AstraZeneca
Case Study: AstraZeneca Start a Journey With Persistent Identity
AstraZeneca are a global research-based biopharmaceutical company. Their skills and resources are focused on discovering, developing and marketing medicines for some of the world’s most serious illnesses, including cancer, heart disease, neurological disorders such as schizophrenia, respiratory disease and infection.
Bringing a new drug to market takes time – more than 10 years. It also takes money – up to $1 billion before the first sales are made. But most importantly, it takes skill, experience and collaboration – working with stakeholders and partners to identify needs and meet them with medicines that make a real difference.
R&DI engaged Talis Systems Ltd, to investigate how a Semantic Web based technology approach called Linked Data could allow their systems to identify and share data more easily
Around 15,700 people work in AstraZeneca’s Research and Development (R&D) organisation and they have 14 major R&D centres in eight countries, including Sweden, the US and the UK. This means that as an organisation they handle and generate massive amounts of data. This data is typically locked away behind propriety systems that may or may not be easy to interface with. In some cases AstraZeneca have built their own systems to merge sets of data in ways that are useful to specialists in their research and development communities.
AstraZeneca’s Research and Development Information group (R&DI) have internal customers throughout the AstraZeneca business and are required to provide targeted information views that deliver data to support project tracking and decision management. The systems they deploy are required to share data and common taxonomies.
An array of independent information storage silos requires complex workflows to extract data from the source system, transform it into another format and then load to destination systems. Every time AstraZeneca want to share data between organisations, they have to scope and develop customised processes. These ‘start from scratch’ approaches are expensive to produce and maintain.
AstraZeneca, like any other large organisation need to keep track of where things are and what they are. For example, they want to be able to track which drugs have been used in which trials because reinventing the wheel is costly. This has given AstraZeneca a focus on persistent identity – being able to always refer to something with the same identifier throughout the drug development lifecycle.
R&DI have been investigating the use of new systems to simplify the identification and sharing of relevant data across systems and have been aware of the pharmaceutical informatics community’s explorations around Semantic Web technologies for some time.
R&DI engaged Talis Systems Ltd, to investigate how a Semantic Web based technology approach called Linked Data could allow their systems to identify and share data more easily. Linked Data is Tim Berners-Lee’s label for an approach to making data reusable by linking individual data points with descriptive relationships.Talis’ Tim Hodson and Keith Alexander worked with Mike Westaway to explore possible areas in which R&DI could show that Linked Data methodologies could provide the flexible, low risk data reuse their internal customers required. The consultants used workshops and interviews to explore AstraZeneca’s business needs and to identify some initial areas in which R&DI could start to use a Linked Data approach. The initial report provided management with a set of options to focus attention toward achievable proof of concept ideas.
Interviews between Tim Hodson and R&DI’s domain experts, explored persistent identity and what that meant to AstraZeneca’s research. The technical mechanisms for implementing a persistent identity scheme were explored using proof of concept deliverables that allow R&DI to take the conversation forward internally.
The hands on approach that Talis brought to their consultancy was appreciated by those involved in the proof of concept building. Mike Westaway, Chief Domain Architect for Strategic Programmes, R&DI says “Traditionally, AstraZeneca has used an Extract Transform Load paradigm and Data Warehousing to analyse Research and Development data. However, there is growing interest in applying semantic approaches to integrate internal and external sources and to provide connections between targets, pathways, compounds, and diseases. We are assessing the value of persistent identity as an enabler for solutions which query across multiple information sources.”
Print this page
Who worked on it?
Work with us
Talis Consulting are a team of expert Consultants who have experience in delivering Linked Data projects; this is imperative when working to help navigate the intricacies and nuances of implementing a successful Linked Data project, without this expertise we aren't able to work to ensure your success.
In support of the Consulting services, Talis also provides a service enabling the hosting of Linked Data, utilising the features of the Talis Platform as a Software as a Service solution.
If you would like to start your Linked Data adoption, please contact us.
