Dear All,

I am sending in the attachment an example for representing substructures
as promised in todays meeting. It contains substructures from Andreas
BBRC algorithm, which have additional properties (p_value: statistical
significance, effect: activating or deactivating). All information about
the feature (smarts, p_value, effect) can be found in the feature
representation.

The presence of a substructure in a compound is indicated by a feature
value of "true" in the data entries. This allows us to represent
multiple occurrences of the same feature in different compounds within
the current API.

What we would need for representing substructures are additional
ontology entries for:

ot:smarts (ot:pValue and ot:effect for fminer features)
ot:Substructure

Best regards,
Christoph

PS: With such a representation ot:Tuple and ot:ComplexValue can be
removed from the ontology (I think no one apart from me was using them
anyway).
-------------- next part --------------
A non-text attachment was scrubbed...
Name: substructure_dataset.turtle
Type: application/octet-stream
Size: 126681 bytes
Desc: not available
URL: <http://lists.opentox.org/pipermail/development/attachments/20101122/c8eaa6d3/attachment.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: substructure_dataset.rdfxml
Type: application/octet-stream
Size: 454733 bytes
Desc: not available
URL: <http://lists.opentox.org/pipermail/development/attachments/20101122/c8eaa6d3/attachment-0001.obj>