GWAS in Structured Populations by Magnus Nordborg
International Centre for Theoretical Sciences via YouTube
Overview
Syllabus
Third Bangalore School on Population Genetics and Evolution
GWAS in structured populations
The agenda
Like begets like Sir Francis Galton et al.
How many genes?
The simplest model possible Fisher 1918
Dominance and epistasis
Average effect
Variance decomposition
Heritability
Typically estimated from parent-offspring regression
How do we map genes?
Linkage mapping
So why do it?
...if sample size is large enough...
The Human Genome Project meant markers were no longer limiting...
The promise
The HapMap Project
Coronary artery disease
The GWAS debacle
...height has a heritability of 80%!
A random-effects model is used to estimate variance components
In other words. ..
Could have told you so!
"The world's most expensive test of the mutation-selection balance hypothesis"
There were Cassandras. ..
Population structure confounding in GWAS
GWAS works!
Science
Dealing with "population structure"
We understand this now...
The problem of fine mapping
..but peak is far from the obvious candidate...
Including two co-factors eliminates spurious peak
A simulation example. ..
Some obvious extensions
Plants are not humans
Adaptation is different from disease
What about skin color?
Cabo Verde
Genome-wide markers explain much more
Taught by
International Centre for Theoretical Sciences