Enhancing SNPs selection via statistical methodologies and population structure study

• Grant number: 21/12618-5
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Effective date (Start): March 01, 2022
• Effective date (End): June 30, 2022
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Ricardo Cerri
• Grantee: Juliana Ferreira Alves
• Supervisor: Luiz Fernando Brito
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Research place: Purdue University, United States
• Associated to the scholarship: 20/08634-2 - Machine learning for selection of SNPs related to the diagnosis of Alzheimer's Disease, BP.IC

Abstract

Single Nucleotide Polymorphism (SNP) is the variation at a single position of a nucleotide chain where the DNA is formed. The association of SNPs and phenotypes is pertinent to assess the genetic basis of human health and the investigation of the genomes. The information provided by SNPs enables predicting the responses of an individual to certain drugs, as well as for the search of genes related to hereditary diseases. SNPs have been associated with complex diseases such as cardiovascular diseases, diabetes, cancer, and Alzheimer's Disease. The association between phenotypes and SNPs can be studied and evaluated using supervised Machine Learning methods. However, the literature shows that data preprocessing techniques can improve the predictive ability of the models. Since the current FAPESP project in Brazil focus on finding the relationship between Alzheimer's Disease and SNP variants using Machine Learning, the BEPE research will complement the current research, evaluating the impact of data preprocessing using statistical filters for SNPs selection. Moreover, this project plans to execute and benchmark Machine Learning models to comprehend the population structure and the accuracy in the genomic prediction. This will improve the current results already obtained and deepen the student knowledge on Data Science and Bioinformatic techniques, and on genome studies applied to the selection and treatment of SNPs. (AU)