Network perturbation analysis of gene transcriptional profiles reveals protein targets and mechanism of action of drugs and influenza A viral infection

Noh, Heeju; Shoemaker, Jason E; Gunawan, Rudiyanto

Another group of network-based analytical methods utilizes cellular network graphs, either curated from the literature knowledge or inferred from gene expression data, to formulate statistical hypothesis tests for ranking drug targets (3, (16) (17) (18) (19) (20) (21) (22) . Several methods in this category prioritize targets based on the enrichment of the downstream or neighbouring molecules in the network for differentially expressed genes, following the principle of 'guilt by association' (3, (16) (17) (18) (19) (20) . Another set of methods rank targets by scoring hypotheses that are generated based on causal relationships in the biological networks (21, 22) . A recent method called Detecting Mechanism of Action by Network Dysregulation (DeMAND), combines the GRN and PIN to create a molecular interaction network, where the drug targets are scored based on the statistical significance of drug-induced alterations in the joint gene expression distribution between two connected genes in the network (23) . The methods in this group make use of only the (static) topology of cellular networks without much consideration of the dynamics of the gene transcriptional process, and thus are unable to fully exploit information contained in time-series datasets.