LD-Based Mapping by Cluster Analysis of the GWAS To Determine a Single Variant Most Associated With Alzheimer's Disease
ISEF · 2022 Mathematics
Overview
My novel method of machine learning that I utilized with cluster analysis can be done on any computer and is fast and easy to learn. My novel method investigated a genome-wide association study (GWAS). The method involves scanning the genomes from many different people and looking for genetic markers that can be used to predict the presence of a disease. It was hypothesized that if the variants that are highly associated (r>.80) with Alzheimer's disease (from the GWAS database) are placed in a hierarchical cluster analysis algorithm, then the number of variants can be greatly reduced and linear regression of medoid analysis will further reduce that number to a single variant that is most likely to cause Alzheimer's disease thereby creating a novel method of machine learning. I used variants that have at least nine other variants associated with Alzheimer's disease, placed these variants into a hierarchical cluster analysis algorithm that reduced that number to a few clusters. The cluster's dendogram showed two distinct, significant clusters of the variants. This dendogram reduced the variants responsible for Alzheimer's disease from 16 to 2. Then, placed these clusters into a medoid partitioning algorithm to find the centrally located variants. Medoid cluster 2 showed that LD1 has the higher location. Using a linear regression model of those medoids, the highest r2 value identified the most likely variant that causes Alzheimer's disease: r2 for Variant rs290358 =.9060 r2 for Variant rs573521= .9693 Therefore, Variant rs573521 is the best variant to study for the most likely cause of Alzheimer's disease because of its higher r2 value. My hypothesis was substantiated since my novel method of machine learning narrowed down a large number of variants to a single variant.
Awards (2)
- American Statistical Association: In-Kind membership to ASA for all winners, including honorable mentions
- American Statistical Association: Honorable Mention
Competition history
- ISEF 2022
Resources
Related projects
ISEF · 2020
Functional Genetic Biomarkers of Alzheimer's Disease and Gene Expression from Peripheral Blood
ISEF · 2022
A Machine Learning Approach to Identifying Blood-Based Biomarkers for Differential Diagnosis of Alzheimer's Disease
ISEF · 2017
ALZCan: A Statistical Machine Learning Based Framework to Predict Future Onset of Alzheimer's Disease Using Genome-Wide Association Analyses, Polygenic Risk Scoring, and Multiple Neuroimaging Modalities (rs-fMRI, FDG-PET, Florbetapir-PET)
ISEF · 2020
Developing a Pre-Risk Assessment Incorporating Machine-Learning and Biomarkers to Diagnose Alzheimer's Disease
Closest projects by meaning, across every fair and year in the corpus.
Source: Regeneron International Science and Engineering Fair