IO INFORMATICS & ASTRAZENECA SIEVE - "Safety Information Evaluation and Visual
Exploration" - COLLABORATION PRESENTED AT THE ISWC IN BETHLEHEM, PA
"Questions that used to take my team six months to answer... are now answered in six seconds."
BERKELEY, CA, BOSTON, MA, PHILIDELPHIA, PA - October 14, 2015
Robert Stanley, CEO of IO Informatics ("IO", io-informatics.com), the leader in applying
Semantic technology for Healthcare and Life Sciences (HCLS), presented the results of IOI's continuing
collaboration with AstraZeneca ("AZ", astrazeneca.com), one of the top pharmaceutical
companies globally. The 14th Annual "International Semantic Web Conference is currently underway in
Bethlehem, PA from October 11-15 (iswc2015.semanticweb.org).
"AstraZeneca's SIEVE system, building on IOI's Sentient software and services for data curation, integration
and search has been successfully applied to improve retrieval and analysis of valuable but previously
underutilized historical clinical trials data at AstraZeneca," stated Stanley. He added, "SIEVE provides a web-
based environment suitable for cross-study analysis. The environment aligns data across clinical data
including treatments, various molecular assessments and outcomes. The new SIEVE integrated datasets
support over-arching goals for biomarker identification and qualification, trial design, concomitant
medication analysis, translational and precision medicine".
IOI's Knowledge Explorer software was used by IOI's Knowledge Engineers in collaboration with AZ's
Subject Matter Experts to explore the content of these records as linked data networks using formal
Worldwide Web Consortium (w3c) semantic standards. The team has analysed and integrated over 42,000
trials records, identified by unique subjectIDs. Over 36,000 records contained valid data, each including a
unique trial, patient, and at least one row of laboratory data. IO integrated this complex and widely diverse
data under a semantic data model or "ontology", with automated import and curation functions to meet
SIEVE requirements. The data was harmonized by the application of machine reasoning or "inference", with
growing data integration and curation rules. The resulting dataset was aligned with AZ data dictionaries as
well as with public ontologies and thesauri to ensure long-term interoperability with data from other sources
within AZ. The integrated and curated data was then connected to IOI's "Web Query" software to support
rich scientific searching, reporting and analysis.
The result is a web-based User Interface accessible to end users for cross-study searching, reporting,
charting and sub-querying. Methods include "Quick Search" options, shared searches and query building
containing nesting, inclusion / exclusion, ranges, etc. Advanced Queries are presented as filters for user
entry to search subjects ("views" or "facets") including Clinical Assays, Therapy Areas, Adverse Events and
Subject Demographics. Reports include exporting, charting, hyperlink mapping and results-list based
searches. Overall results include reduced time to evaluate data from clinical trials and to facilitate forward
looking decisions relevant to portfolios.
"Our semantic technology solves difficult integration and curation problems that were prohibitively time
consuming and challenging using older methods," said Bill Hayden, IOI's Global Director of Business
Development, "but new levels of research and business efficiency and power are the outcomes that matter."
The AZ lead for the SIEVE Project stated, "Questions that used to take my team six months to answer... are
now answered in six seconds."
Substantive immediate benefits currently realized include:
- Content for critical uses is available to researchers within seconds or minutes rather than weeks / months
- User interface provides research-facing, rapid access to actual study data rather than just an index of trials
Targeted views, variables (including clinical assay, therapy area, adverse event, subject
demographics, ...), faceted search and reporting provides sensitive and specific identification of
potential biomarkers, information about adverse events, etc.
Technical longer term benefits of creating and employing the SIEVE environment at AZ include:
Semantic resource supports information scientists' needs for multi-path queries, inference-based
querying, exploratory querying
Data curation, routines and loading procedures are prepared for application to other clinical data
projects, with a growing integration and curation rule-set that draws on advanced machine
reasoning or "inference" using semantic methods
The SIEVE data resource is positioned for rapid, agile extension / federation with complementary
data repositories and databases
Both the immediate and longer term benefits result in saved time and resources for AZ. The system
provides an enhanced research environment that feeds deeper understanding of diseases, treatment
efficacy, safety and other issues required for improved drug discovery and precise, effective treatments. On
the informatics/IT side, having a semantic technology based core for data acquisition, integration and linking
will also reduce the time and resources needed to maintain the database and to connect it to other data sets
in the future.
Stanley stated,"For IO Informatics, the SIEVE collaboration represents another successful application of
semantic or "linked-data" technology to the HCLS community. The project demonstrated that semantic
technologies can provide effective integration and search across "linked-data" that is extremely diverse and
variable in structure and nomenclature. This example of how semantic technology can and does allow for
better data integration and knowledge harvesting opens up unlimited possible uses including aiding to solve
challenges in personalized medicine and big data within the pharmaceutical industry".
ISWC: October 12-15, 2015 in Bethlehem, PA http://iswc2015.semanticweb.org/
ABOUT IO INFORMATICS
IO is at the forefront of the semantic technology wave. The Company provides software and services for uniquely efficient data integration and transformative knowledge applications in life sciences, biotechnology, pharmaceutical, and medical environments. IO is headquartered in Berkeley, California with offices in Canada, Western and Eastern Europe and Asia that partner with life science and healthcare customers in industry, academia and non-for-profit foundations.
For more information visit: http://www.io-informatics.com