- Open Access
PDON: Parkinson’s disease ontology for representation and modeling of the Parkinson’s disease knowledge domain
© Younesi et al. 2015
- Received: 12 May 2015
- Accepted: 14 September 2015
- Published: 22 September 2015
Despite the unprecedented and increasing amount of data, relatively little progress has been made in molecular characterization of mechanisms underlying Parkinson’s disease. In the area of Parkinson’s research, there is a pressing need to integrate various pieces of information into a meaningful context of presumed disease mechanism(s). Disease ontologies provide a novel means for organizing, integrating, and standardizing the knowledge domains specific to disease in a compact, formalized and computer-readable form and serve as a reference for knowledge exchange or systems modeling of disease mechanism.
The Parkinson’s disease ontology was built according to the life cycle of ontology building. Structural, functional, and expert evaluation of the ontology was performed to ensure the quality and usability of the ontology. A novelty metric has been introduced to measure the gain of new knowledge using the ontology. Finally, a cause-and-effect model was built around PINK1 and two gene expression studies from the Gene Expression Omnibus database were re-annotated to demonstrate the usability of the ontology.
The Parkinson’s disease ontology with a subclass-based taxonomic hierarchy covers the broad spectrum of major biomedical concepts from molecular to clinical features of the disease, and also reflects different views on disease features held by molecular biologists, clinicians and drug developers. The current version of the ontology contains 632 concepts, which are organized under nine views. The structural evaluation showed the balanced dispersion of concept classes throughout the ontology. The functional evaluation demonstrated that the ontology-driven literature search could gain novel knowledge not present in the reference Parkinson’s knowledge map. The ontology was able to answer specific questions related to Parkinson’s when evaluated by experts. Finally, the added value of the Parkinson’s disease ontology is demonstrated by ontology-driven modeling of PINK1 and re-annotation of gene expression datasets relevant to Parkinson’s disease.
Parkinson’s disease ontology delivers the knowledge domain of Parkinson’s disease in a compact, computer-readable form, which can be further edited and enriched by the scientific community and also to be used to construct, represent and automatically extend Parkinson’s-related computable models. A practical version of the Parkinson’s disease ontology for browsing and editing can be publicly accessed at http://bioportal.bioontology.org/ontologies/PDON.
- Parkinson’s disease
- disease modeling
- knowledge engineering
Parkinson’s disease (PD), a progressive movement disorder, is the second most common neurodegenerative disease . The molecular etiology of sporadic PD has not been resolved yet and therefore PD is often called an “idiopathic” disease. In recent years, several attempts at elucidating the molecular etiology of PD have generated large omics data sets . The emerging systems view on the pathology of neurodegenerative diseases (NDDs) requires an efficient strategy to aggregate, standardize, represent, and communicate biomedical information through controlled vocabularies and ontologies . An ontology is defined as “an explicit specification of a conceptualization”, which aims to facilitate knowledge sharing .
Ontologies are the basis for automated reasoning , for large-scale annotation of entire genomes [6, 7], for data mining in microarray data , for prediction of biomolecular interactions , and for semantic and ontological search in poorly structured information sources [10, 11].
A large portfolio of widely accepted and widely used ontologies including Gene Ontology , the Sequence Ontology  and the Microarray Gene Expression Database Ontology  has evolved in the life sciences. Gene ontology (GO) is the most frequently used ontology in biomedical sciences, which provides standard functional annotations for genes and gene products. Although GO has facilitated understanding of high-throughput results by means of enrichment analysis, one of its significant limitations is that it does not capture domain-specific biological complexity . For example, GO is devoid of any disease-specific context. It can not be used for answering questions like “which disease subtypes or syndromes are over-represented in my gene or protein set?” Hence, a more useful GO ideally should contain: i. disease-specific annotations, ii. disease-specific categories, and iii. semantics that cover disease knowledge domains.
To include disease-specific biological processes, functionalities, and categories, disease-specific ontologies that cover a broad spectrum of relevant knowledge are required. Disease ontologies may reference source terminologies and vocabularies with a hierarchical concept classification such as the SNOMED CT nomenclature , the ICD ontology  and the human disease ontology . These ontologies contain human disease concepts but their high-level, broad coverage and the lack of depth in these ontologies restrict their usage for specific disease domains. Malhotra and colleagues addressed this issue in the area of NDDs by construction of Alzheimer’s disease ontology (ADO) to cover clinical, etiological, molecular and cellular mechanism aspects of AD . The aim of developing ADO was to enable semantic mining of patient records and literature for effective retrieval and extraction of accurate AD-related information, which could be used for modeling disease processes.
Similarly, there is a need for organizing the knowledge domain of PD. In response to this unmet need, we aimed at creating a disease ontology for PD (PDON) that spans from the molecular biology of the disease to clinical readouts. PDON has been created with a subclass-based hierarchy that – for the majority of concepts - uses subsumption relation (i.e. is_a). However, based on demand from biomedical experts for richer relations, partonomy relation (i.e. part_of) was also introduced, as a concession to better usability of the ontology by interdisciplinary experts. We demonstrate the ontology's usability through our use cases, keeping in mind that, re-usability by other teams is an important aspect in ontology construction and its adoption by the community. As a proof of concept, we performed evaluation of PDON performance by measuring its ability to recover pre-existing, expert-curated information from the knowledge space of PD in the literature with the aim of generating novel insights and hypotheses.
The power of the ontology can be applied to several scenarios, e.g. building recommendation systems by mapping drug failure events to mechanisms and stages of disease/stages of drug discovery, or distinguishing proven facts from hypotheses and speculations . Furthermore, in the emerging era of systems analysis of NDDs, such a knowledge-driven approach is expected to support the integration of multiscale and multilevel information across different biological scales, from molecular networks to clinical readouts.
To this end, we have proposed a model-driven approach to integrating biomedical knowledge and data into mechanistic models that represent cause-and-effect relationships among molecular entities, biological processes and their corresponding clinical outcomes. Using this strategy in the current study, we demonstrate how PDON can be utilized not only to causally link molecular etiology of PD to impaired biological processes and their corresponding disease outcome (Application scenario 1) but also to annotate experimental datasets with their corresponding knowledge description for further integration into disease models (Application scenario 2).
PDON represents a range of key concepts specific to the knowledge domain of Parkinson’s disease through different views, which have been integrated in the ontology. Views are root super-classes that organize concepts within a certain knowledge domain, as they are realized and seen by experts in reality. PDON is represented by nine views:
The view ‘Clinical aspects’ describes a broad range of motor and non-motor features displayed by PD patients. These features have been classified into three upper-level categories that capture clinical concepts related to “diagnostics”, “symptomatology” and “treatment” of PD.
The view ‘Etiology’ captures both genetic and environmental factors that are known to cause familiar PD and Parkinsonism due to toxic dopaminergic cell death, respectively. Toxic and infectious agents as well as genetic mutations are classified within this view; in addition, epidemiologically confirmed factors such as smoking and pesticides have been also included.
The view ‘Model of Parkinson’s disease’ contains various in-vivo and in-vitro disease models that are in routine use in PD research.
The view ‘Neuropathology’ was included to highlight two prototypic hypotheses of PD-related mechanisms, namely synucleopathy and the emerging tauopathy. It is expected that this view is populated further and enriched with more neuropathological concepts by the PD research community.
The view ‘Familial neurodegenerative disease’ includes those hereditary disorders that are clinically associated with PD, such as Huntington’s and Wilson’s diseases.
‘Idiopathic Parkinson’s disease’, ‘Primary parkinsonism’, ‘Secondary parkinsonism’, and ‘Parkinson-plus Syndrome’ represent four separate views as per recommendation of the clinical expert panel. These views provide a categorized overview on distinct syndromes associated with PD based on their origin of cause. For example, the primary parkinsonism class represents parkinsonian syndromes for which a definite cause has been identified (e.g. mutations in PARK genes), whereas secondary parkinsonism syndromes are induced by a hypothetical cause that is potentially identifiable. Those syndromes with unknown causative factor have been clinically assigned to the Parkinson-plus view.
Summary of the structural parameters and their corresponding values measured for PDON
No. of classes
No. of synonyms
Functional evaluation and gain of knowledge measurement:
a query with all the PDON concepts, which resulted in a list of 16333 human genes/proteins; and
PDON branch queries were formulated as ([PDON Node: “<THE BRANCH CONCEPT>”]) AND [MeSH Disease:”Parkinson Disease”] AND [Free Text:” < THE BRANCH NAME>”]
Parameter values and the final value of the knowledge gain calculated for three major branches of PDON. TM: number of relevant genes/proteins to the branch by PDON; GS: number of genes/proteins extracted from the PD map as gold standard; N: total number of genes/proteins for each branch retrieved by PDON. The queries were performed on Human Genes/Proteins and SCAIView returned lists of unique genes specific to each branch. Numbers represent counts of retrieved genes by SCAIView using PDON
Knowledge domain branches
TM ∩ GS
Gain of novel knowledge
Enriched pathways in the content of new knowledge
Etiology of PD
MAPK, Chemokine, Adipocytokine, Neurotrophin, Insulin signaling
Clinical aspects of PD
GPCR signaling, Neuroactive ligand-receptor interactions, Rhodopsin-like receptors, Peptide ligand-binding receptors, Gastrin-CREB signaling
Neuropathology of PD
Immune system, Signaling by GPCR, Endocytosis, Toll-like receptor signaling, Hemostasis
Table 2 lists parameters of the knowledge gain metric and corresponding novel pathways for selected views of the PDON. The highest percentage of new knowledge by PDON is gained in the branches representing neuropathology, etiology and clinical concepts, respectively. Functional analysis of these additionally identified genes/proteins shows that a couple of statistically significant pathways (in terms of both member proteins and p-values) could be added to the existing PD map.
Results of PDON evaluation based on expert questions. For both competency questions, PDON-driven search in SCAIView retrieved less number of abstracts than simple queries in PubMed but more relevant to the questions (i.e. less noise). This performance efficiency for the PDON-driven search has been calculated in percent as shown in the last column
Competency question number
Total no. of abstracts retrieved by PDON in SCAIView
No. of PDON-derived abstracts answering the questions
Total no. of abstracts retrieved by PubMed
No. of PubMed-derived abstracts answering the questions
PDON-driven retrieval efficiency (% PDON retrieval - % PubMed retrieval)
28.7 %-21 %: 7.7 %
83.3 %-33.3 %: 50 %
Competency question 1. Return all literature references mentioning drugs used to treat 'freezing' in PD.
Query in SCAIView: [MeSH Disease:"Parkinson Disease"] AND [Parkinson Ontology:"Freezing"] AND [Parkinson Ontology:“Gait disturbance“]) AND [ATC:"ANTI-PARKINSON DRUGS"]) AND [Free Text:“therapy“]
Query in PubMed: (("parkinson disease"[MeSH Terms] AND gait[Text Word]) AND freezing[Text Word]) AND drug[Text Word])) AND therapy[Text Word]
Competency question 2. Return literature references containing genes that provide resistance to PD in the animal model MPTP.
Query in SCAIView: [MeSH Disease:"Parkinson Disease"] AND [Full Text:"resistance"] AND [Parkinson Ontology:"MPTP model"]) AND [Mouse Genes/Proteins]
Query in PubMed: (((parkinson disease[MeSH Terms]) AND mptp[MeSH Terms]) AND resistance[Text Word]) AND animal model[MeSH Terms]
Results of PDON-driven search in response to expert competency questions. In contrast to PubMed queries, PDON-driven search in SCAIView generated a list of entities that precisely answer the competency questions
Return all literature references mentioning drugs used to treat 'freezing' in PD.
6858781, 16222436, 12217618, 15262734
12112107, 22324564, 18937611
Return literature references containing genes that provide resistance to PD in the animal model MPTP.
As the results of benchmarking indicate in Table 3 as well, PDON-driven search outperforms PubMed when it comes to specificity (i.e. less noise in terms of the number of retrieved documents) and coverage (i.e. higher number of relevant documents). For less specific queries, PDON performs at the level of PubMed baseline or better. When compared to PubMed, PDON showed higher performance (greater specificity) in retrieving relevant document, i.e. documents containing appropriate answers in the form of hypotheses related to the expert question.
Application scenario 1: linking etiology view of the PDON to cause-and-effect mechanistic models
One of the biggest scientific challenges we face in the area of NDDs such as AD or PD is the lack of a clear understanding of the alterations at the molecular level that leads to disease manifestation and the very cause of those alterations. As a consequence, the molecular etiology of neither AD nor PD can be described in a formal knowledge representation. In a recent “comment” in Nature Reviews Drug Discovery, Ismael Kola and John Bell called for a new approach towards classifying diseases . They proposed a new taxonomy of disease based on the (presumed) underlying mechanisms. The underlying fundamental notion is that any personalization of treatment requires a deep understanding of the pathophysiological mechanisms and - as a consequence – should be based on a mechanistic understanding of disease etiology.
In the work presented here, we undertake a first attempt at including etiological knowledge in the knowledge representation describing PD. Although this is not yet the “mechanism-based taxonomy” that Kola and Bell called for, it may represent a first step towards integrating mechanistic information on PD – as far as this information can be represented by means of causal and correlative relationships. A dedicated modeling syntax for the representation of causal and correlative relationships is provided by the Biological Language Expression, OpenBEL . OpenBEL is ideally suited to capture knowledge on causal and correlative relationships that may exist at any level of biomedicine. The OpenBEL language can be used for multi-scale and multi-level modeling of any chain of events that may underlie disease.
In the “etiology” section of the PDON, we have encoded those PD-specific disease processes that are comparably well understood in OpenBEL and included these representations directly as “small models” in the PDON. These models can be easily extended, automatically enriched and extracted from the PDON. To exemplify this approach, we have chosen “inherited familial case of PD”, which provides a straightforward scenario for “cause-effect” relationships.
Altogether, this model illustrates the chain of cause-and-effect events (“chain of causation”) starting from the causal effect of defined mutations on cellular processes and ending in clinical manifestations of disease outcomes.
Application scenario 2: Re-annotation of gene expression data sets with the PDON concept classes
One of the obvious benefits of disease ontologies is their ability to harmonise the annotation of data sets with well-curated terms in their controlled vocabularies. Consistently annotated gene expression data sets will facilitate the automated identification of comparable studies in systematic meta-analyses that aim to identify patterns of expression or interesting relationships between expressed genes that may come to light in the context of multiple studies. Gene expression studies can be found in the GEO (Gene Expression Omnibus) database, which is publicly available . To demonstrate the use of PDON for the re-annotation of gene expression data sets we have selected from the GEO database two examples for gene expression data sets with rather heterogeneous annotations:
GSE 16658 ; as an example of a data set with rather limited annotation
GSE 32037 ; as an example of a GEO data set with comparably rich annotation
The observed lack of detail in the description of gene expression data fields is a common phenomenon with many primary data repositories; ArrayExpress – a database for functional genomics experiments including gene expression studies - therefore is systematically re-annotating all data sets, but that is a manual process done by expert curators who usually follow annotation guidelines that are valid for a broad spectrum of indication areas and thus do not represent specific information at great detail.
More detailed information about the cohort of patients is, however, often available from the primary publication describing the data set. In our example, the respective data set has been described in a publication that is available as open access. The terms and conditions of the journal do allow automated analysis of the full text. Therefore, the complete full text publication describing GSE 16658 was subjected to automated analysis with PDON terms .
The complete full text publication describing GSE 16658 was subjected to automated analysis with terms from the PDON terminology using a UIMA-based annotation workflow. Unstructured Information Management Architecture (UIMA) is content analytics software, which is used for natural language processing in large sets of unstructured text such as biomedical publications . Detailed information on the patient cohort underlying the study is tagged and indicated by mark-ups (Additional file 1). Moreover, the ubiquitin pathway, which has been identified as a pathogenic pathway in this publication, was highlighted by PDON repeatedly. As demonstrated in this example, curation teams can readily make use of automatically pre-annotated text to enrich the annotation of GSE 16658 with the PDON terminology.
Disease-specific ontologies are emerging as powerful semantic tools, which go beyond the scope of GO and offer high semantic specificity and sensitivity by placing the disease context at the center. This trend is indicative of the fact that successful development of novel biomarkers and drugs requires an integrated systems view of the pathophysiological mechanisms that underlie disease, which is highly geared towards understanding connections across multiple biological scales, from molecular interactions to disease phenotypes. Among chronic and complex diseases, NDDs are the most challenging diseases to deal with and present a high level of heterogeneity ranging from diverse molecular mechanisms, various cell types and anatomical regions to different levels of clinical manifestation. Thus, harmonization of such high diversity at the data level and enormous heterogeneity at the clinical level needs high-resolution representation of the knowledge domains specific to NDDs.
In the field of NDDs disease-specific ontologies such as epilepsy ontology  and Alzheimer’s disease ontology  have been recently introduced. However, most potential users of such ontologies including molecular biologists and clinicians find the notations of such formal representations counter-intuitive and non-reusable. PDON not only provides a standard terminology and platform for interoperability of the PD knowledge domain among research groups, but also supports users via literature search through its lexicalized form. The broad coverage of PD-specific concepts at various granularity levels ensures that PDON enables users to capture and integrate PD-related information for systems analysis purposes. In terms of specificity of the PDON search, we found that the maximum overlap of the PDON-derived gene list with the PD map-derived genes is reached in each query within the top 50 genes returned by SCAIView and the most interesting genes missing in the PD model were then identified after this rank position. The PDON search covered more than 90 % of the gold standard entity list from the PD map reference when the entire knowledge domain of PD was queried, which shows the high sensitivity of the PDON. However, the low gene overlap between the gold standard (the reference PD map) and the PDON-derived list of genes (Table 2) is indicative of the large amount of peripheral information that has been used in the construction of the PD disease map complementary to the core pathology. For instance, among the novel knowledge gained by PDON is ATP5I protein, which is a part of the ATP synthase complex in the mitochondrial ROS cascade and complements the core cascades involved in mitochondrial dysfunction under PD condition. In the reference PD map, however, this protein has not been shown and only complex V has been presented. On the other hand, this comparison suggests that PDON-derived results contain rich information content relevant to the core pathophysiology of PD, which is still absent in the PD disease map and can be potentially included in the next version. Introduction of the gain of knowledge measurement metric made it possible to quantitatively measure the gain of novel information content captured by PDON. Moreover, representation of the gained knowledge by enriched pathways confirms the novelty of the gained knowledge in comparison to the reference PD map, as these pathways are not present in the current reference PD map.
An important feature of the PDON is its capacity to project multiple perspectives of the PD knowledge domain among stakeholders in both academia and industry from molecular biology to epidemiological and clinical studies. For instance, during the construction and separate rounds of curation with both clinicians and molecular biologists, special attention was paid to bridging the gap between two higher-level perspectives, namely the perspective of clinicians and the perspective of drug developers. Due to the representation of different perspectives in the domain of PD research, we do hope that the PDON will be broadly adopted and used for the exchange of data, annotation of existing data sets and the communication of knowledge.
The unique value that application of PDON can provide to the PD research community was demonstrated by using PDON for the rapid construction of mechanistic BEL models. This ability can be used for representation as well as visualization of the PD knowledge subdomains in the form of cause-and-effect systems models. As demonstrated in the application scenario, mutations in the PINK1 gene is well known for their causal association to familial early onset PD but their mechanistic model reveals their potential involvement in the etiology of sporadic late onset PD as well. This feature of the PDON, therefore, can be used for investigating associations between familial and sporadic PD in general, and associations between parkinsonism syndromes in particular at the mechanism level. To this end, the authors foresee that one potential application of PDON could be revision of the current classification of PD syndromes based on etiological mechanism underlying each syndrome.
In the times of re-orientating PD research towards systems modeling and analysis, the Parkinson’s disease ontology serves as a semantic framework for standardization and harmonization of a large amount of heterogeneous data and knowledge in the field of PD. Indeed, PDON delivers the knowledge domain of PD in a compact, computer-readable form, which can be further used to construct, represent and automatically extend PD-related computable models. Beside advantages, PDON has its own limitations such as missing concepts, lack of standard definitions, or incompleteness of synonym list. Addressing these shortcomings requires constant contribution of the PD research community to the betterment of the current version of PDON.
The PDON has been constructed in accordance to the ontology-building life cycle . Such a methodology – compared to other methods - offers a set of defined stages when building ontologies to assist identifying construction principles for each stage as well as relationships among stages.
Knowledge acquisition and conceptualization
A first collection of terms and concepts related to PD was generated by scanning various knowledge sources. Initially we used a list of sources recommended by Parkinson’s experts (i.e. neuroscientists and clinicians) including medical text books such as Parkinson’s Disease and Movement Disorders  and encyclopedias like Encyclopedia of Parkinson’s Disease . After extracting the key concepts manually, we then used the search functionality of Google (such as Books and Scholar) to find online resources that contain additional concepts describing the knowledge domain of PD (e.g. www.parkinsons.org). Other resources including review articles and content of online books were used in this way and any available hierarchical organization (structure) of the concepts was extracted along with the concepts themselves. Corresponding definitions and references were also included. Concept enrichment was assisted by n-gram analysis so that publication abstracts were scanned for 2-grams to 5-grams, describing meaningful terms with 2 to 5 words, by using a java program written for this purpose. These n-grams underwent manual inspection and relevant terms were added to the ontology.
The Protégé OWL editor was used as a tool for building the PDON in Web Ontology Language (OWL) format . Concept classes were further annotated with synonyms.
Structural features of the ontology were computed using an existing java script . These features include topological and logic properties such as: depth and breadth (related to the cardinality of paths in a graph), tangledness (related to multi-hierarchical nodes), and fan-outness (related to the dispersion of nodes).
Functional performance of the ontology was measured through a novel model-based evaluation approach. In this approach, the assumption is that the expert-curated molecular map of PD, which has been recently published by Fujita et al. (2013), represents a good part of the current knowledge about molecular processes related to PD and is considered as the gold standard . A list of genes and proteins in this map was used for benchmarking the list of genes and proteins that had been retrieved from the literature through PDON-supported literature mining.
The expert panel’s revision of the ontological structure is considered as a genuine evaluation for disease ontologies . Revisions of the PDON drafts were initially performed in the presence of an expert panel of 6 experts in PD from UCB Pharma, composed of molecular biologists, clinicians and physicians, who generated the “pharma view” of the ontology. In order to generate the “clinical view” of the ontology, the PDON underwent a manual curation by clinician experts in the field of PD (UW and DS).
Ontology-driven information retrieval and extraction
Transformation of the ontology OWL format into a dictionary file was achieved using a Java program that extracts the concept names and the corresponding synonyms from the ontology OWL structure and assigns unique identifiers to each concept which can be stored in form of a dictionary. This dictionary was incorporated into the text-mining tool ProMiner  and results were deployed in the semantic search engine SCAIView for context-sensitive visualization of query results. SCAIView is a semantic search engine that provides a text mining-based environment for information retrieval and extraction from PubMed publications using various terminologies and ontologies . SCAIView environment can be freely accessed under http://www.scaiview.com/scaiview-academia.html; PubMed portal can be accessed through http://www.ncbi.nlm.nih.gov/pubmed.
LTM = the “curated” result list from text-mining (TM) algorithm
LGS = the result list from the reference gold standard
NTM = the total number of entities retrieved from TM
TM: Relevant (true positive) entities (i.e. genes/proteins)
GS: Gold standard list of entities
TM ∩ GS: Overlap between entities retrieved by TM and entities extracted from GS
To show the biological content of the novel knowledge gain by this approach, the list of novel genes/proteins retrieved by PDON (i.e. |TM| - |TM ∩ GS|) was subjected to the pathway enrichment analysis using MsigDB .
Mechanistic model building
(([SNPs]) AND [Human Genes/Proteins:”PINK1”]) AND [PDON]
The authors wish to thank Jonathan Van Eyll, Javier garcia-Ladona, Anne Michel, Patrik Foerch, Michel Gillard and Frank Vanderdonck from UCB Pharma for their contribution to building the ontology. Theo Mevissen, Jürgen Klein and Christian Ebeling from Fraunhofer Institute SCAI are acknowledged for their support in automatic annotation of GEO entries and full-text publications. This work was conducted using the Protégé resource, which is supported by grant GM10331601 from the National Institute of General Medical Sciences of the United States National Institutes of Health. This study was supported by the German Federal Ministry of Education and Research (BMBF) within the Biopharma Neuroallianz consortium, and by UCB Pharma GmbH (Monhein, Germany).
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- de Lau LM, Breteler M. Epidemiology of Parkinson's disease. Lancet Neurol. 2006;5:525–35.View ArticlePubMedGoogle Scholar
- Scholz SW, Mhyre T, Ressom H, Shah S, Federoff HJ. Genomics and Bioinformatics of Parkinson's Disease. Cold Spring Harbor Perspect Med. 2012;2:a009449.View ArticleGoogle Scholar
- Rubin DL, Shah NH, Noy NF. Biomedical ontologies: a functional perspective. Briefings Bioinf. 2008;9:75–90.View ArticleGoogle Scholar
- Gruber TR. Toward principles for the design of ontologies used for knowledge sharing? Int J Hum Compu Stud. 1995;43:907–28.View ArticleGoogle Scholar
- Héja G, Surján G, Varga P. Ontological analysis of SNOMED CT. BMC Med Inf Decis Making. 2008;8:S8.View ArticleGoogle Scholar
- Thomas PD, Kejariwal A, Campbell MJ, Mi H, Diemer K, Guo N, et al. PANTHER: a browsable database of gene products organized by biological function, using curated protein family and subfamily classification. Nucleic Acids Res. 2003;31:334–41.PubMed CentralView ArticlePubMedGoogle Scholar
- Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry M, et al. Gene ontology: tool for the unification of biology, The Gene Ontology Consortium. Nat Genet. 2000;25:25–9.PubMed CentralView ArticlePubMedGoogle Scholar
- Whetzel PL, Parkinson H, Causton HC, Fan L, Foster J, Fragoso G, et al. The MGED ontology: a resource for semantics-based description of microarray experiments. Bioinformatics. 2006;22:866–73.View ArticlePubMedGoogle Scholar
- Yoshikawa S, Satou K, Konagaya A. Drug interaction ontology (DIO) for inferences of possible drug–drug interactions. Stud Health Technol Inf. 2004;107:454–8.Google Scholar
- Spasic I, Ananiadou S, McNaught J, Kumar A. Text mining and ontologies in biomedicine: making sense of raw text. Briefings Bioinf. 2005;6:239–51.View ArticleGoogle Scholar
- Doms A, Schroeder M. GoPubMed: exploring PubMed with the gene ontology. Nucleic Acids Res. 2005;33:W783–6.PubMed CentralView ArticlePubMedGoogle Scholar
- Eilbeck K, Lewis SE, Mungall CJ, Yandell M, Stein L, Durbin R, et al. The sequence ontology: a tool for unification of genome annotations. Genome Biol. 2005;6:R44.PubMed CentralView ArticlePubMedGoogle Scholar
- Pal D. On gene ontology and function annotation. Bioinformation. 2006;1:97.PubMed CentralView ArticlePubMedGoogle Scholar
- SNOMED CT. The International Health Terminology Standards Development Organisation (IHTSDO), Copenhagen. 2015. http://www.ihtsdo.org/snomed-ct. Accessed 03.08.2015.
- International Classification of Diseases (ICD). The World Health Organization, Geneva. 2015. http://www.who.int/classifications/icd/en/. Accessed 03.08.2015.
- Shah NH, Cole T, Musen MA. Analyses Using Disease Ontologies. PLoS Comp Biol. 2012;8:e1002827.View ArticleGoogle Scholar
- Malhotra A, Younesi E, Gündel M, Müller B, Heneka MT, Hofmann-Apitius M. ADO: A disease ontology representing the domain knowledge specific to Alzheimer's disease. Alzheimers Dement. 2014;10:238–46.View ArticlePubMedGoogle Scholar
- Malhotra A, Younesi E, Gurulingappa H, Hofmann-Apitius M. ‘HypothesisFinder:’A Strategy for the Detection of Speculative Statements in Scientific Text. PLoS Comp Biol. 2013;9:e1003117.View ArticleGoogle Scholar
- Kola I, Bell J. A call to reform the taxonomy of human disease. Nat Rev Drug Discov. 2011;10:641–2.View ArticlePubMedGoogle Scholar
- OpebBEL. www.openbel.org. Accessed 30 June 2014.
- Ma Y, Mehta SL, Lu B, Li PA. Deficiency in the inner mitochondrial membrane peptidase 2-like (Immp21) gene increases ischemic brain damage and impairs mitochondrial function. Neurobiol Dis. 2011;44:270–6.PubMed CentralView ArticlePubMedGoogle Scholar
- Gene Expression Omnibus database. http://www.ncbi.nlm.nih.gov/geo/. Accessed 10.12.2013.
- ArrayExpress database. https://www.ebi.ac.uk/arrayexpress/. Accessed 10.01.2014.
- Martins M, Rosa A, Guedes LC, Fonseca BV, Gotovac K, Violante S, et al. Convergence of miRNA expression profiling, alpha-synuclein interacton and GWAS in Parkinson’s disease. PLoS One. 2011;6:e25443.PubMed CentralView ArticlePubMedGoogle Scholar
- Ferrucci D, Lally A. Building an example application with the unstructured information management architecture. IBM Syst J. 2004;43:455–75.View ArticleGoogle Scholar
- Almeida P, Gomes P, Sales F, Nogueira A, Dourado A. Ontology and knowledge management system on epilepsy and epileptic seizures. In: A. Burger et al., editors. Proceedings of the 3rd International Workshop on Semantic Web Applications and Tools for the Life Sciences. Berlin; 2010.Google Scholar
- Stevens R, Goble CA, Bechhofer S. Ontology-based knowledge representation for bioinformatics. Briefings Bioinf. 2000;1:398–414.View ArticleGoogle Scholar
- Jankovic J, Tolosa E. Parkinson’s disease and movement disorders. 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2007.Google Scholar
- Mosley AD, Romaine DS, Samii A. The encyclopedia of Parkinson’s disease. 2nd ed. NewYork: Infobase Publishing Inc; 2010.Google Scholar
- A free, open-source ontology editor and framework for building intelligent systems. http://protege.stanford.edu/. Accessed 8 March 2013.
- Gündel M, Younesi E, Malhotra A, Wang J, Li H, Zhang B, et al. HuPSON: the human physiology simulation ontology. Biomed Semantics. 2013;4:35.View ArticleGoogle Scholar
- Fujita KA, Ostaszewski M, Matsuoka Y, Ghosh S, Glaab E, Trefois C, et al. Integrating pathways of Parkinson's disease in a molecular interaction map. Mol Neurobiol. 2013;49:88–102.PubMed CentralView ArticlePubMedGoogle Scholar
- Obrst L, Ceusters W, Mani I, Ray S, Smith B. The evaluation of ontologies. In: Baker CJO, Cheung KH, editors. Semantic Web: revolutionizing knowledge discovery in the life sciences. NewYork: Springer; 2007. p.139-158.Google Scholar
- Hanisch D, Fundel K, Mevissen HT, Zimmer R, Fluck J. ProMiner: Rule based protein and gene entity recognition. BMC Bioinf. 2005;6:S14.View ArticleGoogle Scholar
- Friedrich CM, Dach H, Gattermayer T, Engelbrecht G, Benkner S, Hofmann-Apitius M. @neuLink: a service-oriented application for biomedical knowledge discovery. In: Tony S, Silverstein JC, Joel S, Yannick L, Mary K, Ian F, Vincent B, Robert Beck J, editors. Proceedings of the HealthGrid: 2–4 June 2008. Chicago: IOS Press Amsterdam; 2008. p. 165–72.Google Scholar
- Molecular Signatures Database. http://www.broadinstitute.org/gsea/msigdb/index.jsp. Accessed 7 February 2014.
- OpenBEL bel-editor. https://github.com/OpenBEL/bel-editor. Accessed 30 November 2014.
- Killcoyne S, Carter GW, Smith J, Boyle J. Cytoscape: a community-based framework for network modeling. Methods Mol Biol. 2009;563:219–39.View ArticlePubMedGoogle Scholar