DNAdigest interviews Open PHACTS
Open PHACTS is a member funded project looking to make life science data more easily accessible and interoperable to reduce barriers to research. Today, we interview Nick Lynch from Open PHACTS about the project and the Open PHACTS Foundation.
1. What is the Open PHACTS Foundation? What does it do? What are its goals and missions?
The Open PHACTS Foundation is a charity supported by paying members, it was established to sustain Open PHACTS once the original funded project ends. Our mission is the same as the Open PHACTS project: making life science data more easily accessible and interoperable to reduce barriers to research. More specifically our aim as a charity is, “The advancement of science for the public benefit through the sharing of knowledge and data in relation to life science and biomedical research,” and in practice we do this by sustaining and developing the Open PHACTS Discovery Platform.
2. Could you introduce us to the Open PHACTS Discovery Platform? How did the idea of it arise? When was it built? How do you plan to further improve it?
A major challenge to drug discovery and many other kinds of life science research, is the fact that a lot of the data researchers need is locked up in isolated “silos” of information. A given compound can have many different names or identifiers, each used by different data providers, so finding all the information that’s available about a specific compound is not a trivial problem.
Some pharmaceutical companies have built their own systems for integrating different sources of data, which costs time and money and duplicates effort. For SMEs and academic researchers, this kind of data integration is often too costly to do at all. The aim of the Open PHACTS project was to address this problem by mapping and integrating data from multiple publicly available and pre-competitive sources.
We specifically focussed on researcher needs to build the platform: we sat down and asked what sort of questions they wanted to be able to answer, and came up with a list of prioritised examples like, “Give me all oxidoreductase inhibitors active < 100 nM in human and mouse.” We started building the platform in 2011, and our first public version went live in 2012 with regular releases since. Development is always ongoing. Right now we’re looking into how a company can integrate their proprietary commercial data with Open PHACTS for in-house research, and how we might extend the platform’s coverage – earlier in the year we hosted a workshop to discuss the big data challenges around phenotypic screening, and what Open PHACTS might be able to do in that area.
3. What is your role both in the Open PHACTS Foundation and the Discovery Platform?
Rather than focus on the individual, I rather describe how Open PHACTS has brought together a team of people and partners to support our goals.
- Technical Team: Our semantic platform has a number of key components that support its key goals including the ability to map identifiers for the same concept across a range of data sources allowing the user to search by many different URIs or terms. We also support text search.
- Data Team: Our data team works with new data sources and recommends new data that should be added to the platform to support new questions that will help our research customers. We also make suggestions as to the best ontologies to use that will support future easier integration in & outside Open PHACTS.
- Support Team & Outreach: Outreach and Support to help people get the most from the data and the tools is important for the sustainability of any platform. Often with large informatics projects the support aspect is sometimes downplayed, but if a platform is to be sustainable longer term it has to ensure the users are getting best value.
- Wider community including Semantic and Data Provider experts: From the start of the project we have been lucky to have a wider group from the core data sources (e.g. WikiPathways, ChEMBL, DisGeNet) and equally Semantic experts that have helped the project progress. This has also allowed a range of further research to be done and papers to published – see here
- Developer Community & App EcoSystem: The project has encouraged and enabled an ecosystem of Apps and other informatics tools to be created that use Open PHACTS in combination with their own work and workflows.
4. What are the different tools that the Open PHACTS Discovery Platform offers (e.g. The Explorer, the API etc.)? Would you describe them and their benefits briefly for us, e.g. by an example of a successful use case?
The Explorer is a browser-based interface for researchers to search the data integrated in Open PHACTS. Rather than looking up the same compound in multiple different databases, a single search in the Explorer connects data about a given compound from major publicly available databases, which you can then browse and filter.
For more complex queries, we have Pipeline Pilot components and KNIME nodes for users to access Open PHACTS data directly through our API. Anyone can sign up for up to five free API access keys. If you already use Pipeline Pilot or KNIME, these components and nodes can be integrated right into your existing workflows.
Several Open PHACTS project members have also built customised apps to directly access the API, for specialised visualisation or analysis. ChemBioNavigator is one example which allows researchers to visualise data about groups of related molecules in chemical and biological space.
Plus of course users can access the API directly through our active documentation – or even develop apps and workflow tools of their own to interface with Open PHACTS.
5. Could you tell us about your partnership with the Innovative Medicines Initiative?
Our partnership with the Innovative Medicines Initiative (IMI) started when the initial Open PHACTS project started in 2010 and it was through the partnership and joint funding (Private/Public) of the EU and EFPIA (European Pharmaceutical Companies) members that prioritised the project as part of IMI. More details on IMI can be found here.
IMI is now moving ahead with a second phase of projects and Open PHACTS is involved in some of the early ideas since we want to make the most of the platform that has been created for data hosting and sharing.
6. Why do you think Open PHACTS is important for researchers and what are the benefits of becoming a member of the Open PHACTS Foundation?
By making integrated life science data open and free, Open PHACTS reduces barriers to research in big pharmaceutical companies, and offers many SMEs and academic researchers unprecedented access to integrated data. This makes life science research easier and more accessible for everyone.
Access to the Open PHACTS Discovery Platform and API will always be free, in line with the Foundation’s goals as a charity. However, becoming a member of the Open PHACTS Foundation means you will be directly involved in decisions about the platform’s future development: what features to build, what gaps to address, what kinds of data to focus on, etc. You and your organisation’s needs will be prioritised not just in terms of development, but with priority API access, technical support, training, and access to new features.
7. Do you think that the problem of limited data sharing affects the work of the Open PHACTS Foundation or the Discovery Platform? If yes, how exactly?
Open PHACTS relies on sharing of data through curation of published data into publically available data sources. This process of curation is often hard and time consuming and so not all data from research is necessarily published in a format that makes its sharing easy.
We certainly would like more data to be published in a format that can be added to our linked data store and the easier that is the better it is for our users and hopefully helps them with their questions in the future. Open PHACTS now also allows you to enter your own private data into the platform to allow questions to be asked of Private and Public data at the same time. This helps a company compare its research data with external data.
8. What is your opinion on ethical data sharing and what do you think could be done in the future to improve it?
We believe ethical data sharing should be a key tenet of science and healthcare research as the process of research relies on building on the research of others and being able to reanalyse data created by others.
As an encouragement for those sharing data, we need to reward them with new grants to continue what they are doing and support the services & tools they have put in place. As life science data becomes more integrated and the need to bring together data from clinical studies and also personal data grows, having clear rules for publishing data is very important for building trust in the health research systems. We are certainly keen to participate in this in the future and support these efforts.
Are you part of a project that facilitates data sharing for genomics research? Would you like to be featured on our blog? We would love to hear from you. Drop us an email at firstname.lastname@example.org or use our contact page to get in touch.