Pop. and Landscape genomics workshop, Canberra 2014
Session notes and slides :
Course (.pdf document)
Course – Slides (.pdf document)
Practical work (.pdf document)
The practical work on landscape genomics uses data on Loblolly pine (Pinus taeda) sampled in the US by the Eckert lab http://eckertdata.blogspot.ch/, (Eckert et al., 2010; Eckert et al., 2010). The purpose is to compute association models between SNPs data and environmental variables that will be downloaded or computed in a GIS.
We will use a GIS software mostly to visualize data (Quantum GIS), another one to produce environmental variables from Digital Elevation Models (DEMs) (SAGA GIS), and a third one to compute spatial statistics (OpenGeoda).
Quantum GIS can be found here:
- For windows: http://www.qgis.org/en/site/forusers/download.html
- For Mac OS: http://www.kyngchaos.com/software/qgis
SAGA GIS can be found here:
- For Windows (recommended) : http://sourceforge.net/projects/saga-gis/files/
- For Mac OS : http://sourceforge.net/apps/trac/saga-gis/wiki/Compiling SAGA on MacOSX
Compilation of SAGA GIS for Mac OS is quite complicated. I would recommend to borrow a computer running Windows for this part.
OpenGeoda can be found here:
Landscape Genomics software
We will use are SamBada – based on multivariate logistic regressions -, LFMM, which considers population structure, and Admixture which computes membership coefficients to populations for each individual.
- SamBada can be found here for Linux, Windows and Mac OS:
- LFMM can be found here for Mac OS, Linux and Windows 64 bits (GUI version will be easier but slower): http://membres-timc.imag.fr/Eric.Frichot/lfmm/software.htm
- Admixture can be found here for Linux and Mac OS: http://www.genetics.ucla.edu/software/admixture/download.html
For Windows users we recommend to install a virtual box using Ubuntu or to borrow a computer running Mac OS or Linux. You could also use STRUCTURE to obtain similar coefficients
- Virtual box: https://www.virtualbox.org/wiki/Downloads
- Ubuntu: http://www.ubuntu.com/download/desktop
- STRUCTURE: http://pritchardlab.stanford.edu/structure_software/release_versions/v2.3.4/html/structure.html
We will also use R for statistical analyses. You can install a modified GUI for R such as R studio
- R (Linux, Mac OS, Windows): http://www.r-project.org/
- R Studio (Linux, Mac OS, Windows): https://www.rstudio.com/ide/download/desktop
Following packages should be installed as well:
We transformed genetic data to PLINK format in both binary (BED) and ordinary (PED) format.
Data can be found here:
Sampling locations with aridity variables can be found here:
Several individuals have identical coordinates. In the purpose of visualizing them all in a GIS, we suggest modified coordinates.
- Loblolly Pine Visualization Coordinates
- Shapefile – Loblolly Pine Visualization Coordinates with admixture coefficients
We will use climatic variables from Worldclim datasets.
In the interest of time, we have created a subset for our study zone that you can download from our server : WorldClim_Subset.zip
Original datasets can be downloaded either by thiles or for the entire world:
DEMs can be found on Earth Explorer (subscription is mandatory before download):
- http://earthexplorer.usgs.gov/ We will use GTOPO30 only.
If you dont want to sign up on EarthExplorer, you can download the DEM here
Papers regarding practical work datasets can be found on Eckert’s blog:
- http://eckertdata.blogspot.ch/ at the date of 27 January 2012
We also recommend reading papers and documentation related to the software we will use
- Joost S., Bonin A., Bruford M.W., Despres L., Conord C., Erhardt G. & Taberlet P. (2007) A spatial analysis method (SAM) to detect candidate loci for selection: towards a landscape genomics approach to adaptation. Molecular Ecology 16, 3955-69.
- Frichot E., Schoville S.D., Bouchard G. & Francois O. (2013) Testing for Associations between Loci and Environmental Gradients Using Latent Factor Mixed Models. Molecular Biology and Evolution 30, 1687-99.
- LFMM tutorial