Integration of microRNA sequencing and genotyping data: local ancestry inference adjustment for eQTL analysis in the Brazilian High Risk Cohort

• Grant number: 21/09584-1
• Support Opportunities: Scholarships abroad - Research Internship - Master's degree
• Effective date (Start): January 12, 2022
• Effective date (End): April 11, 2022
• Field of knowledge: Health Sciences - Medicine - Psychiatry
• Principal Investigator: Marcos Leite Santoro
• Grantee: Jessica Honorato Mauer
• Supervisor: Elizabeth Atkinson
• Host Institution: Escola Paulista de Medicina (EPM). Universidade Federal de São Paulo (UNIFESP). Campus São Paulo. São Paulo , SP, Brazil
• Research place: Baylor College of Medicine, United States
• Associated to the scholarship: 20/02247-7 - Evaluation of differential extracellular vesicule miRNA expression associated with the emergence of psychiatric symptoms in adolescents, BP.MS

Abstract

microRNAs (miRNAs) have a role in regulating processes that may lead to the development of psychiatric disorders. Alterations in miRNA expression could be mediated by expression quantitative trait loci (eQTLs). However, since eQTLs are genomic information, ancestry may influence the frequency of these variants, especially important to consider in populations that have a high degree of admixture, such as the Brazilian population. Local Ancestry Inference (LAI) is an approach for assigning genomic regions to a specific ancestry group and separating each ancestry component for subsequent analyses, allowing researchers to adjust statistical models for genomic analysis by local ancestry. This study aims to characterize miRNA eQTLs previously reported in the literature that are present in the miRNA expression data of 120 individuals from the Brazilian High Risk Cohort Study (BHRCS) and describe how ancestry and admixture influence the frequency of eQTL variants through Local Ancestry Inference (LAI). The BHRCS individuals were previously genotyped by Global Screening Array (Illumina). microRNA was isolated and is being sequenced. We will test best practices for LAI in the study population by comparing different reference panels on simulated admixed haplotypes generated through Admix-simu tool. After defining the best reference panel, we will run LAI on BHRCS individuals and describe local ancestry in the cohort. We will use the available haplotypes from 1000genomes and HGDP as reference data and perform LAI using RFMix-v1. With this study, we expect to develop a pipeline for describing miRNA eQTLs that will be applied to more individuals in the BHRCS and in other cohorts. Additionally, the use of LAI in the analysis will reduce any bias that may arise from ancestry and allow us to observe eQTLs that vary in frequency depending on local ancestry. These analyses may contribute to understanding mechanisms by which some genetic variants may influence development of psychiatric disorders through changes in protein translation by influencing miRNA expression. (AU)