglycan-rich matrix is behind the laxity and hyperextension. Biologically compelling and consistent with the pathology, but the original work did not quantify the specific proteoglycan products in affected equine cartilage, tendon or ligament.

Not settled. How much p.Arg17Lys contributes independently of the splice defect; residual enzyme activity tissue by tissue; whether other B4GALT7 variants could give milder or atypical equine phenotypes; the full cardiopulmonary, pain, fertility and longevity picture in affected horses; whether carriers have any very subtle connective-tissue or gait effect at all; any quantitative performance figure; and whether any of this generalises past Friesians and their crosses.