Assessing Relative Bioactivity of Chemical Substances Using Quantitative Molecular Network Topology Analysis
Citations Over TimeTop 10% of 2012 papers
Abstract
Structurally different chemical substances may cause similar systemic effects in mammalian cells. It is therefore necessary to go beyond structural comparisons to quantify similarity in terms of their bioactivities. In this work, we introduce a generic methodology to achieve this on the basis of Network Biology principles and using publicly available molecular network topology information. An implementation of this method, denoted QuantMap, is outlined and applied to antidiabetic drugs, NSAIDs, 17β-estradiol, and 12 substances known to disrupt estrogenic pathways. The similarity of any pair of compounds is derived from topological comparison of intracellular protein networks, directly and indirectly associated with the respective query chemicals, via a straightforward pairwise comparison of ranked proteins. Although output derived from straightforward chemical/structural similarity analysis provided some guidance on bioactivity, QuantMap produced substance interrelationships that align well with reports on their respective perturbation properties. We believe that QuantMap has potential to provide substantial assistance to drug repositioning, pharmacology evaluation, and toxicology risk assessment.
Related Papers
- → Asymmetry of Chemical Similarity(2006)18 cited
- → A new topological method to measure protein structure similarity(2003)27 cited
- → Molecular Similarity Based Estimation of Properties: A Comparison of Structure Spaces and Property Spaces(2001)23 cited
- → Characterization and Comparison of Protein Structures. Part II—Comparison(1999)4 cited
- → Genome-Wide Functional Classification/Identification of Prokaryotic Transmembrane Proteins Based on Transmembrane Topology Similarity(2003)1 cited