Figure 1.

NEXUS and the character-state data model. Some relevant terms and concepts are illustrated with a graphical view of a small family of RAD23-related protein-coding genes (KOG0011 data provided in Additional file 1). In molecular sequence analyses, OTUs often are labelled with a token fusing a species name with a database ID. In a NEXUS file, such OTU labels are declared in a TAXA block. The TREES block may contain one or more trees relating the OTUs, each tree optionally having label (e.g., "Coelomate") and a numeric weight (e.g., a probability). Trees may contain branch lengths and support values. In the matrix of amino acid character data shown here, the 4^th character (i.e., the 4^th alignment column), has the states "V", "I", "I", "L", "V", "I", "L", and "I". Such character-state data are stored in a NEXUS CHARACTERS block, which defines the type of data and the meaning of a gap symbol such as "-". The ASSUMPTIONS block of a NEXUS file provides the means to store a weight or other numeric value for each character, such as the column-wise alignment scores shown here. Many other types of information not shown here can be stored in NEXUS files.