gsi_sim is a program written in C. It can be used to assess what sort of accuracy can be expected for genetic stock identification given a genetic baseline. Its primary advantage over programs like GMA, and SPAM is that the source code is available and it can be compiled for command line use on Unix-like systems—you are not forced to use it with a GUI interface in some Windoze environment. It also implements the essentially unbiased leave-one-out cross-validation method developed by Anderson, E.C., Waples, R. S. and Kalinowski, S. T. (2008) An improved method for predict-
ing the accuracy of genetic stock identiﬁcation. Canadian Journal of Fisheries and Aquatic
Sciences 65:1475–1486. My co-author, Steven Kalinowski, has also made available some Windows software for performing similar calculations. This may be better for some users. It is available from http://www.montana.edu/kalinowski/Software/ONCOR.htm
The main disadvantage of
gsi_sim is that if you don't know your way around the Unix command line, and you aren't familiar with such utilities as
perl, then it is going to be hard to get what you want out of the output. If you are familiar with such utilities, then the voluminous amount of program output can be flexibly harnessed to answer many different questions.
gsi_sim was originally conceived for assessing power given a baseline, and not for analyzing real mixtures given that baseline. As a result, full functionality is not yet available for that. Although
gsi_sim can be used to analyze a real mixture, it does not yet have any sort of bootstrapping or jackknifing implemented to assess the precision of such estimates. This might come along at some point, however, I would recommend that you use the package cBayes available at http://www.pac.dfo-mpo.gc.ca/sci/mgl/Cbayes_e.htm.
What is available here is something of a pre-release of
gsi_sim. It is all mostly functional, and it has been extensively tested; the documentation that comes with the program describes the options fairly completely, but it may not be entirely clear for the uninitiated how to proceed. Therefore I am putting together a tutorial that provides some examples of its use. Be sure to download that tutorial from the link below.