In my current job I do two things.
1. Many of the seed samples in the major genebanks for crops have a description in a database which says where they come from. For instance, something like "5 km West of Buenos Aires". But many of these samples don't have geographical coordinates (latitude and longitude), which we would need if we wanted to map them. We use special software, Biogeomancer, to look up (conjure?) the coordinates. I supervise staff at IRRI who - with a lot of dedication - are going through the locality descriptions one by one. We have two major projects dealing with the CGIAR genebanks and the USDA/GRIN genebanks.
2. Why do we want to have all these data? The second thing I am doing is to develop new methods to analyze crop diversity from a geographical perspective. One of the things I am working on is to develop geographic diffusion models to predict genetic diversity. The map below depicts gene flow after domestication.
