Applications of anatomical ontologies in phylogenetics

RAMÍREZ, M. J.

Quilmes

Congreso; 1er Congreso Argentino de Bioinformática y Biología Computacional; 2010

Asociación Argentina de Bioinformática y Biología Computacional

Background The evolutionary relationships among living organisms are reconstructed using phylogenetic methods. All those methods use heritable traits as data (from various sources, such as molecules, anatomy, ultrastructure, or behavior), and interpret the variation observed in nature through hypothesized evolutionary transformations. The production of such data depends on detailed mapping of homologous elements. Phylogenetic data can be represented as a matrix with species as rows, characters as columns, and cells as observations. The problems (a) Phylogenetic analyses with multiple data sources and remote collaboration need complex documentation of biological terms, characters and observations; (b) traditional formats make such documentation difficult to reuse for further studies; (c) much of the raw data and documentation depends on images, which are hard to interpret automatically. Materials and methods This application was used for a phylogenetic study of all spiders using DNA sequences, morphology and behavior [1]. The morphological matrix has ca. 900 characters, 400 species, documented by 40,000 images, and is scored by an international team of 15 persons in 8 labs. The documentation of anatomical terms and the annotation of images rests on an ontology maintained in the OBO Foundry repository [2, 3]. The images are linked to characters and cells using the SILK [4] module of Mesquite [5, 6]. Results The Spider Ontology now contans 579 terms (plus 92 synonyms), accounting for virtually all morphological and ultrastructural character systems proposed for spiders. The image collection has 42,000 ontology annotations, used for automatic display of relevant images on the 360,000 matrix cells. The ontology takes much of the burden of documentation of anatomical terms and glossary, in a communitary environment, and is already usable for other projects. The images, their metadata and annotatios are deposited in a long-term repository for public access and maintenance [7]. Conclusions The ontology (a) allowed for standardization, documentation, and efficient retrieval of images; (b) makes much more efficient the production, maitenance and recycling of phylogenetic data; and (c), it builds a privileged channel of communication between communities working on diversity and on model organisms. References 1. Assembling the Tree of Life: Phylogeny of Spiders [http://research.amnh.org/atol/files/] 2. The Open Biological and Biomedical Ontologies [http://www.obofoundry.org/] 3. Spider Ontology [http://bioportal.bioontology.org/virtual/1091/] 4. Simple Image LinKing (SILK) [http://mesquiteproject.org/SILK/] 5. Mesquite: A modular system for evolutionary analysis [http://mesquiteproject.org/] 6. Syst. Biol. 56, 283–294 [doi:10.1080/10635150701313848] 7. Morphbank [http://www.morphbank.net/]