Structural analysis on mutations related to Alzheimer’s disease

Abstract

Proteins have a significant role in all biological processes. The functional properties of proteins rely upon their three-dimensional structures. Over the last twenty years substantial advances in genomic technologies have enhanced our knowledge of the genetics of Alzheimer’s disease. To that end the identification of mutations pathogenicity is still of vital importance. The methodology of the present research work focuses on the structural analysis of proteins related to Alzheimer’s disease and the comparative study to create groups with clear structural similarity and pathogenicity. To achieve that, three-dimensional descriptors (fpfh, rsd and 3dsc) were applied along with supervised machine learning classification methods. In total, 62 APP, 286 PSEN1, 68 PSEN2 and 25 MAPT variants were evaluated in our study.The output of the methodology characterized thirty mutations that were unclear at the point of the data collection.