Statistical Genetics of DMD Gene Mutations in a Kazakhstan Cohort: MLPA/NGS Variant Validation and Genotype–Phenotype Modelling

<b>Background</b>: Duchenne muscular dystrophy (DMD) results from pathogenic variants in the <i>DMD</i> gene, one of the most significant and most mutation-prone genes in the human genome. Although global mutation registries are well developed, genetic data from Central Asian populations remain extremely limited, leaving essential gaps in regional epidemiology and in the understanding of genotype-phenotype patterns. <b>Methods</b>: We conducted a retrospective analysis of patients with genetically confirmed dystrophinopathy in Kazakhstan. Variants were identified using multiplex ligation-dependent probe amplification (MLPA) for exon-level copy number alterations and next-generation sequencing (NGS) with Sanger confirmation for sequence-level changes. All variants were classified under ACMG guidelines. Statistical modelling incorporated mutation-class grouping, exon-hotspot mapping, reading-frame status, CPK stratification, chi-squared association testing, Spearman correlations, Kaplan-Meier ambulation survival curves, and multivariable logistic and Cox regression. <b>Results</b>: multi-exon deletions were the predominant mutation class, with a marked concentration within the canonical hotspot spanning exons 44-55. Recurrent deletions affecting exons 46-50 and 45-50 appeared in several unrelated patients. NGS confirmed severe protein-truncating variants, including p. Lys1049* and p. Ser861Ilefs*7. Phenotypic severity followed a consistent hierarchy: hotspot-associated deletions and early truncating variants showed the earliest loss of ambulation, whereas splice-site variants and duplications demonstrated the mildest courses. CPK levels correlated with the extent of genomic involvement, though extreme elevations did not consistently predict early functional decline. Regression models identified hotspot localization and out-of-frame effect as independent predictors of ambulation loss. <b>Conclusions</b>: This study provides the first statistically modelled characterisation of <i>DMD</i> gene mutations in Kazakhstan. While the mutational landscape largely mirrors global patterns, notable variability in clinical severity suggests the presence of population-specific modifiers. Integrating comprehensive molecular diagnostics with statistical-genetics approaches enhances prognostic accuracy and supports the development of mutation-targeted therapeutic strategies in Central Asia.