Genetic and structural characterisation of alzheimer's disease associated variants in an Eastern Indian cohort.

Abstract

Alzheimer's disease (AD) shows substantial clinical heterogeneity often caused by genetic variants contributing to both familial and early-onset forms. However, the genetic landscape of AD in the diverse Indian population remains further characterised. To identify probable pathogenic variants associated with AD, representing individuals from Eastern India and to assess their potential structural effects using molecular dynamics (MD) simulations. Whole-exome sequencing was performed in 29 radiologically confirmed AD patients. Variants identified were screened using standard in silico prediction tools and categorised according to their involvement in amyloid precursor protein (APP)-related and non-APP pathways. Molecular dynamics simulations of filtered variants in PSEN1, SORL1, and ABCA7 genes were conducted using GROMACS 2024 to assess conformational stability, flexibility (RMSD, RMSF), compactness (Rg), and solvent accessibility (SASA). Thirty-six variants across 18 genes were identified, including five pathogenic variants and 31 variants of uncertain significance. Frequently affected genes included ABCA7 (8 variants), SORL1 (3 variants), PSEN1 (two novel variants), and APP (London mutation, Val717Ile). Most patients (83.3%) had a positive family history, with a mean age at onset of 54.17 ± 11.74 years, presenting with typical AD, posterior cortical atrophy, or frontal variants. MD simulations revealed increased rigidity, altered electrostatic interactions, and disrupted ATP-binding dynamics in mutant proteins compared with wild type. This study identifies novel and known AD-associated variants in an Indian cohort and demonstrates their potential structural impact. Integrating genetic and structural analyses provides valuable insights into AD pathogenesis in underrepresented populations.