A molecular representation has already been created for a significant fraction of the COD (above 160,000 entries at the moment of writing this article), this COD subset hence being available for substructure search.
The limitations of inconsistent molecular representations and inconsistent molecular relationship representations have been overcome by our normalization of the data into common gene name representations and common relationship types which are compatible with other database.
Social representations have four main functions [ 35].
The PubChem molecular structure representation has a larger alphabet of bond types than that which is in common use by most software, which can in principle be used to good effect, but in practice this is not necessarily the case.
Our representations had 50 dimensions.
Previous representations had varying levels of success.
A number of molecular representations and similarity coefficients have been proposed [1] to quantify the molecular similarity between single molecular structures and compound libraries.
In addition, as has been well documented, similarity measures depend critically on the molecular representations and similarity coefficients used [31, 32].
However, these representations of molecular networks have the limit of being static and largely qualitative.
Traditionally, representations of molecular pathways have been produced ad hoc and frequently included in reviews and original papers.
