Workshop ASHG2022 - Genetics and Genomics Digital Forum
The Michigan Imputation Server
Data Preparation, Genotype Imputation, and Data Analysis
You can download the slides of all workshop sessions here. Please also have a look at the individual sessions below for additional training material.
- Christian Fuchsberger, firstname.lastname@example.org (Eurac Research)
- Sebastian Schönherr, email@example.com (Medical University of Innsbruck)
- Lukas Forer, firstname.lastname@example.org (Medical University of Innsbruck)
- Xueling Sim, email@example.com (National University of Singapore)
- Saori Sakaue, firstname.lastname@example.org (Broad Institute)
- Albert Smith, email@example.com (University of Michigan)
Workshop Description (from the program)
Genotype imputation is a key component of modern genetic association studies. The Michigan Imputation Server has thus far helped > 9,600 researchers from around the world to impute > 95 human genomes. This interactive workshop is intended for anyone interested in learning how to impute genotypes and to use the imputed genotypes, highlighting recent reference panels, including the multi-ancestry panel from the TOPMed program and a specialized HLA panel.
A brief overview of imputation and the server will be followed by demonstrations and exercises, including:
quality control and preparation of genetic data for use on the MIS with a special focus on diverse ancestries, chromosome X, and the HLA region;
tracking runs and use of the application program interface for larger jobs;
downloading data from the MIS and preparing data for genetic analysis;
performing a GWAS using imputed data and interpreting results, taking into account imputation quality;
using the additional features, such as the polygenic risk score calculation.
We encourage participants to ask specific questions about their own projects. Workshop materials, including slides and example data sets, will be made available before the workshop and will remain online at the MIS website. We expect that this workshop will enable participants to generate high-quality imputed data sets and to effectively analyze them.