Summary of changes to the Plant Ontology Oct2010

Back to 2010 Release Notes Page

Organization and summary of changes to the Plant Ontology (PO)

January through September 2010

Curators (in alphabetical order): Cooper, Laurel D. (1); Elser, Justin (1); Gandolfo, Maria A. (2); Jaiswal, Pankaj (1), Mungall, Christopher (3); Smith, Barry (4); Stevenson, Dennis W. (5); Walls, Ramona L. (5)

1. Department of Botany and Plant Pathology, Oregon State University, Corvallis, OR 2. Department of Plant Biology, Cornell University, Ithaca, NY
 3. Lawrence Berkeley National Lab, Berkeley, CA 4. Department of Philosophy, University at Buffalo, NY 5. The New York Botanical Garden, Bronx, NY

Overview

The primary structure of the PO, including one branch for growth and developmental stages and one branch for plant structures, has not changed. During this round of revisions, there were minimal changes to the Plant Growth and Developmental Stage Ontology (PGDSO), with the addition of only three new high-level terms (see below).

The curators instead focused on re-organization of the Plant Structure Ontology (PSO). The primary purpose of this re-organization was to create a framework that could accept new terms from any green plant, be it an alga or an angiosperm. To this end, new top-level terms were added, new definitions were written for many terms (including all top-level terms), and the second an third levels of the ontology were extensively re-organized. An effort was made to insure that widely studied plant structures such as flower, stem, and gamete had working definitions and correct parentage.

Detailed revisions, including the addition of many new terms, were made to the PSO term tuber (PO:0004543) and its is_a and part_of children. This provided a test case for how build relationships among the parts of a complex structure while at the same time providing much needed terms for annotation of potato genes. Throughout the revision process, the curators worked to incorporate new advances in ontology design and construction. These included the use of logical text and relational definitions and references to external ontologies where appropriate.

Any new terms added by the PO curators during this round of revisions can be identified by term IDs in the range of 0025001 to 0030000.

Plant Growth and Developmental Stage Ontology (PGDSO)

The PGDSO had two main branches: plant structure developmental stages and whole plant growth stages. No changes were made to these branches.

A new top-level term was added: plant life cycle phase (PO:0028001). This term has two new children: sporophytic phase (PO:0028002) and gametophytic phase (PO:0028003). Structures from the PSO that occur in only one of these phases will have a participates_in relation to the appropriate phase.

Plant Structure Ontology (PSO)

Changes to the top level

In the previous version of the PSO, the top level was organized around the traditional characterization of plant anatomical structures (cell, tissue, organ, and whole plant), plus the classes gametophyte, sporophyte, and in vitro cultured cell, tissue and organ.

After extensive discussion, the terms gametophyte (PO:0009004) and sporophyte (PO:0009003) were subsumed within whole plant. This was achieved by making gametophyte and sporophyte obsolete, with a suggestion to consider whole plant. For future releases, plant structures that are specific to the gametophytic or sporophytic phase will have a participates_in relation to the appropriate phase. Click this link for a more detailed discussion of the decision to obsolete sporophyte and gametophyte.

The terms whole plant (PO:0000003) and plant cell (PO:0009002) were retained. Organ, tissue, and in vitro cultured cell, tissue and organ were also retained, but renamed plant organ (PO:0009008), portion of plant tissue (PO:0009007), and in vitro plant structure (PO: 0000004), respectively. Two new top-level terms were added: cardinal organ part (PO:0025001) and collective plant structure (PO:0025007). These new terms were necessary, because there are many commonly-referenced plant structures that are not a cell, organ, or tissue.

All top-level terms, including the root term plant structure (PO:0009011) were redefined, based on consultation of multiple plant anatomy books and consensus of the curators. All top-level terms except in vitro plant structure have a part_of relation to whole plant. It was decided that in vitro plant structures should remain separate from plant structures that occur in vivo, thus in vitro structures may have a derives_from relation to other plant structures, but not an is_a or part_of relation.

Discussions are underway to have all plant cell terms mirrored in the cell ontology (CL) and all in vitro plant structure terms mirrored in the Ontology for Biomedical Investigation (OBI).

New or redefined second and third level terms

There was considerable re-organization of the second and third levels of the ontology. Some of the re-organization was done to provide more structure to the ontology (to make it less flat and easier to navigate). For example, new classes of plant cells or tissues were added. Other changes were made to correct inconsistencies or inaccuracies in the ontology, such as lack of parentage, dual parentage, or incorrect parentage. To date, very few terms have been added that are specific to taxa outside angiosperms, but many second and third level have been structured so that it will be easier to incorporate these terms in future releases. Changes are summarized below by their ancestor term.

Descendants of cardinal organ part: This new class provided a home for many existing structures that did not have is_a parents (see below). Several new classes of cardinal organ part were added, including lamina (PO:0025060), organ margin (PO:0025005), and stalk (PO:0025066), but most terms are direct children of cardinal organ part.

Descendants of collective plant structure: This new class provided a home for many existing structures that did not have is_a parents (see below), plus the new term root system (PO:0025025). The existing term shoot was renamed shoot system (PO:0009006) and made a collective plant structure. The term collective phyllome structure (PO:0025023) was created to encompass phyllomes arranged in either rosettes or whorls, and includes structures like androecium, corolla, gynoecium, and collective leaf structure. The term collective leaf structure (PO:0025022) replaced the obsolete term leaf whorl (PO:0008034).

Descendants of plant cell: Several new children of plant cell were created to help organize cell types, including embryonic cell, ground tissue cell, laticifer cell, megagametophytic cell, and microgametophytic cell. Two important cell classes were added and defined: gamete (PO:0025006) and spore (PO:0025017).

Descendants of plant organ: Terms were reclassified so that plant organ now has only two is_a children: phyllome (PO:0006001) and plant axis (PO:0025004). A phyllome is a lateral plant organ produced by a shoot apical meristem, and includes all types of leaves, petals, sepals, sporophylls, etc. Plant axis is a new term that includes axial organs of the root system [root (PO:0009005)] and axial organs of the shoot system [the new class shoot axis (PO:0025029)]. Shoot axis is the parent of terms like stem, branch and rhizome.

Descendents of portion of plant tissue: There was extensive reorganization of plant tissue types, including many new definitions. Any terms that had the word ‘tissue’ in their name were prefixed with ‘portion of’, so, for example, vascular tissue became portion of vascular tissue.

Terms that did not have the word tissue in their name (like parenchyma or epidermis) were not changed. Many definitions have been reworded, so it is clear when a term is referring to any portion of tissue, a specific portion of tissue, or the maximal portion of tissue in a structure.

A new term portion of ground tissue (PO:0025059) was created, with the goal that most tissues could be classified as ground tissue, epidermis, or vascular tissue. Many existing tissue types were made children of portion of ground tissue, or its children, which include cortex, parenchyma, periderm and schlerenchyma.

The term dermal tissue (PO:0009014) was made obsolete, because it was redundant with epidermis (PO:0005679). While the bulk of plant tissue terms have been put into to the appropriate class, there are still a number of direct children of portion of plant tissue (e.g., chalaza, mesocarp and seed coat) that need more specific classification. This will be completed in the next release.

Descendants of whole plant:

The only is_a children of whole plant are embryo, female gametophyte, male gametophyte and seed. The term seedling was made obsolete and annotation for seedling should use whole plant and the PGDSO term seedling growth (PO:0007131).

All structures (except in vitro structures) have a part_of relations rooted in whole plant.

Progress toward is_a completeness

The number of classes without is_a parents was reduced from 200 to 53. In part this was accomplished by the addition of the new upper level terms cardinal organ part and collective plant structure. These classes provided parentage for terms like petiole, leaflet, shoot node, and stigma (cardinal organ parts) or terms like corolla, flower, fruit, inflorescence, infructescence, and shoot system (collective plant structures). Many of the remaining terms without is_a parents are immaterial entities, such as pores or grooves. All terms should have is_a parents by the following release of the PO.

Addition of terms for tubers

A new subset was created for potatoes, and a number of new terms were added to allow more specific annotations of potato structures. Most of these terms are portions of plant tissue and describe the specific parts of the tuber epidermis, tuber periderm and tuber flesh (=tuber storage parenchyma).

Changes to children of PO:0000055 bud

For a discussion of the options that were explored for how to order the child terms of bud, see the page Bud

Structures that are specific to the gametophyte or sporophyte phase

Although this release of the PSO retains much of its angiosperm bias, we feel that the upper-level ontology structure is now ready to accommodate terms from other plant groups. All of the top-level terms (plant cell, plant organ, portion of plant tissue, etc.) were re-defined precisely so that they could accept descendent terms from any group of plants, not just angiosperms. Rather than grouping lower-level terms by the taxa in which they occur, we chose to group them based on structural similarity. Although we have not added many lower-level terms that are specific to plants outside the angiosperms, we have included a few examples of how our schema would work. For example, the term axial cell (PO:0000081) has descendents hydroid (PO:0025032) and xylem element (PO:0000273). Hydroids, which occur in the gametophytes of bryophytes, would have a participates_in relation to gametophytic phase, while xylem elements, which occur in the sporophytes of vascular plants, would have a participates_in relation to sporophytic phase. Since we have not yet included the participates_in relation in this revision, these relations are currently described in the comments. Another example includes the children of leaf (PO:0025034): vascular leaf (PO:0009025) participates_in sporophytic phase, and non-vascular leaf (PO:0025075) participates_in gametophytic phase. Notice that by choosing to group terms by structural similarity, we are not required to make hypotheses about evolutionary homology (which usually is unknown). This allows researchers to compare gene expression in similar structures, in order to evaluate their own hypotheses.

General changes

All new and many existing textual definitions were written (or re-written) in the genus-differentia form, using language that is understandable for human readers but may also be interpretable by computer algorithms. For example, a shoot epidermis is: A portion of epidermis that is part of a shoot system. The relations described in the textual definitions have been made explicit, so that shoot epidermis has the relations: is_a epidermis, part_of shoot system. For future releases of the ontology, these relations will be written as cross-products (using intersection_of relations in the .obo format), allowing for improved reasoning and easier editing.

Whenever possible, upper-level terms were defined based on existing terms in other OBO Foundry ontologies, and external dbxrefs to those ontology terms were included. For example, plant structure is defined an anatomical structure that is or was part of a plant, or was derived from a part of a plant, with reference to CARO:0000003 (anatomical structure).

Future releases of the ontology will include cross-product definitions for such terms, that is, built in relations to other ontologies.

Many synonyms were reclassified as either exact, narrow, or broad, rather than the default value of related synonym.