--GENERAL INFORMATION--
TITLE: Text-mined orthosteric and allosteric compound dataset
CONTRIBUTOR: Leiden Academic Centre for Drug Research, Leiden University 

This dataset contains the annotation of orthosteric and allosteric compounds, which were retrieved using text mining.  

--METHOD--
Data for this dataset was derived from ChEMBL (version 22) using text mining. The title and abstract of each publication in ChEMBL was searched for keywords to annotate orthosteric and allosteric compounds.  

--DATA SPECIFIC INFORMATION--
The dataset contains the following properties/columns:
CHEMBL_ID_compound - compound identifier, similar as used in the ChEMBL database
year - the publication year
CHEMBL_ID_protein - the protein identifier, similar as used in the ChEMBL database
activity_id - activity identifier, similar as used in the ChEMBL database
L1 - protein familiy level 1, similar as used in the ChEMBL database (version 22)
L2 - protein familiy level 2, similar as used in the ChEMBL database (version 22)
L3 - protein familiy level 3, similar as used in the ChEMBL database (version 22)
binding_type - text-mined annotation of binding type
pchembl_value - activity value of the given compound-protein combination
canonical_smiles - chemical structure of the compound