WFS1 Wolframin — R558C Variant Card

Molecular Atlas Pilot Variant · RareResearch.AI · Windsor Symposium Demo

Prepared: May 26, 2026 · Expert-reviewed & corrected: August 11, 2026 (see Expert Review & Corrections below) · Source artifacts archived in this folder

Identity

Structural Context

Why this matters: Wolframin has eleven transmembrane helices anchoring it in the ER membrane. R558 sits in a short cytoplasmic loop directly preceding TM8. A positively charged arginine at this position likely contributes to membrane-proximal helix packing and electrostatic interactions with anionic phospholipid headgroups. Substituting cysteine removes the positive charge, introduces a free thiol capable of forming aberrant disulfide bonds, and is expected to perturb helix anchoring and ER folding fidelity.

Computational Predictions

AlphaMissense

DynaMut2

Clinical Evidence

Inheritance and scope

Pathogenic/Likely pathogenic — for WFS1-related spectrum (unresolved mode) (dominant or recessive); Hearing loss, inheritance unstated (inheritance not specified)

ClinVar classifies this variant as Pathogenic/Likely pathogenic for WFS1-related spectrum (unresolved mode) (dominant or recessive); Hearing loss, inheritance unstated (inheritance not specified). ClinVar records this under a WFS1-spectrum label that spans both disease families — recessive Wolfram syndrome (biallelic) and the dominant WFS1-related disorders. The label does not resolve which applies, and it is not a severity statement.

Phenotype scope is interpretation of submitted records. The computed values on this card (AlphaMissense, DynaMut2 ΔΔG, pLDDT) are measurements of the model, not of a patient. Presentation cannot be generalised from neighbouring variants: one nucleotide over can present very differently. Structural similarity supports a mechanistic hypothesis, never a phenotype claim.

R558C is one of 326 pathogenic-spectrum variants identified for WFS1 in ClinVar (out of 2,243 total variants catalogued).

Patient Population

R558C is the Ashkenazi Jewish founder mutation in WFS1, carried by approximately 2.32% of Ashkenazi Jewish individuals (~1 in 43) — 1.32% in mixed Ashkenazi/Sephardi cohorts, 0.04% Sephardi (PMID 30014265). The gnomAD global AF of 0.035% reflects the general population, not the at-risk group. This makes it the largest single identifiable carrier group for any WFS1 mutation worldwide and the highest-yield starting point for any clinical intervention targeting WFS1.

gnomAD v4 confirms the population-specific picture independently: the Ashkenazi Jewish allele frequency is 1.327% (393 of 29,608 alleles), against a global allele frequency of 0.035% — a 38-fold enrichment. A carrier frequency derived from that allele frequency (~1 in 38) brackets the published figure of ~1 in 43 (PMID 30014265); the published number remains the one to cite, and the derived one is shown only as an independent cross-check. Source: gnomAD v4, retrieved 2026-08-21.

Homozygotes in the general population: gnomAD v4 reports 4 R558C homozygotes, all Ashkenazi Jewish. gnomAD excludes individuals with severe pediatric disease, so the presence of adult homozygotes is consistent with the milder, later-onset course documented in the cohort literature below, and argues against a fully penetrant severe recessive phenotype. This is interpretive signal, not a reclassification. Source: gnomAD v4, retrieved 2026-08-21; flagged in expert review by Nufar Kotler, 2026-08-11.

Clinical spectrum: the ClinVar "Pathogenic/Likely pathogenic" label and AlphaMissense 0.849 are correct as annotations but incomplete as clinical interpretation. Published cohort data shows R558C homozygotes present a substantially milder, later-onset form of Wolfram syndrome — mean diabetes onset ~17–19 years, optic atrophy ~29 years, Wolfram diagnosis ~30 years — and the R558C protein is more stable than other recessive pathogenic variants (PMID 30014265, Diabetologia 2018; PMID 36134655, JCI Insight 2022; Neurol Genet 2021, doi:10.1212/NXG.0000000000000578). Pathogenicity here is a spectrum, not a single label.

Druggability Assessment

Final classification: Category 3 / Category 4 — Most Druggable.

The convergent evidence:

• pLDDT 84.56 — region is well-structured and modelable (NOT IDR; safe for docking)
• DynaMut2 ΔΔG = −0.5 kcal/mol — mild destabilization. The fold survives the mutation; this is NOT a gross misfolding mutation requiring gene therapy
• DynaMut2 interaction map — R558's guanidinium group makes specific ionic and polar contacts with neighboring residues in the TM7–TM8 region. The R→C swap eliminates these contacts, disrupting the local interaction network without globally unfolding the protein
• AlphaMissense 0.849 (LPath) — confirms the mutation has functional consequence, despite mild structural cost
• ClinVar Pathogenic/Likely pathogenic with no conflicts — clinical penetrance is real

Therapeutic implication: R558C is exactly the kind of variant the Atlas was built to identify — pathogenic but structurally tractable. The protein folds; the lost interactions define a specific local site that can be targeted by small molecules. Priority for docking experiments and pharmacological chaperone screening. Gene therapy is not required and likely not warranted given the largely-intact fold.

This is also the highest-value variant in the Atlas pilot because the carrier population (~1 in 43 Ashkenazi Jewish individuals, PMID 30014265 — the largest identifiable WFS1 carrier group worldwide) is large enough to support a real clinical pathway.

Research Path (decision tree)

ΔΔG < 2  + binding site affected   → CATEGORY 3 — docking experiments
ΔΔG 2–4                            → CATEGORY 2 — pharmacological chaperones
ΔΔG > 4                            → CATEGORY 1 — gene therapy
pLDDT < 50                         → CATEGORY 5 — IDR, experimental validation
Stable fold + functional site hit  → CATEGORY 4 — site-specific docking

R558C pLDDT 84.56 rules out Category 5. Final assignment pending ΔΔG.

Artifacts in this folder

Expert Review & Corrections

This card has been reviewed by an affected-family domain expert. Reviewer: Nufar Kotler (Jewish Wolfram Network), 2026-08-11. Corrections from that review are recorded in the Atlas corrections log (v1, 2026-08-11) and applied to this card:

1. Carrier frequency corrected (order-of-magnitude error). A previous version of this card stated R558C is carried by "approximately 1 in 3" Ashkenazi Jewish individuals, citing a "27–34%" range. That figure had no published basis. The verified published carrier frequency is ≈2.32% (~1 in 43) in Ashkenazi Jews (1.32% mixed Ashkenazi/Sephardi; 0.04% Sephardi). R558C remains the Ashkenazi founder variant and the largest identifiable WFS1 carrier group — the corrected number carries that argument on its own. Ref: PMID 30014265.

2. Pathogenicity presented as a spectrum. The single "Pathogenic/Likely pathogenic" label and AlphaMissense 0.849 stand as annotations but are incomplete as clinical interpretation: published cohort data shows R558C homozygotes present a substantially milder, later-onset form (mean diabetes onset ~17–19 yrs; optic atrophy ~29 yrs; Wolfram diagnosis ~30 yrs; R558C protein more stable than other recessive pathogenic variants). Verified published data outranks the single predictive label. Refs: PMID 30014265; PMID 36134655; Neurol Genet 2021, doi:10.1212/NXG.0000000000000578.

3. Population-specific frequency now shown on every card, not just this one (v2, 2026-08-21). The v1 fix corrected this card's Ashkenazi figure but left every other variant showing only a global gnomAD allele frequency, which is the same error in miniature: a global number standing in for the at-risk population. The Atlas now carries a per-ancestry allele-frequency breakdown on all 2,085 variants and names the highest-frequency population alongside the global figure. Ref: gnomAD v4, retrieved 2026-08-21 via scripts/pull_gnomad.py.

4. Homozygote counts added (v2, 2026-08-21). The reviewer noted that the presence of homozygotes in general-population databases belongs in the clinical picture, and cited four for R558C. The schema had no homozygote field at all — 0 of 2,085 records carried one. gnomAD v4 reports 4 R558C homozygotes, all Ashkenazi Jewish, confirming the reviewer's figure exactly. The field is now present on every variant, with an explicit "not available" where gnomAD has no entry rather than a silent zero. Ref: gnomAD v4, retrieved 2026-08-21.

5. Pathogenicity is now phenotype-scoped Atlas-wide (v2, 2026-08-21). Per the reviewer's second correction, a WFS1 variant may be pathogenic for diabetes or another WFS1-related phenotype without that saying anything about Wolfram syndrome or expected severity. Every card now states what the variant is classified for, under which inheritance mode, and at what ClinVar review strength — with 0–1★ single-submitter entries flagged as unverified submissions rather than documented phenotypes.

Expert review happens in the open — predictions get corrected in the artifact itself, not silently. The full corrections log is published on the Atlas site.

Modeling assumptions (documented per Atlas standard)

1. AlphaFold predicts the monomer structure only. Wolframin's native oligomeric state in the ER membrane is not modeled here.
2. DynaMut2 ΔΔG is computed on the static AlphaFold structure; dynamic ensemble effects are not captured.
3. AlphaMissense is trained on missense variants observed in primates; novel substitutions in less-conserved regions may be less reliable.
4. ClinVar classifications reflect aggregated clinical evidence as of February 2026; new submissions may update.
5. Ashkenazi carrier frequency is cited at ≈2.32% (~1 in 43) per PMID 30014265 (Diabetologia 2018). An earlier version of this card stated "1 in 3" with a claimed "27–34%" range — that figure had no published basis and was corrected on expert review, 2026-08-11 (see Expert Review & Corrections).

Every assumption documented. Every score sourced. The Atlas standard.