Early Detection of Alzheimer’s Disease Based on Single Nucleotide Polymorphisms (SNPs) Analysis and Machine Learning Techniques

Abstract

One of the brain diseases is Alzheimer's disease (AD). It is also known as a degenerative disease, and over time becomes worse. One of the most common risk factors of genetic is Apolipoprotein E (APOE) for AD, whose significant association with AD is observed in different genome-wide association studies (GWAS). Among individuals, the most common genetic variation type is known as Single nucleotide polymorphisms (SNPs). For this disease, SNPs are recognized as significant biomarkers. In the early stages of the disease, SNPs support in understanding and detecting the disease. This paper's primary goal is an early prediction and diagnosis with high classification accuracy that can perform by identifying SNPs biomarkers associated with AD. In this paper, we concentrate on using Machine learning (ML) techniques to identify the AD biomarkers. Naïve Bayes (NB), Random Forest (RF), Logistic Regression(LR), and Support Vector Machine (SVM) learning algorithm have been performed on all AD genetic data of neuroimaging initiative phase 1 (ADNI-1)/Whole-genome sequencing (WGS) datasets. In the whole-genome approach ADNI-1, results revealed that NB, RF, SVM, and LR learning algorithms, overall accuracy is scored 98.1%, 97.97%, 95.88%, and 83%, respectively. The results show that the classification techniques are favorable for the early detection of AD.