Whole-exome sequencing in undiagnosed muscular dystrophies: a high diagnostic yield and novel insights from Iranian families.

Muscular dystrophies are a genetically heterogeneous group; in highly consanguineous populations autosomal recessive inheritance predominates, and linkage disequilibrium between tightly linked genes such as COL6A1 and COL6A2 complicates causal attribution.

Muscular dystrophies are a genetically heterogeneous group that remains diagnostically challenging, particularly in highly consanguineous populations where a substantial proportion of patients stay undiagnosed. The authors performed exome sequencing in one affected individual from each of 10 unrelated Iranian families with clinically diagnosed non-dystrophinopathic muscular dystrophy and inconclusive prior genetic testing, prioritising candidate variants with in silico tools (SIFT, PolyPhen-2, CADD, SpliceAI), population frequency databases and ACMG/AMP criteria, then validating them by Sanger sequencing, MLPA and STR haplotype analysis. Diagnostic yield reached 69.2% (9 of 13 variants across 10 families) after segregation analysis and ACMG-based reclassification, from 13 candidate variants in ten known genes (DYSF, SGCA, TK2, MAP3K20, LMNA, COL6A1, COL6A2, ITGA7, MICU1, SGCB), seven of them novel (54%). Inheritance was autosomal recessive in 84.6% of cases, consistent with 70% parental consanguinity, and a de novo splice-site variant in COL6A2 (c.1053+1G>T) was identified in a sporadic case, confirming autosomal dominant inheritance. The authors highlight interpretation difficulties in families carrying variants in the tightly linked genes COL6A1 and COL6A2, illustrating the role of linkage disequilibrium in founder populations.

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

The figure of 69.2% should not be read as a transferable exome yield: it comes from ten selected, highly consanguineous families already investigated, that is, the most favourable setting possible. The real contribution lies elsewhere, in the warning about COL6A1 and COL6A2, adjacent and in linkage disequilibrium, where attributing causality to one or the other requires haplotype analysis that no bioinformatic filter spontaneously suggests. Seven novel variants out of thirteen finally underline how sparse frequency databases remain for this population, which directly affects ACMG scoring and therefore reporting.

Analysis by Dr Thibaut Benquey