Advances in the biological sciences are allowing pharmaceutical companies to meet the health care crisis with drugs that are more suitable for preventive and tailored treatment, thereby holding the promise of enabling more cost effective care with greater efficacy and reduced side effects. However, this shift in business model increases the need for companies to integrate data across drug discovery, drug development, and clinical practice. This is a fundamental shift from the approach of limiting integration activities to functional areas. The Linked Data approach holds much potential for enabling such connectivity between data silos, thereby enabling pharmaceutical companies to meet the urgent needs in society for more tailored health care. This paper examines the applicability and potential benefits of using Linked Data to connect drug and clinical trials related data sources and gives an overview of ongoing work within the W3C's Semantic Web for Health Care and Life Sciences Interest Group on publishing drug related data sets on the Web and interlinking them with existing Linked Data sources. A use case is provided that demonstrates the immediate benefit of this work in enabling data to be browsed from disease, to clinical trials, drugs, targets and companies.
The Linking Open Drug Data (LODD) task within the W3C's Semantic Web for Health Care and Life Sciences Interest Group1 gathered a list of data sets that include information about drugs, and then determined how the publicly available data sets could be linked together. The review showed that this domain is promising for Linked Data as there are many publicly available data sets, and they frequently share identifiers for key entities. The complete evaluation results are posted on the W3C ESW Wiki2. Participants of the LODD task have undertaken to demonstrate the value of Linked Data to the health care and life sciences
2 http://esw.w3.org/topic/HCLSIG/LODD/Data/DataSetEvaluation domain. This has been achieved by publishing and linking several drug related data sets on the Web, and investigating use cases that demonstrate how researchers in life science, as well as physicians and patients can take advantage of the connected data sets. This paper is structured as follows: Section 2 describes the published data sets, their linkage with other published data sources, and the methods that were used to create the links. Section 3 exemplifies how navigating linked data can be utilized within a competitive intelligence use case. While Section 4 summarizes our findings and experiences from publishing and navigating the data sets.
In this project, data about pharmaceutical companies, drugs in
clinical trials, mechanisms of action of drugs, safety information,
and data about disease gene correlations were added to the Linked
Data cloud. This selection of data sets enabled strong connections
to existing Linked Data resources, while providing novel data of
interest to the pharmaceutical industry. The existing Linked Data
of primary interest to this work includes the many bioinformatics
and cheminformatics data sources published by Bio2RDF [6], and
the information on diseases and marketed drugs in DBpedia [
The Linked Clinical Trials (LinkedCT) data source 3 is derived from a service provided by U.S. National Institutes of Health, ClinicalTrials.gov, a registry of more than 60,000 clinical trials conducted in 158 countries. Each trial is associated with a brief description, related disorders 4 and interventions, eligibility criteria, sponsors, locations (investigators), and several other pieces of information. The data on LinkedCT is obtained by first transforming the XML data provided by ClinicalTrials.gov to relational data using the capabilities of a hybrid relational-XML Relational Database Management System such as IBM DB2. This transformation requires identification of the entities and facts in the XML data and storing them in reasonably normalized relational tables that are appropriate for transformation into RDF. The RDF data is then published using D2R server [8]. The RDF version of the dataset contains 7,011,000 triples and 290,000 links.
DrugBank [
of disorders, disease genes, and associations between them was obtained from the Online Mendelian Inheritance in Man (OMIM)7, a compilation of human disease genes and phenotypes. The data set is published by Diseasome in a flat file representation. The flat files were read into a relational database and made accessible as Linked Data using D2R server. The Linked Data version of Diseasome contains 88,000 triples and 23,000 links8.
DailyMed9 is published by the National Library of Medicine, and provides high quality information about marketed drugs. DailyMed provides much information including general background on the chemical structure of the compound and its mechanism of action, details on the clinical pharmacology of the compound, indication (disorder) and usage, contraindications, warnings, precautions, adverse reactions, overdosage, and patient counseling. The data was originally published in Structured Product Labeling 10 , a XML-based standard for exchanging medication information that has been recently introduced by the Food and Drug Administration in the United States. It was published using the D2R server. The Linked Data version of DailyMed contains 124,000 triples and 29,600 links11. There are many commonly used identifiers in the life sciences that can be utilized for making links between data sets explicit. Links that were generated based on shared identifiers include the connections from LinkedCT to Bio2RDF's PubMed, and from DrugBank to DBpedia. The connections between bioinformatics and cheminformatics data sources are already provided by Bio2RDF allowing us to interlink our drug-related data sets to their work. In cases where no shared identifiers exist, string and semantic matching techniques were applied for link discovery
10 http://www.fda.gov/oc/datacouncil/SPL.html
11 http://www4.wiwiss.fu-berlin.de/dailymed/
[
A use case has been developed that demonstrates the value of Linked Data about drugs to the pharmaceutical industry. Departments within pharmaceutical companies have typically decided independently which data sets need to be brought into their organization for integration and interrogation. Access to the data is provided to employees based upon their roles. The use case describes the value that can be gained by allowing employees to gain access to a more diverse and linked body of data. This approach enables new and novel questions to be explored. The following use case describes a scenario in competitive intelligence.
A neuroscience focused business manager is interested in seeing an update on new clinical trials that competitors are starting in Alzheimer’s Disease (AD). These updates influence future sales forecasts across geographies, and impact portfolio decisions as new drugs needs to demonstrate improved safety and efficacy compared to the existing pharmacopeia.
Using a Semantic Web browser of choice – for instance Tabulator12 or the Marbles data browser13, the manager is able to see all drugs in trials for AD in LinkedCT, including a new phase III trial planned by Pfizer for a drug called Varenicline. The business manager can see that more information is available about the drug, which is unusual because not much data is typically available for drugs that are under investigation. Following the data link the manager sees data from DailyMed that shows that the drug is already on the market for nicotine addiction. As side effects are better understood for drugs that are already on the market, they tend to be more successful in trials. Out of curiosity, the manager scrolls down the page to see that side effects are listed as constipation, sleeping problems, vomiting, nausea, and gas; and that the typical dose is 1mg twice daily. The dose stated on LinkedCT for the trial was no higher than that, so it is unlikely that this drug will have new safety problems. 12 http://www.w3.org/2005/ajar/tab 13 http://beckr.org/marbles Link Type owl:sameAs owl:sameAs rdfs:seeAlso owl:sameAs owl:sameAs owl:sameAs
foaf:page drugbank:possible
DiseaseTarget drugbank:branded
Drug owl:sameAs drugbank: pfam DomainFunction drugbank:enzyme
SwissprotId drugbank:iupacId drugbank:pdbId
Count
27,685
12,127
8,848
444
301
42,219
61,920
8,201
1,593
1,522
19,028
4,660
4,592
Given the promising safety profile, the manager is curious to
discover why a nicotine addiction drug might work for AD.
Linking to DrugBank highlights to the manager that Varenicline
is an alpha-4 beta-2 neuronal nicotinic acetylcholine receptor
agonist. However, Diseasome indicates that the corresponding
genes are only important in nicotine addiction, rather than AD.
This suggests that there is a more complex relationship between
the diseases, than just sharing a drug target. Extending the
browsing to the SWAN Knowledgebase14 [
Using the Linked Data approach a business manager was able to browse data relating to companies, clinical trials, drugs, diseases and genetic variation. More specifically, the manager was able to determine when extra data was available, gain access to data without needing to map different identifiers and synonyms, and gain additional insights as to interesting questions to ask.
This paper describes the mapping of four drug related data sources into the Linked Data cloud, and the ensuing insights that can be gained in the area of competitive intelligence. However, this is just the beginning, because more interesting and novel questions will be able to be addressed as additional data sets are added. As a next step, it would be interesting to incorporate data relating to epidemiology, as that could provide information relating to geographical areas in which diseases are prevalent, and where there is a strong need for the development of a drug that meets the needs of a specific population. It would also be valuable to create links to the AD hypotheses data that is in RDF within the SWAN Knowledgebase.
Pharmaceutical companies need to make decisions based upon both internal and external data, it is therefore important that companies begin to make internal data available in a linked representation, both to break down the internal silos and to easily connect with external data. Such an approach would require organizations to understand where the linkage points occur across internal data sets, but this is ongoing work as it is a critical prerequisite for all data integration efforts relating to the effective tailoring of drugs.
Currently, when pharmaceutical companies bring copies of data
within their organizations for integration, they each need to have
experts who understand the connectivity across data sets.
However, with the Linked Data approach, this responsibility is
shifted to the data providers. This is a much more efficient
approach, as the data providers are the individuals who
understand the data best. It also means that the integration only
has to happen one time. In addition, it becomes possible for data
providers to incrementally add links to new data sets as they
become aware of their existence, rather than needing to design a
model to do everything in one go. As stated in [
As the Linked Data cloud grows, focus in pharmaceutical companies will be moved to approaches for interpretation. One project with potential to utilize the value from Linked Data is the Large Knowledge Collider (LarKC), a platform for massive distributed incomplete reasoning that aims at removing the scalability barriers of currently existing reasoning systems for the Semantic Web15.
The Linked Data approach is very promising for the pharmaceutical industry, and its value will increase as more data sources become available. However, our technical work as well as use case experiments revealed various challenges that need to be mitigated to make this approach robust enough to be deployed within an enterprise environment: 1.
Progress needs to be made in finding links between data
items across data sets where no commonly used identifiers
exist. Discovering such links requires using specific record
linkage [
frameworks for setting links within the LODD data sets, domain experts need to identify linkage points and specific rules required for finding the links.
Work needs to be undertaken to make data browsers more
robust and performant. In addition, the user interface of data
browsers needs to be improved. Life Sciences data
frequently consists of long lists of entities (e.g. genes, trials,
diseases, patients) that need to be browsed, filtered, and
queried. Benefits would be gained if hybrid interfaces that
combine querying and browsing would be available and able
to process the large amounts of data that are typically
relevant within this domain. For such interfaces, it could be
promising to combine live data retrieval with local caching
and in-advance crawling of relevant data sets, as it is
currently done by Semantic Web Search engines such as
Sindice [
A significant challenge within the life sciences and health care is the strong prevalence of terminology conflicts, synonyms, and homonyms. These problems are not addressed by simply making data sets available on the Web using RDF as common syntax but require deeper semantic integration. For applications that focus on discovery and data navigation, having explicit links between data sources is often already a huge benefit even without semantic integration. For other applications that rely on expressive querying or automated reasoning deeper integration is essential. In order to also provide for such applications and lay the foundation for fusing data from several Linked Data sources, it would be beneficial if more community practices on publishing term and schema mappings would be established.
This work was undertaken within the LODD task of the W3C's Semantic Web for Health Care and Life Sciences Interest Group. Significant contributions to the LODD task have also been made by Kei Cheung, Don Doherty, Matthias Samwald, and Jun Zhao. Anja Jentzsch and Chris Bizer received funding for this work from Eli Lilly.
[1] Healthcare 2015: Win-win or lose-lose?
www.Ibm.com/healthcare/hc2015. [2] Gerhardsson de Verdier, M.: The Big Three Concept - A Way to Tackle the Health Care Crisis? Proc. Am. Thorac.
Soc. 5: 800–805, 2008. [3] Andersson B., Momtchev V.: D7a.1.1 LarKC Requirements summary and data repository, http://wiki.larkc.eu/LarkcProject/WP7a. [4] Sharp, M., Bodenreider, O., and Wacholder, N.: A framework for characterizing drug information sources. AMIA Annu. Symp. Proc. 2008 Nov 6:662-666.
http://www.ncbi.nlm.nih.gov/pubmed/18999182. [5] Goble, C., Stevens, R.: State of the Nation in Data Integration for Bioinformatics. J. Biomed. Infor. 41: 687693, 2008. [6] Belleau F., Nolin., M.-A., Tourigny N., Rigault, P., and Morissette, J. Bio2RDF: Towards a mashup to build