From linked open data to molecular interaction

Author(s)
Barbara Zdrazil, Eva Hellsberg, Michael Viereck, Gerhard F. Ecker
Abstract

Retrieval of congeneric and consistent SAR data sets for protein targets of interest is still a laborious task to do if no appropriate in-house data set is available. However, combining integrated open data sources (such as the Open PHACTS Discovery Platform) with workflow tools now offers the possibility of querying across multiple domains and tailoring the search to the given research question. Starting from two phylogenetically related protein targets of interest (the human serotonin and dopamine transporters), the whole chemical compound space was explored by implementing a scaffold-based clustering of compounds possessing biological measurements for both targets. In addition, potential hERG blocking liabilities were included. The workflow allowed studying the selectivity trends of scaffold series, identifying potentially harmful compound series, and performing SAR, docking studies and molecular dynamics (MD) simulations for a consistent data set of 56 cathinones. This delivered useful insights into driving determinants for hDAT selectivity over hSERT. With respect to the scaffold-based analyses it should be noted that the cathinone data set could be retrieved only when Murcko scaffold analyses were combined with similarity searches such as a common substructure search.

Organisation(s)
Journal
MedChemComm
Volume
7
Pages
1819-1831
No. of pages
13
ISSN
2040-2503
DOI
https://doi.org/10.1039/c6md00207b
Publication date
2016
Peer reviewed
Yes
Austrian Fields of Science 2012
301207 Pharmaceutical chemistry
ASJC Scopus subject areas
Biochemistry, Pharmaceutical Science
Portal url
https://ucris.univie.ac.at/portal/en/publications/from-linked-open-data-to-molecular-interaction(7ec9296c-3a4e-454c-9bc0-c72f060ee6e0).html