BAMBE

Bayesian Analysis in Molecular Biology and Evolution

Version 2.03 beta, May 2001

© Copyright 1998-2001 Donald Simon & Bret Larget, Department of Mathematics and Computer Science, Duquesne University.

Bayesian Analysis in Molecular Biology and Evolution (BAMBE) is a free software package for the Bayesian analysis of phylogenies. The package includes programs for analyzing aligned DNA or RNA sequence data and allows data sets with gaps or indeterminate sites. The main program uses a variety of Metropolis-Hastings algorithms to sample from the joint posterior distribution of phylogenetic trees and likelihood model parameters. Other programs in the distribution help in analysis of the sampled trees and parameter values.

Bret Larget and Donald L. Simon of the Department of Mathematics and Computer Science at Duquesne University are the developers of the software.

Software Availability

UNIX

bambe-2.03b.tar.gz source code for UNIX

BAMBE 2.02 beta for UNIX fixes numerical problems that affected some DEC alpha clones and perhaps other machines as well.

bambe-2.02b.tar.gz source code for UNIX

BAMBE 2.01 beta for UNIX is written in C++ and is more efficient than version 1.

bambe-2.01b.tar.gz source code for UNIX

The programs in previous BAMBE packages are written in ANSI C which we continue to make available. We are distributing the sourcecode which you must compile.

bambe-1.02b.tar.gz (contains a few small changes to eliminate warning messages with some compilers)

bambe-1.01b.tar.gz (released in October, 1998)

Windows

Bambe-V2.01beta.zip compiled for Windows

We have tested the algorithms extensively, but have not attempted to compile the code or run the program with many platforms/compilers. If you have trouble compiling, please send us a detailed message. Make sure to include the type of computer, the name of the the compiler, and the error message when compilation failed.

After downloading the software, please register so we can keep you updated about new releases of the software.

When you have experimented with the software, we would appreciate hearing your comments on the features of the programs, ease of use, clarity of documentation, and so on, so we may use your input to create a better product.

A Windows version with a graphical user interface is under development. Macintosh and Java versions are planned for the future.

Documentation

Features

• A choice of likelihood models.
  • HKY85 (Hasegawa-Kishino-Yano, 1985)
  • F84 (as implemented in Felsenstein's PHYLIP)
  • TN93 (Tamura-Nei, 1993)
• A choice of Metropolis-Hastings samplers for proposing new phylogenetic trees (including a new scaled Beta distribution version of local).
• Metropolis-Hastings samplers for proposing new model parameters.
• The choice of separate samplers for burn-in and inference.
• Flexible specification of site categories with separate parameter values for each site category.
• A choice to initialize the search at a random tree, the UPGMA tree, the Neighbor-joining tree, or a user-defined tree.
• Two formats (BAMBE and Newick) for reading and writing trees.
• HTML Help files.

The Windows version includes:

• Graphical real-time monitoring of the tree and model parameters.

Planned features in future releases

• More general likelihood models.
• The discrete-gamma model for site heterogeneity.
• Maximum likelihood.
• Inference from other data types including aligned amino acid sequences, codon sequences, and restriction site data.