ECTS credits:
5 (Masters students) or 3 (PhD students)

Course parameters:
Language: English
Level of course: Masters and PhD
Time of year: September 2021
No. of contact hours/hours in total incl. preparation, assignment(s) or the like: 35 hours contact. Masters students are additionally required to complete an assignment (optional for PhD students)
Capacity limits: Maximum 30 participants

Objectives of the course:
The first objective of the course is to teach students the methods being used to analyse data from genome-wide association studies (GWAS). These will include both basic analyses (e.g., quality control and single-SNP tests of association) and advanced methods (e.g., heritability analyses). The second objective is to teach students how to construct prediction models (i.e., to be able to predict an individual’s risk of developing a disease, given their genetic information). The course will be a mix of lectures and practical sessions (in the latter, students will run commonly-used software on example data).
 

Learning outcomes and competences:
At the end of the course, the student should be able to:

• Understand and describe the aims of a GWAS
• Be able to perform a basic GWAS analysis (single-SNP regression)
• Understand how heritability analyses are used to improve our understanding of complex traits.
• Understand how prediction models are used in the analysis of complex traits and healthcare systems.
• Construct basic prediction models, and understand the ideas underlying advanced prediction models.

Compulsory programme:
The course will run over four days. Each day will be contain four 1.5-hour sessions (running from 8.30-4pm, including breaks). Students should attend all sessions.

Course contents:
Preliminary schedule:

Day 1
Sessions 1+2 - Doug Speed - Introduction to GWAS, including quality control, single-snp analysis.

Sessions 3+4 - Peter Sørensen - Introduction to polygenic risk scores (PRS), including Classical PRS, testing prediction models.

Day 2
Sessions 1+2 - Luc Janss - Bayesian statistics, MCMC, feature models

Sessions 3+4 - Doug Speed - Heritability I - first pedigree based analyses, then snp-based analyses. Heritability partitioning

Day 3
Sessions 1+2 - Peter Sørensen - Successes of GWAS and prediction. Uses for personalized / precision medicine

Sessions 3+4 - Doug Speed - Heritability II - enrichments, summary statistics, selection coefficients, using this information for prediction

Day 4
Sessions 1+2 - Luc Janss - Introduction to the software BayZ

Sessions 3+4 - Doug Speed - Introduction to the software LDAK

Prerequisites:
Aimed at Masters and PhD students. Ideally students will have a basic understanding of statistics (e.g., linear regression) and some experience with the software R.

Name of lecturers:
Doug Speed, Peter Sørensen, Luc Janss

 
Type of course/teaching methods:
Lectures and computer practicals

Literature:
None

Course homepage:
None

Course assessment:
Masters students will be required to submit a written assignment 3 weeks after course date.

Provider:
Department of Quantitative Genetics and Genomics (QGG)

Time:
27 - 30 September 2021 

Place:
Main Aarhus Campus (Fredrik Nielsens Vej 2-4)    

Registration:
The course is free for Masters, Research Year or PhD students currently at Aarhus University.

For others, the registration fee is 3600 DKK.

 

If you have any questions, please contact Doug Speed, e-mail: doug@qgg.au.dk.