Open PHACTS. An Introduction and Explanation March Acknowledgements: Contains contributions from across the Open PHACTS partners.

Transcription

1 Open PHACTS An Introduction and Explanation March 2012 Acknowledgements: Contains contributions from across the Open PHACTS partners.

2 Public Domain Drug Discovery Data: Pharma are accessing, processing, storing & re-processing Literature Patents PubChem Genbank Databases Downloads x each company Data Integration Data Analysis Firewalled Databases Why?

3 The Innovative Medicines Initiative EC funded public-private partnership for pharmaceutical research Focus on key problems Efficacy, Safety, Education & Training, Knowledge Management The Open PHACTS Project Create a semantic integration hub ( Open Pharmacological Space ) Delivering services to support on-going drug discovery programs in pharma and public domain Not just another project; Leading academics in semantics, pharmacology and informatics, driven by solid industry business requirements 23 academic partners, 8 pharmaceutical companies, 3 biotechs Work split into clusters: Tehnical Build (focus here) Scientific Drive Community & Sustainability The Project

4 Open PHACTS Project Partners Pfizer Limited Coordinator Universität Wien Managing entity Technical University of Denmark University of Hamburg, Center for Bioinformatics BioSolveIT GmBH Consorci Mar Parc de Salut de Barcelona Leiden University Medical Centre Royal Society of Chemistry Vrije Universiteit Amsterdam Spanish National Cancer Research Centre University of Manchester Maastricht University Aqnowledge University of Santiago de Compostela Rheinische Friedrich-Wilhelms-Universität Bonn AstraZeneca GlaxoSmithKline Esteve Novartis Merck Serono H. Lundbeck A/S Eli Lilly Netherlands Bioinformatics Centre Swiss Institute of Bioinformatics ConnectedDiscovery EMBL-European Bioinformatics Institute Janssen Pharmaceutica OpenLink The Open PHACTS Foundation

5 A user-friendly, full featured interface that allows scientists to explore and interrogate integrated biological and chemical data What will users see?

6 A Precompetitive Infrastructure Begin the task of creating an environment that can also power future collaborative efforts (public & industry) Expose Industry Experience: Create drug-discovery focused tools outside of the firewall, influenced by decades of practical experience A Pharmacology Use Case Showcase one application this technology: a stable, responsive, user-orientated system for Pharmacology Analysis A Data Publishing Methodology Develop standards and methodologies to promote good data sharing and interoperability An exemplar project for the use of the Nanopublication concept A technical approach that can be repeated in other areas OPS OPS 3 Pillars

7 Major Work Streams Build: OPS service layer and resource integration Drive: Development of exemplar work packages & Applications Sustain: Community engagement and long-term sustainability Consumer Firewall Supplier Firewall OPS Service Layer Assertion & Meta Data Mgmt Transform / Translate Integrator Corpus 1 Target Dossier Db 2 Compound Dossier Db 3 Db 4 Pharmacological Networks Std Public Vocabularies Business Rules Corpus 5 Work Stream 2: Exemplar Drug Discovery Informatics tools Develop exemplar services to test OPS Service Layer Target Dossier (Data Integration) Pharmacological Network Navigator (Data Visualisation) Compound Dossier (Data Analysis) Work Stream 1: Open Pharmacological Space (OPS) Service Layer Standardised software layer to allow public DD resource integration Define standards and construct OPS service layer Develop interface (API) for data access, integration and analysis Develop secure access models Existing Drug Discovery (DD) Resource Integration

8 Number sum Nr of 1 Question All oxido,reductase inhibitors active <100nM in both human and mouse Given compound X, what is its predicted secondary pharmacology? What are the on and off,target safety concerns for a compound? What is the evidence and how reliable is that evidence (journal impact factor, KOL) for findings associated with a compound? Given a target find me all actives against that target. Find/predict polypharmacology of actives. Determine ADMET profile of actives For a given interaction profile, give me compounds similar to it The current Factor Xa lead series is characterised by substructure X. Retrieve all bioactivity data in serine protease assays for molecules that contain substructure X. Retrieve all experimental and clinical data for a given list of compounds defined by their chemical structure (with options to match stereochemistry or not) A project is considering Protein Kinase C Alpha (PRKCA) as a target. What are all the compounds known to modulate the target directly? What are the compounds that may modulate the target directly? i.e. return all cmpds active in assays where the resolution is at least at the level of the target family (i.e. PKC) both from structured assay databases and the literature Give me all active compounds on a given target with the relevant assay data Give me the compound(s) which hit most specifically the multiple targets in a given pathway (disease) Identify all known protein-protein interaction inhibitors Business Question Based Requirements

9 A use case driven approach Main architecture, technical implementation and primary capabilities driven by a set of prioritised research questions Based on the main research questions define prioritised data sources Three Exemplars will be developed to demonstrate the capabilites of the OPS System and to define interfaces and input/output standards Three Use cases have been defined to benchmark the OPS system towards current standard workflows in data retrieval and mining

10 Example Research questions Give all compounds with IC50 < xxx for target Y in species W and Z plus assay data What substructures are associated with readout X (target, pathway, disease, ) Give all experimental and clinical data for compound X Give all targets for compound X or a compound with a similarity > y%

11 Exemplar Services Chem-Bio Navigator: querying and visualization of sets of pharmacologically annotated small molecules, on basis of chemical substructures, pharmacophores, biological activities Target Dossier: in silico dossiers about targets, incorporating related information on sequences, structures, pathways, diseases and small molecules Polypharmacology Browser: map coverage of the chemo-biological space, to facilitate the polypharmacological profiling of small molecules

12 The Semantic Web & Linked Data Annotated OPS with ontologies Linked Knowledge Harmonising data sets Concept vocabulary services Identity resolution services Concept maps Semantic Normalisation Syntactic Normalisation OPS in RDF Linked Data KEGG URI CheBI URI BRENDA URI Open Data

13 Nanopublications Capturing scientific information in the Triple Store Nanopublications & OPS (1)

14 Nanopublication facilitates the identification of individual assertions in RDF graphs. (A) represents a standard RDF conversion. In (B), colours represent corresponding data in row and graph representations. Nanopublications are an additional layer on top of RDF An implementation is the use of Named Graphs to identify specific scientific assertions This offers two major benefits 1. An ability to circle the triples which make up an individual scientific assertion (which makes viewing and understanding the central fact by both human and computers a lot easier) 2. The assignment of provenance and attribution to facts, allowing for data citability (see value of data paper on previous slide OPS has produced a series of guidelines to producing Nanopublication-RDF for data contributors Nanopublications & OPS (2)

15 Lash Up Demo Play

16 OPS GUI OPS Framework Architecture. Dec 2011 App Framework Identity & Vocabulary Management Web Service API Sparql OPS Data Model Semantic Data Workflow Engine Web Services RDF Data Cache Chemistry Normalisation & Registration Descripto r RDF 1 Descripto r RDF 2 Descripto r Nanopub RDF 3 Descripto r Nanopub RDF 4 Public Vocabularies Data 1 Data 2 Data 3 Data 4

17 OPS Framework Implementation Summary A Semantic Data Publishing approach. Individual data sets exist across the web, each is converted to an RDF representation (ideally by the data owner themselves) which includes a data set descriptor (data set meta data, including last updated information). RDF is created according to commonly used principles, including good URIs from stable, open vocabularies from the NCBO This data may also be published as Nanopublications a mechanism for assigning provenance and providing citeability and credit for data usage The Harvester monitors the sources of interest (with data-owners consent). When it identifies new data, via the data set descriptor, it loads this into the triple store. The triple store works as an application-specific cache. Only data specified by the project, required to serve specific use-cases is loaded. The cache may be obliterated and rebuilt; no primary data is stored here Chemicals are normalized and validated via the ChemSpider registration system at data loading time. Vocabulary management is provided by ConceptWiki. Resolution of multiple identifiers for the same entity and multiple URL forms for the same identifier is provided by a semantic identifier mapping service using BridgeDb. Mappings between vocabularies are provided by individual datasources and major systems such as the NCBO and ConceptWiki The Large Knowledge Collider (LarKC) is a semantic workflow engine and sits on top of the triple store. Its wraps the vocabulary and identity services as plugins, such that they can be access in-line in Sparql queries. LarKC is compatible with many triple stores allowing for different optimizations. Structure similarity searches are provided by ChemSpider web services, also wrapped as LarKC plugins and thus accessible via Sparql. Other services are similarly integrated as required User interfaces access the system via RESTful/JSON web services which provide simplified access to optimised commonly requested queries The OPS Graphical User Interface is built using a Ruby-On-Rails, ExtJS framework. The scaffold for this was seeded from the Lundbeck LSP4All system

18 RDF is our chosen format; it is well suited to describing complex data, open and supported by a growing body of tools and scientists Critical data sources are published as RDF and include data set descriptors (see here & here), which are an existing standard, promoted by Open PHACTS. This plays a major role in identifying and maintaining content in integration systems such as OPS Providing richer meta data for each dataset increases the value and scope for reuse of that data beyond Open PHACTS Producers of RDF can choose to enhance their RDF with information required to create Nanopublications, and promote the citability of data We aim to contribute to standards around RDF publishing that promote interoperability and data reuse Data Publishing Methodology

19 Main chemical needs are the registration of molecules (including validation/sanitisation) and ability to perform structure searches This is an example of an external service, provided by ChemSpider (RSC) Each database that contains molecules (SMILES, INCHI, SDF, MOL formats) must be registered with ChemSpider This process validates structures, and assigns them a unique CSID CSIDs, preferred name, synonyms, Inchi string & key and SMILES are published back into the cache Chemical structure standardisation rules are is based upon the FDAs Food and Drug Administration Substance Registration System Standard Operating Procedure, available at Some modifications for OPS have been made and these will be published shortly. ChemSpider also provides validated synonyms for many chemicals to ConceptWiki to enable entry to the system via free text typing of a compound name Chemistry Representation

20 Increased value to Pharma through The ability to use commercial and private (internal) data The ability to switch off internal systems to achieve the same aims as OPS Thus, providing a platform that allows for secure access of such data is essential Subject of a pilot study, spring 2012 The Need

21 The core platform will be built on open source technology. This includes the data harvester, the semantic workflow engine/api code, the Open PHACTS GUI and associated widgets The standards for producing RDF/Nanopublications will all be open and available An open version of the system will be available at openphacts.org, fully functional with public data Interested parties will also be able to download the core platform and instantiate it on their own servers, having everything they need to run a local system should this be required Note: OPS is decoupled from any specific RDF database engine. It should be possible to run the platform on a range of free and commercial platforms that meet certain criteria (to be published) What Is Open?

22 Associated partners Organisations, most will join here Support, information Exchange of ideas, data, technology Opportunities to demo at community webinars Need MoU Development partnerships Influence on API developments Opportunities to demo ideas & use cases to core team Need MoU and annexe MoU +Annexe Associated partners Development partnerships Consortium Consortium 22 current members Open PHACTS and the scientific community

23 Open systems need an engaged community, to grow, develop and sustain. Joining as an Associated Partner is the first step. Associated Partners: we have a MoU ready to sign for mutual support and exchange of ideas, data or technology Most will fit here, regular contact, support and review Opportunity to present ideas and use cases to core Open PHACTS team Development Partnerships: when we want to do some more specific work together, develop APIs, new data, algorithms etc. Greater access to the core of the project through agreed collaborative annexe to MoU Eg we are working actively on a certain business question How to get involved.

24 We actively manage our partners, and the wider community We term this the Open PHACTS Waiting Room, managed by a Gatekeeper Our relationships with all partners are visible at openphacts.org What we are doing together and why Opportunities to engage and develop will open and based on project needs We hold regular community workshops and events Learn more about Open PHACTS and the Open Pharmacological Space Participate in new ideas and functions Engage in development of new use cases, help us answer new questions Contribute to development, and engage in plans for sustaining the Open Pharmacological Space Please contact us to join the debate Open PHACTS Waiting Room