Estimated phi

Version 1.02 of bamova was released on 27 September 2011. See Changelog below.

bamova implements a Bayesian Analysis of Molecular Variance and different likelihood models for three different types of molecular data (including two models for high throughput sequence data), as described in detail in Gompert and Buerkle (2011) and Gompert et al. (2010). Use of the software will require good familiarity with the models described in this paper. It will also likely require some programming to format data for input and to analyze the MCMC output.

Gompert, Z. and C. A. Buerkle. 2011. A hierarchical Bayesian model for next-generation population genomics. (article).
Gompert, Z., M. L. Forister, J. A. Fordyce, C. C. Nice, R. J. Williamson and C. A. Buerkle. 2010. Bayesian analysis of molecular variance in pyrosequences quantifies population genetic structure across the genome of Lycaeides butterflies. Molecular Ecology 19: 2455–2473. (article)

Download and installation

We distribute bamova as source code and as a compiled binary for Mac OS X. Each compressed archive contains a small sample data set and a software manual. The software manual can also be downloaded separately.

  1. bamova 1.02 source code
  2. bamova 1.02 binary for Mac OS X (10.6, Intel-based)
  3. bamova 1.0 software manual

bamova depends on the prior installation of the free and open-source GNU Scientific Library. Instructions for compiling the software from the source code are given in the manual.

Questions and feedback

We have tested bamova with many simulated and empirical data sets on Mac OS X and linux computers. If you encounter problems that you think are associated with an error in the code, please let us know. We welcome your feedback and will do what we can to assist you with the software.

If you have a question, a feature request, or encounter a bug, please consult the manual for bamova, the manuscript that describes the models and thoroughly investigate your concern or problem. If you are interested in a new feature, we encourage you to consider writing it and contributing it as a proposed change or addition. If our assistance is required, please send an email and provide as much information as possible. Over time we will likely build a FAQ.


  • 1.02 (27 September 2011) relative to 1.01
    1. fixed a bug associated with initializing phi-statistics with group structure (they could be initialized to NaN, not a number, with certain input data).
  • 1.01 (31 March 2011) relative to 1.0 (18 February 2011)
    1. The parameters used to initialize alpha and beta for genome-level phi distributions are now ensured always to be positive.
    2. All probabilities are checked to ensure they are non-zero (or finite on a log-scale). Zero probabilities are replaced with DBL_MIN from gsl.
    3. The calculation of the multivariate Polya pdf is now performed on the log-scale to retain numerical precision.
    4. Some typographicals errors in the output were fixed.

Terms of use

bamova is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

bamova is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

Grant support for work

This material is based upon work supported by the National Science Foundation under Grant No. 0701757 (Plant Genome Research Program). Any opinions, findings and conclusions or recomendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation (NSF).