As with any new technology, semantic science carries great promise, as well as pitfalls for the casual user. Predictive Medicine knows the domain thoroughly and has professional contacts throughout the space.
In our judgement, the greatest single hazard in applying semantic science is the temptation to connect data in an undisciplined fashion. The technologies offer, on the surface, nearly unlimited degrees of freedom, but this freedom masks deeper issues that can burden your project with serious but unrecognized opportunity costs. There is a bit of an art to developing semantic solutions that are fit-to-purpose.
We will address:
grounding in agreed-upon use cases
ontological scope and definition of concepts
public vocabulary discovery and mappings
discrimination of concepts into individuals, classes and relations
granularity of descriptions
preparation for inferencing, if warranted
build and thorough documentation of ontologies,
preparation for launching their life cycle, and
embedding them into the technical workflow.
As a particular example application, translational medicine is a broad attempt to do two things:
1) to bring the '-omics' knowledge of the scientific disciplines to bear on the needs of the clinical workflow in a manner that is tailored to the individual patient; and then
2) to re-harvest the knowledge gained from aggregated patient outcomes data and feed it back into the scientific enterprise.
There are important 'impedance mismatches' between the domains of science and of clinical work that must be handled carefully in order to realize the maximum utility and value from the effort.
Practiced in translational medicine, Predictive Medicine excels in integrating knowledge across heterogeneous domains.
We are active in the W3C and we stay abreast of the latest semantic web developments.
We are relied upon to judge issues of semantic fit-to-purpose, granularity, provenance and more.
Our principals have helped build ontologies in biopathways, influenza and translational medicine.
Please see our paper, with Dexter Pratt from Selventa, A Comparison of BEL V1.0 and BioPAX Level 3. Selventa commissioned us to investigate relationships between their emerging Biological Expression Language (BEL) and the BioPAX standard language for describing biological pathways.