Early detection of Alzheimer's disease using single nucleotide polymorphisms analysis based on gradient boosting tree

Area of Science:

• Genetics
• Neuroscience
• Computational Biology

Background:

• Alzheimer's disease (AD) is a neurodegenerative disorder impacting cognitive functions and decision-making.
• Single nucleotide polymorphisms (SNPs) are key genetic variations serving as biomarkers for complex diseases like AD.
• Early detection of AD through SNP biomarkers is crucial for timely intervention and management.

Purpose of the Study:

• To develop and evaluate a comprehensive framework for the early prediction and diagnosis of Alzheimer's disease using SNP biomarkers.
• To identify the most significant genes associated with AD through advanced SNPs analysis.
• To leverage machine learning techniques for discovering novel AD biomarkers.

Main Methods:

• Proposed a machine learning framework integrating two feature selection techniques: information gain (filter) and Boruta (wrapper).
• Applied Gradient Boosting Tree (GBT) algorithm on Alzheimer's Disease Neuroimaging Initiative (ADNI-1) and Whole-Genome Sequencing (WGS) datasets.
• Evaluated feature selection methods based on their effectiveness in identifying significant AD-related genes and improving classification accuracy.

Main Results:

• The GBT algorithm combined with Boruta feature selection achieved a high classification accuracy of 99.06% on the ADNI-1 whole-genome dataset.
• The information gain feature selection method resulted in an average accuracy of 94.87%.
• The Boruta wrapper method demonstrated superior performance compared to the information gain filter technique for AD detection.

Conclusions:

• The proposed machine learning framework shows significant promise for the early and accurate detection of Alzheimer's disease.
• The Boruta feature selection technique is highly effective in identifying relevant genetic markers for AD, outperforming the information gain method.
• This approach facilitates the identification of significant genes, aiding in the early diagnosis and potential therapeutic strategies for Alzheimer's disease.