Fig. 1
From: Network-based analysis of transcriptional profiles from chemical perturbations experiments
Analysis workflow. DrugMatrix liver samples are used to infer chemical-specific Compound Networks (a) and a Control Network (b). Similarity in terms of network structure is evaluated to identify groups of compounds, whose samples are pooled to infer Aggregate Compound Networks (c-e). Modules of tightly connected genes both in the Control network and in Compound Aggregate networks are identified and compared across conditions in terms of Module Differential Connectivity (MDC) (f). Modules with a change in connectivity that is highly specific to each compound group are investigated through pathway enrichment analysis (g-h)