NeXML

NeXML is an XML-based exchange standard for phyloinformatic data, developed by Rutger A. Vos and collaborators and formally published in 2012. It was conceived as a modern successor to the venerable but loosely specified NEXUS format (Nexus), reusing the same conceptual "nexus" — a set of operational taxonomic units (OTUs) around which trees and character data are organized — while replacing NEXUS's block-and-token grammar with a validatable XML vocabulary. Its structure is defined by an XML Schema Definition, so documents can be checked syntactically for correct data types, cardinality, and identifier uniqueness before any downstream tool consumes them, addressing a long-standing source of silent parse errors and dialect drift in NEXUS files.

The data model centers on a small set of top-level containers. An otus block lists the taxa; one or more trees blocks hold trees (and networks) whose nodes reference OTUs by ID; and characters blocks carry state matrices, supporting both discrete (standard, DNA, RNA, protein, restriction) and continuous character types. Because every object — a tree, a node, an edge, a matrix cell, an OTU — carries a stable identifier, essentially any element can be annotated. NeXML uses RDFa-style meta annotations to attach ontology predicates and structured values (drawing on vocabularies such as Dublin Core and PRISM), which makes the data self-describing and linkable into the semantic-web ecosystem rather than confined to opaque comment blocks.

In the broader format-conversion landscape NeXML sits alongside other phylogenetic serializations: the compact bracketed Newick for topology alone, the XML-based phyloXML aimed at richly annotated trees, and the legacy Nexus it was designed to supplant. Tooling support is good within evolutionary biology — Perl's Bio::Phylo (the reference implementation), DendroPy, and Biopython read and write it, and the ete3 toolkit and R's RNeXML package provide further access, the latter bridging NeXML to the standard ape phylo object.

Its strengths are rigorous validation, extensibility without breaking backward compatibility, and first-class structured metadata — qualities that make it a durable archival and interchange format for comparative data. The trade-offs are the usual XML ones: files are considerably more verbose and less human-editable than Newick or NEXUS, and the richer model carries a steeper learning curve, so it is favored for data sharing, repositories, and provenance-sensitive workflows more than for quick interactive analysis.