We believe RDF/OWL and
LSID can help Pharma out from the InfoMaze!
Position
paper for, W3C Workshop on Semantic Web for Life Sciences
Kerstin Forsberg
and Bo Andersson, AstraZeneca R&D
One
Challenge
It’s a well known fact
that pharmaceutical industry are facing huge challenges in bringing new complex
drug candidates to the market. The break through in genome technology have
resulted in many new potential drug targets, however our knowledge about these
new opportunities and the system biology behind them are limited. We have a
knowledge gap to fill, however available tools are from the pre genome era.
Pharma need a new toolbox
addressing these challenges, where short sighted fixes are replaced by long term
foundations!
What we want to
do
We want to establish an
innovative and collaborative environment where scientists from discovery and
development navigates between different information sources needed, utilizing
appropriate tools to give them the ability to take informed
decisions.
The toolbox shall provide
not only the tools to access, analyse and visualise, but also the protocols and
models giving the ability to discover, combine and interpret the information
from heterogeneous sources. Information sources necessary to handle through this
“plug & play” alike foundation are both internal and external including
information from the health care environment. The toolbox shall address needs in
both the submission process and for explorative research.
In this vision we can
imagine;
§
Clinical documents being
semantically annotated both for regulatory purpose and from different scientific
perspectives thus being made easily exploitable. This will open up the huge
source of knowledge within existing document management systems for both
scientific questions and best practice advances.
§
Clinical study data being
readily available for scientific inference and reasoning anywhere and at any
time as needed. Clinical data sources would feed the explorative environments
with data required for analysis and reporting as well as facilitating hypothesis
generation throughout the drug development process.
Where W3C fit
in
Pharma need a foundation
based on technologies to organise data and documents in prospective structures
to improve submission process and open up for evolutionary knowledge creation
based on enrich information with the concept of
assertions.
Semantic interoperability
is a key driver for this foundation to enable a “plug & play” interface to
information; to be useful cross contexts and time, to be machine
interpretable and human
understandable.
We believe the power of
Semantic Web, and the standards, RDF, OWL as well as the LSID approach and the
ISO11179 model for metadata registries, lies in the principle of minimalist
design based on a common model of great generality. Only when the common model
is general can any prospective application be mapped onto the model. Therefore
we foresee these standards being core both for health care and clinical domain
specific standards as their representation format.
Summary
We imagine the challenge,
the idea how to address it and how we believe W3C play a role are
uncontroversial.
However, how can we make
sure that the health care environment and the pharmaceutical industry adopt this
thinking about filling existing gaps in our toolbox.
To discuss at the
workshop;
§
Do we have the standards
we need.
§
How do we secure
interoperability cross contexts and time.
§
How can we get a global
adoption of these standards.
§
How can we secure
available tools for handling data according to these standards, i.e. open not
proprietary.
§
Etc.