|
Abstract:
|
Integrating gene expression profiles with certain proteins can improve our understanding of the fundamental mechanisms in protein-ligand binding. This paper spotlights the integration of gene expression data and target prediction scores, i.e., a score quantifying the compound's binding likelihood to the target protein given its chemical structure, providing insight into the compounds' mechanism of action (MoA). Compounds are clustered based on the similarity of their predicted targets and each cluster is linked to a set of differentially expressed genes using Linear Models for Microarray Data (LIMMA). MLP analysis is used to identify gene sets based on their biological processes that are enriched in small p-values from LIMMA. A qualitative search is also performed on the target-based compound clusters to identify pathways. Three compound clusters are studied: (i) the target-driven cluster involving HSP90 inhibitors, geldanamycin and tanespimycin, (ii) The antipsychotic cluster and (iii) the thiazolidinediones drug cluster. This analysis, reveals the underlying MoA of a compound by establishing the link between its target prediction, affected genes and pathways.
|
