ATGC: BioNJ

BIONJ: an improved version of the NJ algorithm based on a simple model of sequence data.

Gascuel O. Molecular Biology and Evolution. 1997 14:685-695.

Please cite THESE papers if you use BIONJ.

Download BIONJ binaries

The package contains executable files for Linux, Mac OSX and Windows systems.

The distance matrix must be in PHYLIP square format. The program asks for the input and output file names. When the input file contains several matrices given one after the other, as obtained when combining SEQBOOT and DNADIST from PHYLIP to perform a bootstrap analysis, BIONJ returns the same number of trees, written one after the other in the output file; this file may be given to CONSENSE to obtain the bootstrap tree.

Sources

This is the C source code of BIONJ, compatible with Linux, Windows and PowerMac. It reads distance matrix(ces) in PHYLIP square format, and return tree(s) in standard Newick format.

Datasets

DATATEST is a small 9-taxon data set that you can use to test the program. It contains 3 replicated matrices, the first of which corresponds to Case's (1978, Syst. Zool. 27, 299) data. The other two are fictive; they are given in order to indicate the input and output formats, and to test the program.
TREETEST is the 3 trees that you should obtain with DATATEST. Note that the first tree (which corresponds to Case's data) is very different from the NJ tree (see Saitou and Nei, 1987), which is due to the fact that the data clearly do not conform to the molecular clock hypothesis. When the data conform to this hypothesis, BIONJ and NJ trees are usually close, at least when the number of taxa is low. Explanations are given in the MBE 1997 paper.