From Statistical Genetics Courses
Advanced Gene Mapping Course
The Rockefeller University, New York
Monday through Friday, January 10-14, 2022
An Advanced Gene Mapping course will be held in online from Monday through Friday, January 10-14, 2022. The cost of the 5-day course is $100 for student, academic, and government researchers and $1,500 for researchers working in industry. This fee covers tuition and course-related expenses (cloud computing, etc.).
The course emphasis is on analyzing sequence and other omics data to elucidate the genetic etiology of complex human disease traits. Topics will include: data quality control of sequence and other types of data; single variant and aggregate rare variant association analysis of whole-genome data (genotype, sequence, and imputed) for qualitative and quantitative traits (population and family-based data); controlling for population admixture and substructure; generalized linear mix models and linear mixed models; meta-analysis; sample size estimation and power calculations; detecting gene x gene and gene x environmental interactions; analysis of epigenomic data, e.g methylation, and chromatin; heritability estimation using variant and RNA-Seq data; analysis of RNA-Seq data; eQTL mapping; elucidating pleiotropy; functional prediction and variant annotation; estimation of polygenic risk scores; Mendelian randomization; mediation analysis; and fine mapping. As mandated by the NIH there will also be a special session on responsible conduct of research that will include sessions on conflict of interest, research ethics, data management (security), and ethical use of human research subjects.
A variety of freely available software will be used to perform the practical exercises, due to differences in their functionality. PSEQ and VAT will be used to analyze sequence data to perform annotation, quality control, rare variant association analysis, and meta-analysis. FaST-LMM, GCTA-MLMA, REGENIE will be implemented to analyze population- and family-based data by applying generalized linear mixed models (qualitative traits) and linear mixed models (quantitative traits). For rare variant association analysis of trio data, RV-TDT will be applied. MultiPhen (multivariate) and PLINK (univariate) will be contrasted in their ability to detect pleiotropy; Mediation analysis will be performed using R to aid in distinguishing between biological, mediated, and spurious pleiotropy. To make inferences on causality, Mendelian randomization will be performed using MR-base. Estimation of polygenic risk scores will be performed using LDpred and non-parametric shrinkage. SuSie will be used for fine mapping to aid in the detection of causal susceptibility variants. Heritability estimates will be performed using GCTA. For analysis of eQTLs, Matrix eQTL will be used. Analysis of imputed expression data will be performed by applying PrediXCan; To perform analytical and empirical power analysis for single and rare variant aggregate tests, a variety of tools will be used that includes the Armitage Power Tool and the SKAT R library will be used. Additionally, variant annotation will be performed with ANNOVAR as well as directly using a variety of functional prediction and conservation tools, e.g. CADD, GERP, MutationTaster, MutPred, Polyphen-2, and SIFT.