Ontology modeling of genetic susceptibility to adverse events following vaccination Ontology modeling of genetic susceptibility to adverse events following vaccination Yu Lin, Yongqun He Unit for Laboratory Animal Medicine, Department of Microbiology and Immunology, Center for Computational Medicine and Bioinformatics, and Comprehensive Cancer Center, University of Michigan Medical School, Ann Arbor, MI 48109, USA ABSTRACT may be quite common in the healthy population. Moreover, Administration of different vaccines triggers a variety of adverse events the main determinants of susceptibility may be different in in some groups of people but not in others. This phenomenon may be due to the variation of genetic factors that affects the susceptibility to vaccine different populations [6]. With current technological adverse events. In this study, we introduce the development of an Ontology advances and new biostatistics approaches to understanding of Genetic Susceptibility Factor (OGSF) that is aligned with the Basic a large number of databases of information, we can now Formal Ontology (BFO). OGSF represents the genetic susceptibility, better understand how genetic variations become critical to genetic susceptibility factors and vaccine adverse events using formal ontologies. Two case studies were used to test and validate the model. One vaccine-induced positive host responses and adverse case study represents a human gene allele DBR1*15:01 as a genetic reactions. susceptibility factor to vaccine Pandemrix related multiple sclerosis. An Ontology of Genetic Susceptibility Factor (OGSF) Genetic polymorphisms associated with smallpox vaccine adverse event was previously developed for our formalization of the was analysed as the second use case. A SPARQL query, visualization of extracted data as a network and the social network analysis of the network, definitions of ‘genetic susceptibility’ and ‘genetic further provide insights on the evaluation and application of the ontology. susceptibility factor’ using the TCF7L2 gene and its susceptibility to Type 2 Diabetes as an example [7]. The 1 INTRODUCTION entities important for the representation of genetic susceptibility to diseases include: genetic polymorphism, Vaccines have enabled tremendous decreases in infectious the population and geographical location, the disease diseases and remain among the most effective of our public entities, and related statistical entities (e.g., odds ratio and p- health initiatives. At the same time, as an ever increasing value). Here we consider that a vaccine adverse event is a number of vaccines is administered globally, many vaccine- pathological bodily process, and we extend the former work associated adverse events and reactions have been identified to model the genetic susceptibility to adverse event. and threaten the public health successes attributable to Based on previous studies, we have now developed a new vaccines [1]. As defined in the Vaccine Adverse Event version of genetic susceptibility-focused ontology, the Reporting System (VAERS) and Ontology for Adverse Ontology of Genetic Susceptibility Factor (OGSF) by using Event (OAE), a vaccine adverse event is an adverse event Basic Formal Ontology (BFO) 2.0 as its upper ontology. following vaccination and does not assume a causal OGSF is used to study the susceptibility factors associated association [2]. Vaccine-related adverse events often occur with vaccine adverse events. in some populations but not in others, which has led to the hypothesis of genetic susceptibility to vaccine adverse events [3, 4]. 2 METHODS Genetic susceptibility, also called genetic predisposition, 2.1 Ontology editing is an increased likelihood or chance of developing a The format of OGSF ontology is W3C standard Web particular disease due to the presence of one or more gene Ontology Language (OWL2) (http://www.w3.org/TR/owl- mutations and/or a family history that indicates an increased guide/). For this study, many new terms and logical risk of the disease. The allele that confers the increased definition were added into original OGSF [7] using the risk/susceptibility may be inherited but the disease itself will not. The single locus genotype is usually insufficient to Protégé 4.3.0 build 304 OWL ontology editor cause a disease. For the disease to appear, impaired (http://protege.stanford.edu/). expressions of alleles at other gene loci and/or 2.2 Ontology term reuse and new term generation environmental factors are often needed [5]. OGSF imports the whole set of the Basic Formal Genetic susceptibility factors are the genetic entities, most Ontology (BFO) [8]. To support ontology interoperability, likely genetic variations, which influence the susceptibility. many terms from reliable ontologies are reused. For this The genetic susceptibility factors contributing susceptibility purpose, OntoFox [9] was applied for extracting individual to a disease may not be obvious mutations. It is more likely a combination of subtle changes on several genes, which terms from external ontologies. For those genetic susceptibility-specific terms, we generated new OGSF IDs with the prefix of “OGSF_” followed by seven-digit auto- * To whom correspondence should be addressed: incremental digital numbers. yuln@med.umich.edu 1 Ontology modeling of genetic susceptibility to adverse events following vaccination 2.3 Evaluation of OGSF studies, and population-based studies are conducted in order Use case studies were designed based on literature survey. to determine whether or not a genetic variation mediates the SPARQL was performed using the SPARQLquery plug-in diseases outcome such as a vaccine adverse event. embedded with Protégé4.3.0 build 304. Graphed data was Fig. 1 shows how we use OGSF terms and BFO relations extracted using the OntoGraf plug-in [10] Gephi 0.8.2 beta to represent genetic susceptibility to vaccine adverse event. (http://gephi.org)[11] was used to conduct social network data analysis and visualization based on the extracted graph textual conclusion of genetic susceptibility has_specified_output genetic association investigation data from instances of OGSF. is_about has participant at some time 2.4 Availability and access has genetic genetic The website for OGSF project is available at susceptibility material basis susceptibility factor case group http://code.google.com/p/ogsf/. The source of the ontology is_about is_a genetic susceptibility to is also available in the NCBO Bioportal: vaccine adverse event part of continuant has member part at some time at some time http://bioportal.bioontology.org/ontologies/3214. realized in human vaccinee carrying vaccine 3 RESULTS adverse event actively participates in susceptibility allele for adverse event 3.1 OGSF is aligned with BFO is_preceded_by inverse ('vaccine immunization for host') The development of OGSF follows the OBO Foundry vaccination is specified input of vaccine principles, including openness, collaboration, and use of a Fig. 1. Design pattern for representing genetic susceptibility to a common shared syntax [12]. The early version of OGSF vaccine adverse event (VAE). was not well aligned with BFO. To align OGSF with BFO The set of core terms representing the whole topic are 2.0 Graz version, we started with key terms and render them ‘genetic susceptibility factor’, ‘genetic susceptibility’, using BFO's terms as parent terms. ‘adverse event’ and ‘textual conclusion of genetic There are two core terms in OGSF: 'genetic susceptibility’. In Fig.1, the 'genetic susceptibility factor' is susceptibility' and 'genetic susceptibility factor'. The OGSF the material basis of 'genetic susceptibility', which has a term 'genetic susceptibility' (OGSF_0000000) is a subclass subclass ‘genetic susceptibility to vaccine adverse event’. of 'disposition' (BFO_0000016). The alternative term for The genetic susceptibility is realized in the process of 'genetic susceptibility' is 'genetic predisposition'. Note that ‘vaccine adverse event’. The ‘genetic susceptibility factor’ in BFO 2.0 the term 'predisposition' is not included, so we is the part of a 'human vaccinee carrying susceptibility allele put genetic susceptibility directly as the child term of for adverse event’, which ‘actively participates in’ the 'disposition'. The first level child terms of 'genetic ‘vaccine adverse event’. On the other hand, a ‘genetic susceptibility' include: 'genetic predisposition to disease of association investigation’ has participant ‘case group’ with type X' (OGMS_0000033), 'genetic susceptibility to the ‘human vaccine carrying susceptibility allele for adverse pathological bodily process' (OGSF_0000001), and 'genetic event’ as its member. The ‘genetic association investigation’ susceptibility to biological process' (OGSF_0000002). The has ‘textual conclusion of genetic susceptibility’ as its term that reveals our use case is 'genetic susceptibility to specified output, and the conclusion ‘is about’ both ‘genetic adverse event following vaccination' (OGSF_0000010) and susceptibility factor’ and ‘vaccine adverse event’. An it is the third level child term of 'genetic susceptibility'. inverse of VO relation: ‘vaccine immunization for host’ Another core OGSF term 'genetic susceptibility factor' interlinks the human vaccinee and ‘vaccine’. ‘Vaccine’ is a (OGSF_0000004) is a subclass of 'material entity' specified input of the process of ‘vaccination’. Relation ‘is (BFO_0000040). An allele, gene, genotype, and haplotype preceded by’ linking ‘vaccination’ and vaccine adverse can be genetic susceptibility factors. The relation: 'material event’ indicates that ‘vaccination’ happens before the basis of at some time' (BFO_0000127), is used to link ‘vaccine adverse event’. genetic susceptibility factor and genetic susceptibility. 3.3 Modeling genetic association study 3.2 Modeling genetic susceptibility to adverse event Studies have provided many supporting evidences for following vaccination asserting susceptibility factors to adverse event outcomes. The genetic susceptibility to vaccine adverse events is used Based on the OBI framework, we specially modeled the as a use case for OGSF redesign. genetic association study designs according to our use case. Genetic susceptibility reflects the relation between a The textual definition of OGSF term 'genetic association genetic factor (e.g. allele) and risk of condition, disease or investigation' was given as: 'an investigation that aims to responses to vaccines or drugs. Different levels of genetic test whether single-locus alleles or genotype frequencies (or association studies, such as family studies, genetic linkage more generally, multilocus haplotype frequencies) differ 2 Ontology modeling of genetic susceptibility to adverse events following vaccination between two groups of individuals (usually diseased Vrethem et al. reported the occurrence of severe subjects and healthy controls)'. Different types of those narcolepsy with cataplexy and multiple sclerosis (MS) in a studies exist, such as 'case-control study', 'GWAS study' previously healthy young male in association with (Genome-Wide Association Study) and 'case report'. Pandemrix vaccination [14]. The investigators found that 'GWAS study' is a type of 'case-control study' and has two that those patients carrying HLA allele DBR1*15:01 were subclasses 'initial GWAS study' and 'replicate GWAS associated with MS and those having HLA allele study'. The statistical method conducted in a study is DQB1*06:02 were associated with narcolepsy. It was also modeled as 'data analysis' that is a part of an investigation as concluded that the genetic susceptibility in this patient is a asserted in OBI. 'Case group' and 'control group' are clue that an immune-mediated mechanism and a common subclasses of 'human study subject group'. The 'human study etiology for both diseases in this patient. subject group' is the participant of the 'genetic association The DBR1*15:01 as a genetic susceptibility factor investigation'. responsible for Pandemrix-induced MS was modeled in the A statistical analysis of the genetic susceptibility is based class level using OGSF, and the particular study was on the choice of a statistical study design, which depends on modeled in instance level using OGSF (Fig 2). several factors related to the phenotype: the population, the At the class level, 'DBR1*15:01' is an 'allele of HLA accurate measurement of environmental factors, and known gene', which is also the material basis of (BFO 2: 'material genetic background among other factors. Due to the basis of at some time') 'genetic susceptibility to vaccine presence of many different cofounders, it is often difficult to adverse event'. The instance of ‘DBR1*15:01’ is a part of detect and verify genetic susceptibility factors associated the MS patient instance. In class level, 'multiple sclerosis with specific adverse event outcomes. Observed statistically AE patient' 'actively participates in' the 'multiple sclerosis significant genetic susceptibilities may be contradictory AE' process. Multiple Sclerosis adverse event is preceded among different studies [13]. More and consistent by the 'Pandemrix vaccination'. 'Pandemrix' is a participant observations in different populations may give stronger of 'Pandemrix vaccination' and it is related to the MS patient evidence to support the true causal relation between a using a short relation from Vaccine Ontology (VO): 'vaccine 'genetic susceptibility factor' and an observed outcome. immunization for host', which relates a vaccine with a Well-designed experiments may be applied to verify the vaccinee. association. In order to store the result from genetic association studies, we use 'textual conclusion of genetic publication_PMID case only study _22841884 susceptibility' to be asserted as 'specified output of' a is_about instance of 'genetic association investigation'. The 'textual conclusion positive conclusion of genetic association of genetic susceptibility' is a 'textual entity'. The 'is about' genetic susceptibility_1 has specified output study_1 relation was used to link the conclusion with 1) 'genetic genetic susceptibility to susceptibility factor' and 2) ‘vaccine adverse event’ process. is_about vaccine adverse event material basis of at some time Three terms: 'positive conclusion of genetic is_about is allele HLA DRB1 has participant susceptibility', 'negative conclusion of genetic susceptibility' is_about DRB1*15:01 of gene gene at all times instance of and 'neutral conclusion of genetic susceptibility' are asserted DRB1*15:01 (instance) 'part of continuant as subclasses of 'textual conclusion of genetic susceptibility'. at all times that whole exists' A 'positive conclusion of genetic susceptibility' means that a multiple actively multiple sclerosis instance multiple sclerosis AE positive conclusion is drawn based on a significant sclerosis AE participates in AE patient of patient_1 statistical association of a genetic factor and a vaccine is preceded by inverse ('vaccine immunization for host') adverse event as studied in this paper. A 'negative Pandemrix participates in Pandemrix vaccination at some time conclusion of genetic susceptibility' a denied association between a genetic factor and an adverse event. Sometimes, Fig. 2. OGSF modeling of vaccine-associated multiple sclerosis depending on the data, an investigator may draw a conclusion of a non-significant association but without a Since it is a case report, this study gives one specific clear deny of a possible association. This situation is positive supporting evidence to the genetic susceptibility of captured using ‘neutral conclusion of genetic susceptibility’. DBR1*15:01, which is asserted at the instance level. We use 'genetic association study_1' to represent the study, which 3.4 Case study gives a specific output 'positive conclusion of genetic Case studies are used for two purposes: 1) to validate the susceptibility_1'. This specific conclusion is about the entity modeling, 2) to test possible applications of the ontology. 'DBR1*15:01' and the 'multiple sclerosis AE'. 3.4.1 Case study 1: HLA allele DBR1*15:01 is genetic 3.4.2 Case study 2: genetic polymorphisms associated with susceptibility to Pandemrix related multiple sclerosis adverse events after smallpox vaccination 3 Ontology modeling of genetic susceptibility to adverse events following vaccination Reif et al. reported that genetic polymorphisms in an The instance level representation representing two enzyme methylenetetrahydrofolate reductase (MTHFR) and independent studies provide the statistical supporting an immunological transcription factor (IRF1) were evidence to the genetic susceptibility (Fig. 3). associated with AEs after smallpox vaccination [15]. In this Fig 3 illustrated that two ‘positive conclusions of genetic study, two independent clinical trials were conducted as susceptibility’ from clinical trail 1 and trail 2 support the ‘T initial and replicating genetic association studies separately. allele of rs1801133 SNP’ as the ‘material basis of at some The Odds Ratio was used to measure the association time’ the ‘genetic susceptibility of adverse event following between genotypes and systematic adverse event. Only vaccination’. The datatype properties ‘hasOddsRatio’ and strong association supported by a statistically significant ‘hasPvalue’ are properties of the ‘positive conclusion of the Odds Ratio in both studies was considered and asserted as a genetic susceptibility’. Using these datatype properties, the true positive genetic association. real data denotes the statistical power was represented in the In this case, the important information to be stored is the ontology. susceptibility allele of the SNPs and the statistical power in positive conclusion of genetic adverse event hasOddsRatio genetic two studies. Those information was curated and summarized hasPvalue susceptibility_trail2 has specified output association clinical trail_2 in Table 1. hasSize participates Table 1. Statistical summary of genetic susceptibility factors with is about is about case group at all times is about systematic adverse event following smallpox vaccination in trail2 genetic material systematic adverse T allele of GSF& Allele Gene Odds Ratio P-value Study susceptibility basis of rs1801133 event after smallpox publication_PMID _18454680 of vaccine at some vaccination SNP 1 or 2 adverse event time case group rs1801133 SNP T MTHFR 2.3 (1.1–5.2) 0.04 1 is about in trail1 participates is about is about at all times rs1801133 SNP T MTHFR 4.1 (1.4–11.4) 0.01 2 hasSize positive conclusion of genetic adverse event rs9282763 SNP G IRF1 3.2 (1.1–9.8) 0.03 1 hasOddsRatio genetic has specified output association clinical hasPvalue susceptibility_trail1 trail_1 rs9282763 SNP G IRF1 3.0 (1.1–8.3) 0.03 2 rs839 SNP A IRF1 3.2 (1.1–9.8) 0.03 1 Fig. 3. Modeling Case Study 2 using OGSF rs839 SNP A IRF1 3.0 (1.1–8.3) 0.03 2 Haplotype 1* G,A IRF1 3.2 (1.0–10.2) 0.03 1 3.4.3 SPARQL query Haplotype 1* G,A IRF1 3.0 (1.0–9.0) 0.03 2 A SPARQL script was developed to query against inferred Haplotype 2# T,C,A IL4 2.4 (1.0–5.7) 0.05 1 OGSF ontology. The query led to the retrieval of the genetic Haplotype 2# T,C,A IL4 3.8 (1.0–14.4) 0.06 2 susceptibility factors, as shown in Table 1. (Sparql query Notes: script shown in Supplemental material if allowed). & GSF stands for Genetic Susceptibility Factor * Haplotype 1 contains G allele of rs9282763, A allele of rs839 in IRF1 gene. # Haplotype 2 contains T allele of rs2070874,C allele of rs2243268, A allele of 3.4.4 Visualization and social network analysis rs2243290 in IL4 gene. In order to give a better view of the terms and links between The class level assertion is similar to case study 1. For terms, data from case study 2 was extracted using OntoGraf example, the constrains representing one of the genetic and visualized using Gephi as following (Fig. 4 and 5). susceptibility factors, A allele of rs839, are as follows: 1. 'material basis of at some time' some 'genetic susceptibility to adverse event following vaccination' This axiom denotes that the A allele of rs839 is the material basis of the genetic susceptibility to AE induced by vaccination 2. 'part of continuant at all times that whole exists' some ('human vaccinee experiencing systemic adverse event' and inverse('vaccine immunization for host') some 'Smallpox virus vaccine') This axiom denotes that the ‘A allele of rs839’ is part of some human who is experiencing systemic adverse event and had vaccinated by Smallpox vaccine 3. isContainedIn some 'IRF1 gene' This axiom denotes that the ‘A allele of rs839’ is contained in IRF1 gene 4. 'alternative allele of SNP' This axiom denotes that the ‘A allele of rs839’ is an alternative allele 5. 'susceptibility allele' (inferred) Fig. 4. All related nodes within case study 2. This axiom denotes that the ‘A allele of rs839’ is a susceptibility allele, so it is a genetic susceptibility factor. Fig. 4. shows how the data and terms interlinked with each other in the network of case study2. The most 4 Ontology modeling of genetic susceptibility to adverse events following vaccination connected node is ‘systematic adverse event after smallpox with the same adverse event C. To add this fact into OGSF vaccine’, since there are 10 conclusions related to it as would strengthen the belief that gene G is related to the shown in table1. All the genes, relevant SNP alleles and genetic causal association. haplotypes are interlinked with each other, and can be The notion of genetic susceptibility can be expressed captured as a community within the network, which using OWL classes, whereas each study is modeled in indicated by colors of the node. instance level as data item. To simplify the connections, the Running Gephi’s ‘filter’ function, two different views of relation 'is_about' was used to bridge the individual level the network of case study 2 were yield as shown in Fig 5. 'textual' conclusions from an individual study to a 'genetic susceptibility factor' (class level) and specific vaccine adverse event (class level). The efficiency and applicable Fig. 5. Two views of the aspects of these relations need to be tested using more genetic susceptibility complicated datasets and SPARQL query. network in case study2. 4.2 The granularity of genetic susceptibility factor is (A). EgoNetwork filter at allele level view of the network, which shows entities that Nowadays, thousands of Single Nucleotide polymorphisms are directly linked to (SNPs) can be tested efficiently in large population-based 'genetic susceptibility to studies. Researchers are using various entities to describe vaccine adverse event'. genetic susceptibility bearers, such as genotype, SNP, LD (B) Closeness centrality block, haplotype and so on. Except for LD block, other filtered view of the genetic susceptibility factors can be represented by notion of network. All the dots in allele. As defined in our previously developed Ontology for the figure have closeness Genetic Interval (OGI) [16], ‘allele’ is ‘an alternative form centrality value equal to of a genetic interval that is located at a specific position on a 0. specific chromosome’. In OGI, term 'allele' has following subclasses: 'allele of gene', 'allele of polymorphism', 'allele of SNP', 'allele of phenotype', 'allele shared by sibs'. OGSF fully imports OGI, thus inherited the OGI's allele classes Combining Fig. 5A and 5B, it indicates that: 1) in OGSF, and definitions. OGI gives formalized topological relations the genetic susceptibility is directly related with variants, between alleles and genes, so that the relations between such as SNPs and haplotypes. 2) Gene is indirectly linked to alleles and genes can be logically calculated [14]. Adopting genetic susceptibility via variants. The in-directed those relations ensure the example discussed in the section connection can be captured by centrality network analysis in 4.1 can be reasoned in OGSF. the given data set. In our specific case study 2, the closeness 4.3 Visualization of sub network of OGSF data centrality calculations of genetic susceptibility, adverse The ontology's instance level data can be visualized as event and genes are the lowest. directed graph. The visualization and network analysis results provide deep insights in terms of ontology designing. 4 DISCUSSION Representing the genetic susceptibility can be addressed 4.1 Representing genetic susceptibility requires the using three layers of information depending on researchers' notion of instance level evidence interest. The first layer is the direct link of types of genetic The purpose of representing the knowledge of genetic factors and investigated adverse event. In our representation, susceptibility here is to extend existing beliefs by adding it is grounded to allelic variant level. The second layer is new facts. For example, if in one study A1, the genetic the supporting conclusion that provides positive evidence to factor SNP B is statistically significant related to an adverse the direct link. The third layer is the linking between a gene event C, then the SNP B as a genetic susceptibility factor and the investigated adverse event. Since in OGSF, gene will be represented using the OGSF framework. This and adverse event are not directly linked, the social network knowledge will become an existing belief, when another analyses shows that this indirect link can be measured study A2 reached the same conclusion, this fact will be mathematically and thus provide the foundation for added into the OGSF knowledgebase and hence provide prediction. It is noted that usually only significant stronger supporting evidence to the genetic causal associations were reported in the literature, and many association. Another example is that suppose gene G is negative results may not be available. The network analysis related with both SNP B and SNP E, when another study A3 may be biased. gave the conclusion of SNP E statistically significant related 5 Ontology modeling of genetic susceptibility to adverse events following vaccination In conclusion, based on the formalization of genetic 8. Grenon P: Spatio-temporality in Basic Formal susceptibility, OGSF provides a frame work to represent the Ontology. In: IFOMIS reports. Edited by Grenon genetic allelic variants, genes and pathological processes. It P. Leipzig: 2003: 89. requires ontological scientific discourse representations as 9. Xiang Z, Courtot M, Brinkman RR, Ruttenberg A, those developed in SWAN ontology[17]. Furthermore, a He Y: OntoFox: web-based support for ontology large numbers of databases have been established in order to reuse. BMC Res Notes 2010, 3:175. establish the relation between genotypes and phenotypes. 10. [http://protegewiki.stanford.edu/wiki/OntoGraf] Some of them, such as SNPedia [18], Bio2RDF [19], Leiden 11. Bastian M, Heymann S, Jacomy M: Gephi: an Open (source) Variation Database (LOVD) [20] and GWAS open source software for exploring and central [21], support sematic web and open data technology. manipulating networks. International AAAI OGSF is aim to be an intermediate layer between Conference on Weblogs and Social Media, 2009. applications and above existing resources. 12. Smith B, Ashburner M, Rosse C, Bard J, Bug W, Ceusters W et al: The OBO Foundry: coordinated evolution of ontologies to support ACKNOWLEDGEMENTS biomedical data integration. Nat Biotechnol This project was supported by a NIH-NIAID grant 2007, 25(11):1251-1255. (R01AI081062). We would like to acknowledge with 13. Bellamy R: Susceptibility to infectious diseases : appreciation Dr. Wei Zhang, a biostatistician expert from the importance of host genetics. Cambridge, UK ; University of Michigan School of Public Health, for his New York, NY, USA: Cambridge University advice and consultation. Press; 2004. 14. Vrethem M, Malmgren K, Lindh J: A patient with REFERENCES both narcolepsy and multiple sclerosis in association with Pandemrix vaccination. J Neurol Sci 2012, 321(1-2):89-91. 1. Poland GA, Ovsyannikova IG, Jacobson RM: 15. Reif DM, McKinney BA, Motsinger AA, Chanock Adversomics: the emerging field of vaccine SJ, Edwards KM, Rock MT, Moore JH, Crowe JE: adverse event immunogenetics. Pediatr Infect Dis Genetic basis for adverse events after smallpox J 2009, 28(5):431-432. vaccination. J Infect Dis 2008, 198(1):16-22. 2. He Y, Xiang Z, Sarntivijai S, Toldo L, Ceusters W: 16. Lin Y, Sakamoto N: Genome, Gene, Interval and AEO: A Realism-Based Biomedical Ontology Ontology. In: 2nd Interdisciplinary Ontology for the Representation of Adverse Events. In: Conference: Feb 28-Mar. 1 2009; Tokyo: Keio International Conference on Biomedical Ontology: University Press Inc.; 2009: 25-34. July 26-30, 2011; Buffalo, NY, USA: http://ceur- 17. Ciccarese P, Wu E, Wong G, Ocana M, Kinoshita ws.org/; 2011: 313-315. J, Ruttenberg A, Clark T: The SWAN biomedical 3. Siber GR, Santosham M, Reid GR, Thompson C, discourse ontology. J Biomed Inform 2008, Almeido-Hill J, Morell A, deLange G, Ketcham 41(5):739-751. JK, Callahan EH: Impaired antibody response to 18. Cariaso M, Lennon G: SNPedia: a wiki Haemophilus influenzae type b polysaccharide supporting personal genome annotation, and low IgG2 and IgG4 concentrations in interpretation and analysis. Nucleic Acids Res Apache children. N Engl J Med 1990, 2012, 40(Database issue):D1308-1312. 323(20):1387-1392. 19. Belleau F, Nolin MA, Tourigny N, Rigault P, 4. Black FL, Hierholzer W, Woodall JP, Pinhiero F: Morissette J: Bio2RDF: towards a mashup to Intensified reactions to measles vaccine in build bioinformatics knowledge systems. J unexposed populations of american Indians. J Biomed Inform 2008, 41(5):706-716. Infect Dis 1971, 124(3):306-317. 20. Fokkema IF, Taschner PE, Schaafsma GC, Celli J, 5. Alghabban A: Dictionary of pharmacovigilance. Laros JF, den Dunnen JT: LOVD v.2.0: the next London ; Chicago: Pharmaceutical Press; 2004. generation in gene variant databases. Hum Mutat 6. Strachan T, Read AP: Human molecular genetics 2011, 32(5):557-563. 3, 3rd edn. New York ; London: Garland Science; 21. Thorisson GA, Lancaster O, Free RC, Hastings 2004. RK, Sarmah P, Dash D, Brahmachari SK, Brookes 7. Lin Y, Sakamoto N: Ontology driven modeling AJ: HGVbaseG2P: a central genetic association for the knowledge of genetic susceptibility to database. Nucleic Acids Res 2009, 37(Database disease. Kobe J Med Sci 2009, 55(3):E53-66. issue):D797-802. 6 7