Biomedical Engineering Reference
In-Depth Information
published within Cancer Commons, and these can be discussed as well. In
addition to displaying and computing with the molecular disease model, the
platform provides a means for members of the community to discuss and
suggest refi nements to the model. These are fed back to the curators for con-
sideration and potential revision of the model and can be read by any member
of the community.
The rapid communitywide interactive publication model of Cancer Com-
mons overcomes three major issues that weigh down traditional biomedical
communications. First, while formal, peer-reviewed, professionally copy-edited
research articles are integral to medical research, they are not an effi cient
means of disseminating timely knowledge that may have immediate clinical
utility for patients. Second, with tens of thousands of peer-reviewed publica-
tions, abstracts, conference proceedings, and the like, there is far too much
information for anyone to absorb and meaningfully apply in making treatment
decisions. Finally, in the process of preparing formal papers, much information
is delayed or, worse, lost. Interim data and results are typically discarded,
especially the results of failed experiments and clinical trials, dooming others
to waste time rediscovering them again—a tragedy when patients are involved.
Although journals and funding agencies are committed, in principle, to requir-
ing that data associated with publications be made available, in practice, this
only succeeds in the few cases for which community-endorsed repositories
exist, for example, those maintained by the NCBI. Beyond access to data, there
is the deeper issue of making the conclusions conveyed in a scientifi c paper
available in a structured form that can be understood and manipulated by
computers as well as by human scientists.
In Cancer Commons, the equivalent of peer review occurs continuously
through discussion forums. The experts who curate the molecular disease
model aggregate this information and make it actionable for patients with
specifi c molecular subtypes of cancer, and revisions are disseminated through
the Career Commons rapid communitywide interactive publication model. As
such, the molecular disease model becomes, in effect, a dynamic review article
subject to continuous review and revision based on the latest clinical and labo-
ratory fi ndings, with all conclusions and supporting data available in a struc-
tured format suitable for computation.
All contents of Cancer Commons are available for semantic annotation so
that they can be brought immediately to the attention of all researchers, clini-
cians, and patients for whom that information may be relevant. Beyond ensur-
ing timely access to knowledge by humans, semantic annotation is also the key
to making that knowledge machine understandable.
The Cancer Commons platform includes applications that enable partici-
pants to not just annotate papers but also comment on the annotations of
others and weave them into structured arguments that support and refute
hypotheses [9]. Applications can use this structured knowledge, for example,
to track and prioritize targets, leads, and trials or suggest experiments to vali-
date clinically observed responses. Capturing and using this structured dis-
Search WWH ::
Custom Search