=Paper=
{{Paper
|id=None
|storemode=property
|title=A plant disease extension of the Infectious Disease Ontology
|pdfUrl=https://ceur-ws.org/Vol-897/session1-paper01.pdf
|volume=Vol-897
|dblpUrl=https://dblp.org/rec/conf/icbo/WallsSEGSJ12
}}
==A plant disease extension of the Infectious Disease Ontology==
https://ceur-ws.org/Vol-897/session1-paper01.pdf
A plant disease extension of the Infectious Disease Ontology
Ramona Walls1, Barry Smith2, Justin Elser3, Albert Goldfain4
Dennis W. Stevenson1, Pankaj Jaiswal3
1.
New
York
Botanical
Garden,
Bronx,
NY,
USA,
2.
Department
of
Philosophy,
University
at
Buffalo,
Buffalo,
NY,
USA,
3.
Department
of
Botany
and
Plant
Pathology,
Oregon
State
University,
Corvallis,
OR,
USA,
4.
Computer
Science
Department,
Blue
Highway,
Inc.,
Syracuse,
NY,
USA
ABSTRACT Tester, 2011). Complete genome sequences already exist for
Plants
from
a
handful
of
species
provide
the
primary
source
of
food
for
25 green plant species, of which 17 are agriculturally im-
all
people,
yet
this
source
is
vulnerable
to
multiple
stressors,
such
as
disease,
drought,
and
nutrient
deficiency.
With
rapid
population
portant (Anon, 2012), along with expression sequence tags
growth
and
climate
uncertainty,
the
need
to
produce
crops
that
can
(EST), unigene, mutant phenotype, and other data sets for
tolerate
or
resist
plant
stressors
is
more
crucial
than
ever.
Traditional
hundreds of plant species. Additionally, a vast quantity of
plant
breeding
methods
may
not
be
sufficient
to
overcome
this
chal-‐ information on plant diseases is available in resources like
lenge,
and
methods
such
as
high-‐throughput
sequencing
and
automat-‐
manuals, textbooks, extension program highlights, and crop
ed
scoring
of
phenotypes
can
provide
significant
new
insights.
Ontolo-‐
gies
are
essential
tools
for
accessing
and
analysing
the
large
quantities
management databases, but almost always in natural lan-
of
data
that
come
with
these
newer
methods.
As
part
of
a
larger
project
guage form. Access to and analysis of the growing quanti-
to
develop
ontologies
that
describe
plant
phenotypes
and
stresses,
we
ties of genomic, phenomic, and free-text data can be greatly
are
developing
a
plant
disease
extension
of
the
Infectious
Disease
On-‐ facilitated when data are annotated using ontologies. The
tology
(IDOPlant).
The
IDOPlant
is
envisioned
as
a
reference
ontology
designed
to
cover
any
plant
infectious
disease.
In
addition
to
novel
development of ontologies can also foster consistency in the
terms
for
infectious
diseases,
IDOPlant
includes
terms
imported
from
description of plant diseases, including aspects such as envi-
other
ontologies
that
describe
plants,
pathogens,
and
vectors,
the
geo-‐ ronmental factors, areas of endemism, phenotypes associat-
graphic
location
and
ecology
of
diseases
and
hosts,
and
molecular
func-‐ ed with diseases, and development stages of both plants and
tions
and
interactions
of
hosts
and
pathogens.
To
encompass
this
range
of
data,
we
are
suggesting
in-‐house
ontology
development
comple-‐
pathogens. Finally, the standardization and reasoning power
mented
with
reuse
of
terms
from
orthogonal
ontologies
developed
as
provided by using ontologies enhances data sharing among
part
of
the
Open
Biomedical
Ontologies
(OBO)
Foundry.
The
study
of
biomedical researchers, allowing the results of research in
plant
diseases
provides
an
example
of
how
an
ontological
framework
plant pathology to be translated into applications for human
can
be
used
to
model
complex
biological
phenomena
such
as
plant
or other animal diseases, and vice versa.
disease,
and
how
plant
infectious
diseases
differ
from,
and
are
similar
to,
infectious
diseases
in
other
organism.
A plant disease is traditionally defined as a deviation
from normal physiological functioning that is harmful to a
1 INTRODUCTION plant (Manners, 1993). Biotic factors or stressors such as
pests or pathogens and abiotic factors such as low tempera-
Plants are the primary food source on which almost every
ture, air pollution, or nutrient deficiency, may cause plant
other organism on earth depends, either directly or indirect- diseases. Infectious plant diseases are caused by pathogens,
ly, and six plant species – wheat, rice, corn, potato, sweet such as fungi, bacteria, and viruses. As part of a larger pro-
potato, and cassava – provide 80% of calories consumed by ject to develop ontologies that describe both biotic and abi-
humans worldwide (FAO, 2012; Goudie & Cuff, 2001). It is otic plant stresses, we are developing a plant disease exten-
imperative to develop higher-yielding crop varieties to sup- sion (IDOPlant) of the Infectious Disease Ontology (IDO)
port the growing human population. This can be done in two
(Cowell & Smith, 2010) as a reference ontology for plant
primary ways, (1) by increasing, e.g., the number or size of
disease. The goals are to provide plant scientists with the
grains on a cereal plant or tubers on a potato plant, and (2)
means to identify genomic and genetic signatures of host-
by reducing losses due to diseases and pests. Pre-harvest pathogen interactions, resistance, or susceptibility, and to
disease and pest damage in the eight most important food help agronomists and farmers by developing tools to identi-
and cash crops in the world account for ~42% of attainable fy disease phenotypes and gather epidemiological statistics.
production, and infectious plant diseases also threaten plant
IDOPlant will integrate and interoperate with member
conservation and human health (Anderson et al., 2004).
and candidate ontologies of the Open Biomedical Ontolo-
Many challenges in plant pathology (the study of plant
gies (OBO) Foundry (Table 1), such as: the Plant Ontology
diseases) can potentially be met through advances in meth-
ods such as high-throughput sequencing and automated (PO; describes the plant structures and the development
scoring of phenotypes (Studholme et al., 2011; Furbank & stages at which infections happen or signs of disease are
observed), the Plant Trait Ontology (TO; describes pheno-
types or entities that are evaluated in plants, such as leaf
* To whom correspondence should be addressed: color or grain yield), and the Gene Ontology (GO; describes
jaiswalp@science.oregonstate.edu
1
Walls et al.
ID Ontology Name Domain
2 METHODS
PO Plant Ontology1 plant structures and development
stages Throughout this paper, words in italics are ontology terms,
GO Gene Ontology2 biological processes, sub-cellular e.g., pathogen. If the source ontology is not evident from
components, molecular function the context, then we prefix with the ontology ID, as in:
TO Trait Ontology3 plant traits IDO:pathogen. The IDOPlant and the Plant Phenotype and
PATO Phenotypic Quality On- phenotypic qualities
tology4
Stress Ontology are being constructed in web ontology
OBI Ontology for Biomedical protocols, instrumentation, materials, language (OWL), using the Protégé 4.1 software
Investigations5 data, types of analysis (http://protege.stanford.edu) The annotation standard for-
ENVO Environment Ontology6 environmental features and habitats mat will follow the GO and PO model with the GAF2.0
ChEBI Chemical Entities of chemical entities format (Gene Ontology, 2012).
Biological Interest7
GAZ Gazetteer8 geographical information
We began by reviewing whether the terms in the IDO
were adequately structured for describing plant infectious
NCBItaxon NCBI Taxonomy Classi- taxonomic classification of living
fication9 organisms
diseases, including discussion with the developers of IDO
Table 1. Some of the external ontologies needed to describe plant diseas- and the Ontology for General Medical Science (OGMS;
es. References: 1. Ilic et al., 2007, 2. Gene Ontology Consortium, 2010, 3. http://code.google.com/p/ogms/). Next, following the strat-
Jaiswal, 2011, 4. Mungall et al., 2010, 5. http://obi-ontology.org/ egy used in other IDO extensions, we created terms for the
page/Main_Page, 6. http://http://www.environmentontolo gy.org/, 7.
Degtyarenko et al. 2007, 8. http://bioportal.bioontolgy.org/ontologies IDOPlant, such as plant infectious disease, specific to the
/1397, 9 http://www.obofoundry.org/wiki/index.php/NCBI Taxon: needs of plant pathology. More specific terms, such as rice
Main_Page. bacterial blight disease were added as an example of how
the molecular functions of interacting genes from host and to model a specific disease and to provide terms to be used
pathogen as well as biological processes involving either in annotating existing gene expression data available
host, pathogen, or both). The multi-organism process branch through Gramene (http://www.gramene.org). Textual defini-
of the GO, developed as part of the PAMGO project (Giglio tions and relationships among terms are drawn from plant
et al., 2009), is especially relevant to the IDOPlant. To en- pathology textbooks or journal articles, and are reviewed by
sure compatibility with research on non–plant diseases, the plant disease experts.
IDOPlant is created by downward population from the up- Logical definitions for IDOPlant terms are being con-
per-level terms of the IDO. The IDOPlant differs from exist- structed in OWL. Many of the terms needed for logical def-
ing IDO extensions, because the latter focus on specific dis- initions already exist in other ontologies. To access these
eases or pathogens, like Malaria or Brucellosis, that affect terms, we could import entire ontologies into the IDOPlant,
human or other animal health (Lin et al., 2011; Topalis et but this would result in many unnecessary terms and may
al., 2010). IDOPlant, in contrast, is designed to encompass cause problems if the resultant ontology is too large. Import-
any plant infectious disease. Furthermore, the IDOPlant is ing a selected subset of terms creates problems as well. If
being developed as part of the larger Plant Phenotype and we import individual terms from external ontologies, then
Stress Ontology Project, which is not limited to infectious we lose the ontology structure they are associated with and
diseases but encompasses any plant stress. Our approach the reasoning power that comes with it. If we import select-
calls for a multi-pronged strategy that includes creating new ed terms through the MIREOT process (Courtot et al.,
terms, as well as importing terms from, and building links 2009), which imports the minimum information to reference
to, other ontologies. an external ontology, we have to update the IDOPlant
The study of plant diseases provides an excellent ex- whenever there is a change to the source ontology.
ample of how the framework of the OBO Foundry (Smith et To cope with these issues, we use a multi-pronged
al. 2007) allows us to describe complex biological phenom- strategy that includes directly importing some terms and
ena using terms from multiple ontologies. By constructing building bridge files to link to external ontologies.
the IDOPlant within the OBO framework, we eliminate re- • Terms specific to plant diseases are added to the
dundant efforts, have a head start in ontology term devel- IDOPlant and assigned unique IDOPlant IDs, e.g.,
opment, and yield outcomes compatible with databases that IDOPlant:#######.
already annotate their data using OBO Foundry ontologies, • Terms falling near the bottom of the IDOPlant hierarchy
such as the Arabidopsis Information Resource (TAIR) that are drawn from ontologies from which only a few
(Swarbreck et al., 2008), Gramene (Youens-Clark et al., terms are needed are imported as single terms, using the
2011; Jaiswal, 2011) and Uniprot (The UniProt Consortium, original ontology ID. When appropriate, the MIREOT
2010). In this paper, we describe our plans for the overall method is used.
structure of the IDOPlant, provide an example of how to • Content treated in ontologies from which many terms are
model plant disease data, and discuss the types of data that needed are accessed by simultaneously loading multiple
can be annotated with the IDOPlant. ontologies and creating relations among them using
2
A plant disease extension of the Infectious Disease Ontology
bridge files (Mungall et al., 2010). This applies specifi- The IDO consists primarily of terms specific to infec-
cally to the three main ontologies (PO, TO, and PATO) tious disease, together with relevant terms imported from
whose terms are needed to describe plant stresses. Users other ontologies, such as organism from OBI; disease, dis-
will be required to open the entire suite of these ontolo- order, and disease course from OGMS; habitat from
gies when annotating data with the IDOPlant. ENVO; macromolecular complex, reproduction, and entry
• Taxonomic entities require special treatment, because we into host from GO; bacterium and virus from NCBItaxon;
will ultimately need to import many terms for plant spe- and molecular entity from ChEBI. The IDO has created
cies, disease organisms, and vector species. However, the many new terms, such as resistance to drug, infectious
NCBItaxon ontology is very large and can be impractical agent, and infectious disease epidemic. The bulk of the
to work with when loaded. Therefore, we will manually unique IDO terms can be used for the IDOPlant without
import the necessary taxa into the IDOPlant from either modification. For example IDO:infectious disease is defined
NCBItaxon or uBio (http://ubio.org). as “A disease whose physical basis is an infectious disor-
• In the event that a term imported into the IDOPlant is der”. This in turn is based on the OGMS definition of dis-
made obsolete in the source ontology, we will replace the ease: “A disposition (i) to undergo pathological processes
term either with the term suggested by the source ontolo- that (ii) exists in an organism because of one or more disor-
gy or with a new term created for the IDOPlant.
ders in that organism” (Scheuermann et al., 2009). Although
the wording of definitions such as this may be unfamiliar to
3 RESULTS AND DISCUSSION plant pathologists, the meaning is consistent with traditional
Researchers should contact the curators before using the treatments of plant disease (e.g., Manners, 1993).
IDOPlant, because it is under active development. A draft is IDO terms such as transition to clinical abnormality
available at http://purl.obolibrary.org/obo/idoplant.owl. or subclinical infection required careful consideration, be-
3.1 Using IDO for plant infectious diseases cause the word “clinical” is not commonly used for plants.
We decided that the meaning of their definitions was appro-
Our review of the IDO suggests that it is generally appropri- priate for plants, despite the names. For example, a feature
ate as a foundation for the description of plant diseases. The of an organism is clinically abnormal when it: “(1) is not
IDO is rooted in the Basic Formal Ontology (BFO) (Arp & part of the life plan for an organism of the relevant type …
Smith, 2008) and in the OGMS, which increases compatibil- (2) is causally linked to an elevated risk either of pain or
ity with other OBO Foundry ontologies and helps to ensure other feelings of illness, or of death or dysfunction, and (3)
logically consistent use of type-subtype relations. For ex- is such that the elevated risk exceeds a certain threshold
ample, IDO:pathogen cannot be classified as a subtype of level” (Scheuermann et al., 2009). All three conditions can
IDO:process of establishing an infection, because they be- be met in plants. Although we cannot know if plants experi-
long to disjoint super-classes (BFO:continuant and ence pain or feelings of illness, we can assess death or dys-
BFO:occurrent, respectively). function in plants.
Another potential limitation
of the IDO for use in plant science
is the meaning of terms from the
OGMS that were defined within
the scope of clinical encounters
involving humans. In particular,
the definition of symptom from
OGMS requires a host of a type
that can report its experiences, and
so is restricted to sentient hosts. In
plant pathology, “symptom” is
commonly used to describe the
phenotypes that are associated
with a plant disease. The pheno-
types are independent of the dis-
ease and the same phenotype or
“symptom” may be associated
with many different diseases. Fur-
Fig. 1.
Selected terms from the upper-level type-subtype hierarchy of the IDOPlant, with top-level terms thermore, diagnosis generally de-
imported form the IDO (including terms imported to the IDO from BFO and other ontologies) and lower- pends on a collection of pheno-
level terms that were added as part of the IDOPlant (in bold). All arrows represent is_a relations. Dashed types, and not every instance of a
arrows indicate several skipped intermediate steps in the ontology hierarchy.
3
Walls et al.
disease will display every phenotype that is typical of the the material basis of an infectious disease, e.g., rice
disease. Rather than use the OGMS definition of symptom, bacterial leaf blight disease has_infectious_agent Xan-
we developed a new term for the IDOPlant: thamonas oryzae.
plant disease symptom =def. A feature of a plant that is In addition we are developing the following for IDOPlant:
of the type that can be hypothesized to be involved in has_plant_disease_symptom: This relation is used to indi-
the realization of a plant disease. cate a phenotype, process, or independent continuant
Comment: Features include phenotypes such pale yel- that is evaluated to diagnose a disease. For example,
low leaf color, processes such as sudden wilting, and “rice bacterial leaf blight disease has_plant_disease_
independent continuants such as leaf lesion. symptom leaf color pale yellow” means that pale yel-
The terms plant disease symptoms already exist in other low leaf color is a plant disease symptom (see above)
ontologies (primarily the TO), and will be linked to plant of rice bacterial leaf blight disease, but it does not
diseases using the relation has_plant_disease_symptom (see mean that every instance of rice bacterial leaf blight
section 3.3). disease has pale yellow leaves.
3.2 Terms from external ontologies 3.4 Modeling disease in the IDOPlant
The study of plant diseases encompasses many do- Much of the information available on plant diseases is in a
mains. In addition to IDO terms common to all infectious natural language or free text form, such as: “Bacterial leaf
diseases, like pathogen or resistance, the IDOPlant needs blight disease of rice is caused by Xanthomonas oryzae. It
terms to describe the taxonomy of host plants, pathogens, produces pale yellow leaves in mature plants. In one report
and vectors, genomic and genetic data, the geographic loca- the pathogen and the disease were reported in the Northern
tion and ecology of diseases and hosts, plant and fungal Territory of Australia.” Using ontologies to process such
anatomy, plant and pathogen development, biological pro- descriptions in a standardized form makes them comprehen-
cesses, and molecular functions. To encompass this range, sible to computers and reasoners. For example, the descrip-
the IDOPlant is not only creating new ontology terms spe- tion above could be converted (using natural language pro-
cific to its domain, but also integrating and linking to exist- cessors or other mechanisms) to:
ing terms from multiple sources (Table 1). Whenever possi- disease: rice bacterial leaf blight dis-
ble, existing ontology terms are being used to create logical ease | host species: Oryza sativa (rice)
| caused by: Xanthomonas oryzae | has
definitions for IDOPlant terms. For example, rice bacterial symptom: pale yellow leaves | reported
leaf blight is defined as “A bacterial blight disease (in in: Northern Territory
IDOPlant), that has as infectious agent Xanthomonas oryzae This standardized text could then be made even more pow-
(from NCBItaxon)” (fig. 3). Terms from external ontologies erful using ontology terms and relations (Fig. 3).
will also be used for relations such as rice bacterial leaf
blight disease has_plant_disease_symptom pale yellow leaf 3.5 Integrating data into the IDOPlant
color (from TO). Logical definitions allow us to use auto- The current situation in the plant disease research communi-
mated reasoners to ensure that the ontology hierarchy is ty is similar to that in the animal community when the GO,
sound and to infer sub-types and relations implied by the MeSH (Savage, 2000), and CARO (Haendel et al., 2007)
definitions. These can then be added to the ontology
if correct or eliminated if incorrect or redundant
(Meehan et al., 2011).
3.3 IDOPlant relations
The IDO imports the Relation Ontology (RO)
(Smith et al., 2005), which includes basic relations
such as SubClassOf (is_a), part_of, participates_in,
inheres_in, bearer_of, has_disposition, has_role,
and has_function. In addition, we plan to incorpo-
rate several new relations:
has_material_basis: This relation is under devel-
opment by the OGMS and will be added to the
BFO. It is used to indicate the material basis of
a disease. For infectious diseases, we use the
has_infectious_agent relation. Fig. 3.
Some of the terms and relations needed to model rice bacterial blight disease
in the IDOPlant. Following the IDO, a disease is treated as a disposition of an infected
has_infectious_agent: This relation, which is under organism, which has a particular infectious agent. The IDOPlant can also be used to
consideration by the IDO, is used to indicate define terms in the TO, such as rice bacterial blight disease resistance, which is a
resistance to infectious disease that inheres in Oryza sativa.
4
A plant disease extension of the Infectious Disease Ontology
projects were being initiated: A number of organism- REFERENCES
specific databases are faced with large amounts of molecu- Anderson, P.K. et al., 2004. Emerging infectious diseases of plants: patho-
lar, germplasm (stock), genotype, and phenotype data asso- gen pollution, climate change and agrotechnology drivers. Trends in
ciated with function, phenotype, or environment. The shar- Ecology & Evolution, 19(10), pp.535–544.
Anon, 2012. Phytozome v7.0. Available at: http://www.phytozome.net/.
ing of the task of building a set of controlled vocabularies
Arp, R. & Smith, B., 2008. Function, role, and disposition in Basic Formal
such as GO and PO has helped enormously to address the Ontology. Nature Precedings. Available at: http://hdl.handle.net
needs of multiple individual databases. The IDOPlant con- /10101/ npre.2008.1941.1.
trolled vocabulary for plant infectious diseases will allow Courtot, M. et al., 2009. MIREOT: the Minimum Information to Reference
an External Ontology Term. Nature Precedings, (713). Available at:
database curators to store and retrieve the results of experi-
http://precedings.nature.com/documents/3574/version/1.
ments related to diseases, including quantitative trait loci, Cowell, L. & Smith, B., 2010. Infectious Disease Ontology. In Infectious
pathogen and host germplasm descriptions, microarray ex- Disease Informatics. New York, NY: Springer, pp. 373–395.
pression studies, gene knockouts, reporter gene expression Degtyarenko et al. 2007. ChEBI: a database and ontology for chemical
patterns, and gene-gene interactions from host and patho- entities of biological interest. Nuc Acids Research 36, pp.D344-D350.
FAO, 2012. FAOSTAT, Food and Agricultural Organization of the United
gen. The Plant Phenotype and Stress Ontology Project aims Nations. FAOSTAT. Available at: http://faostat.fao.org/default.aspx .
to overcome the obstacles in annotating data for complex Furbank, R.T. & Tester, M., 2011. Phenomics – technologies to relieve the
biological concepts that span multiple ontologies by devel- phenotyping bottleneck. Trends in Plant Science, 16(12), pp.635–644.
The Gene Ontology Consortium, 2010. The Gene Ontology in 2010: exten-
oping both the ontology terms and the software tools needed
sions and refinements. Nuc Acids Research 38, pp.D331-D335.
to annotate data from all aspects of plant diseases. Gene Ontology, 2012. GO Annotation File Format Guide. Available at:
To annotate plant infectious disease description data, http://www.geneontology.org/GO.format.annotation.shtml.
the IDOPlant is reaching out to resources such as the Food Giglio, M.G. et al., 2009. Applying the Gene Ontology in microbial annota-
and Agriculture Administration’s AGROVOC tion. Trends in Microbiology, 17(7), pp.262–268.
Goudie, A. & Cuff, D.J., 2001. Encyclopedia of global change: environ-
(http://aims.fao.org/website/AGROVOC-Thesaurus/sub) mental change and human society, Oxford University Press.
and the International Rice Research Institute Haendel, M. et al., 2007. CARO - The Common Anatomy Reference On-
(http://www.knowledgebank.irri.org/rice.htm). These re- tology. In Anatomy Ontologies for Bioinformatics: Principles and
sources will enrich the IDOPlant by providing a wealth of Practice. Computational Biology Series. Springer.
Ilic, K., et al. 2007. The Plant Structure Ontology, a Unified Vocabulary of
information that can be incorporated into the ontology and Anatomy and Morphology of a Flowering Plant. Plant Physiology:
by identifying gaps and errors. The IDOPlant can benefit 143(2), pp.587-599
these organizations by making their content more easily Jaiswal, P., 2011. Gramene Database: A Hub for Comparative Plant Ge-
accessible to semantic applications. nomics. In A. Pereira, ed. Plant Reverse Genetics. Totowa, NJ: Hu-
mana Press, pp. 247–275.
Lin, Y., Xiang, Z. & He, Y., 2011. Brucellosis Ontology (IDOBRU) as an
4 CONCLUSIONS extension of the Infectious Disease Ontology. Journal of Biomedical
Semantics, 2(1), p.9.
As the growing human population and climate change place Manners, J.G., 1993. Principles of Plant Pathology, Cambridge, U.K.:
even more uncertainty on food supply, the need to under- Cambridge University Press.
stand the linkages between plant disease, environment, and Meehan, T. et al., 2011. Logical development of the Cell Ontology. BMC
Bioinformatics, 12(1), p.6.
yield is greater than ever. The IDOPlant and the Plant Phe- Mungall, C. et al., 2010. Integrating phenotype ontologies across multiple
notype and Stress Ontology Project can contribute to this species. Genome Biology, 11(1), p.R2.
challenge by making data on plant diseases more accessible. Savage, A., 2000. Changes in MeSH Data Structure, NLM Technical Bul-
We are taking advantage of the interoperability of OBO letin. Available at:
http://www.nlm.nih.gov/pubs/techbull/ma00/ma00_mesh.html.
Foundry ontologies to leverage existing terms to enhance Scheuermann, R.H., Ceusters, W. & Smith, B., 2009. Toward an Ontologi-
the new IDOPlant extension, simultaneously enriching all cal Treatment of Disease and Diagnosis. Summit on Translational Bi-
ontologies involved by filling in terms needed for logical oinformatics, 2009, p.116.
Smith, B. et al., 2005. Relations in biomedical ontologies. Genome Biology,
definitions. By expanding the core terms of the IDO to 6(5), p.R46.
plants, we can learn how plant diseases differ from, and are Smith, B. et al., 2007. The OBO Foundry: coordinated evolution of ontolo-
similar to, infectious diseases in general. gies to support biomedical data integration. Nat Biotech, 25(11),
pp.1251–1255.
Studholme, D.J., Glover, R.H. & Boonham, N., 2011. Application of High-
ACKNOWLEDGEMENTS Throughput DNA Sequencing in Phytopathology. Ann Rev of Phyto-
pathology, 49(1), pp.87–105.
RW, JE, DWS, and PJ are supported by NSF-IOS: 0822201 Swarbreck, D. et al., 2008. The Arabidopsis Information Resource (TAIR):
to the Plant Ontology Project. BS is supported by NIH:U54 gene structure and function annotation. Nuc Acids Research, 36(suppl
HG004028 (National Center for Biomedical Ontology) and 1), p.D1009 –D1014.
Topalis, P. et al., 2010. IDOMAL: an ontology for malaria. Malaria Jour-
NIH / NIAID R01 AI 77706-01 (Immune System Biological
nal, 9(1), p.230.
Networks). Thanks to Lindsay Cowell, Laurel Cooper, Rob- The UniProt Consortium, 2010. Ongoing and future developments at the
ert Hoehndorf, and three anonymous reviewers for com- Universal Protein Resource. Nuc Acids Research, 39(Database),
ments on earlier versions of this manuscript. p.D214–D219.
Youens-Clark, K. et al., 2011. Gramene database in 2010: updates and
extensions. Nuc Acids Research, 39(suppl 1), p.D1085 –D1094.
5