What is GARM?
We are currently working on GARM, a Genetic Algorithm for creation of Resistance Maps. The main motivation behind GARM is to begin to help remove the influence of ‘expert opinion’ on landscape resistance maps and allow users to input large numbers of environmental layers.
GARM takes as inputs a wide range of environmental layers representing landscape features (e.g. elevation, forest cover, landuse type, etc.) and individual locations with observed genetic information (see Figure below). GARM works by adjusting the weights for assigned classes of each environmental layer, while searching the solution domain for the resistance map which best matches the empirical genetic structure. The fitness function is the Mantel correlation between genetic distances and landscape distances (partial and simple). These landscape distances are calculated each generation using Dijkstra’s shortest path algorithm in the UNICOR program. GARM converges upon the optimal resistance map, which is a weighted combination of the user’s initial environmental data.

Solving Problems with GARM:
The aim of GARM is to facilitate a more inclusive, less-biased, or “expert opinion” driven process for the creation of resistance maps, as well as to provide a promising exploration and optimization tool for landscape genetic studies. Conservation professionals (e.g. land managers), researchers and students can use GARM to model functional connectivity in their system.
Example Simulation:
Disclaimer:The software is in the public domain, and the recipient may not assert any proprietary rights thereto nor represent it to anyone as other than a University of Montana-produced program (version 1.x). GARM is provided "as is" without warranty of any kind, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. The user assumes all responsibility for the the accuracy and suitability of this program for a specific application. In no event will the authors or the University be liable for any damages, including lost profits, lost savings, or other incidental or consequential damages arising form the use of or the inability to use this program.
We strongly urge you to read the entire documentation before ever running GARM. We wish to remind users that we are not in the commercial software marketing business. We are scientists who recognized the need for a tool like GARM to assist us in our research on landscape genetic issues. Therefore, we do not wish to spend a great deal of time consulting on trivial matters concerning the use of GARM. However, we do recognize an obligation to provide some level of information support. Of course, we welcome and encourage your criticisms and suggestions about the program at all times. We will welcome questions about how to run GARM or interpret the output only after you have read the entire documentation. This is only fair and will eliminate many trivial questions. Finally, we are always interested in learning about how others have applied GARM in ecological investigation and management application. Therefore, we encourage you to contact us and describe your application after using GARM.
We hope that GARM is of great assistance in your work and we look forward to hearing about your applications.
Download:
GARM is available for download from the downloads page
Acknowledgements:
This program was developed by Brian Hand.· The reference to cite is:
Computational Ecology Laboratory
Division of Biological Sciences (DBS)
The University of Montana
32 Campus Drive, HS 507
Missoula MT, 59812-1002
Phone: (406) 243-2393
Fax: (406) 243-4184