New version of the tutorial on heritability and MCMCglmm
Doing quantitative genetics, is hard. Well, not nightmarish-hard, but still: you need to understand what you are doing, what are the limits of what you are doing, and remember what kind of factors you could have missing that would create a bias on your parameters. All of this while handling relatively complex models like the animal model. And (why not?) doing Bayesian statistics for the first time of your life.
Because of this, I decided a looooong time ago (about 10 years ago, time flies…) to write a little tutorial on heritability, animal model and Bayesian statistics, explaining how to use the R package MCMCglmm to do all of this. The idea was to make the enterprise slightly less intimidating, and detail all of the pitfalls while explaining how to handle the model complexity.
It might not seem so, but in 10 years, quantitative genetics advanced quite a lot, and so did my understanding of it and of the animal model. So I decided it was time to update the tutorial, make it prettier (very important!), get rid of the very few things I happened to disagree with nowadays and add a lot of new cool stuff:
- More theoretical explanations on generalised animal models (for non-Gaussian traits) and tutorial on a few of them (binary and Poisson, basically), using my R package QGglmm.
- More theoretical explanations and tutorial on multi-traits models, including multi-traits with non-Gaussian traits.
- Explanations and tutorial on how to deal with fixed effects and include their variance in the phenotypic variance (including for non-Gaussian traits)
- A few other things that I’m thinking about right now, I guess…
The new tutorial is largely based on the Phoenix dataset that I created for this article, which hopefully is grounding the examples in something a little bit more concrete that before (provided you believe that phoenixes do exist).
I hope you will enjoy this new version of the tutorial, please do not hesitate to let me know of anything wrong (or super cool) with it!
