I am pleased to annouce the first version of Rejector2, an overhaul of Rejector, is now available for download. Rejector2 contains the following changes from the original Rejector software:

• Rejector, Rejstats and Prior are now one executable
  • REJECTOR2 now performs all the funcitons of the above three programs from the original REJECTOR release
  • To access the separate summary statistic calculation funcitons of Rejstats, invoke REJECTOR2 with the –rs switch
  • The only other executable you need is your coalescent simulation
• REJECTOR2 works with the coalescent simulator msHOT (SNPs only)
  • The simulation program msHOT can now be used with REJECTOR2 (see page 9)
• REJECTOR2 can now handle much larger subsets of data
  • The data model of REJECTOR2 has been rewritten to handle much larger numbers of loci
  • NOTE: While REJECTOR2 can deal with 500,000 or more loci, be sure that the simulation you use can also handle this much information
    • msHOT can quickly simulate hundreds of thousands of SNPs, but do not work with STRs
    • msHOT is designed to work with segments of DNA, but REJECTOR2 does not yet know how to pass this information (should be added soon)
    • Simcoal2 can work with STRs, but in my tests cannot handle more than a few thousand loci
• REJECTOR2 does NOT work with multiple CPUs or over Xgrid (yet)

I have also updated the original Rejector package, including support for altering the transition fraction and recombination rate, as well as the addition of a new statistic.

Rejector is a software package comprised of four separate executables designed to act in concert for the purposes of parameter value estimation and comparison of alternate models of population history via rejection-based approximate Bayesian inference. This inference method involves the calculation of summary statistics from experimental data, and then the calculation of those same statistics from numerous randomized replications of the data generated from a user-specified model. Comparison of the simulated summary statistics with those generated from the experimental data will indicate the probability that the parameter values used to generate the simulations are close to the real conditions that created the experimental data.

Please consult the Rejector User Guide for a detailed description of the method, examples of use and suggested applications.

Rejector can be compiled and run under Mac OS X, Windows, and Linux. For installation instructions, see the Rejector User Guide, or if using the binaries, just make sure all binaries are in the same folder. Please read the guide before using Rejector! You can view the changelogs for the software here.

Rejector uses SIMCOAL2 for coalescent simulation. All references to Rejector should also include the following reference:

Laval, G. and L. Excoffier (2004). "SIMCOAL 2.0: a program to simulate genomic diversity over large recombining regions in a subdivided population with a complex history." Bioinformatics (Oxford) 20(15): 2485.

The Rejector package, apart from SIMCOAL2, was written by Matthew Jobin at the Department of Anthropology, Stanford University, Stanford, CA.