T a p<0.05 (T-test control vs MMS-treated) and fold>2 or <-2 in at least one time point were selected as MMS-altered genes. Repository information: GSE57801; http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token= apmtuguyplcnbgh acc=GSERNAi screening datasetGenes whose knockdown caused sensitivity to MMS in Kc167 cells were obtained from our previously published wide-genome RNAi screening in D. melanogaster Kc167 cells as detailed in [1]. This RNAi screening tested 13,826 open reading frames, of which 996 FBgn (FlyBase gene ID, denotes known genes) affected MMS survival; 537 of which were further examined with 202 validating [1]. Microarray detected 12,363 FBgn, with 95 of these had a matched probe with the RNAi screening platform therefore making fusion of RNAi and microarray gene lists feasible.PLOS ONE | DOI:10.1371/journal.pone.0153970 April 21,3 /Gene Expression and RNAi Data FusionPathway Enrichment Analysis (PEA)For determination of pathways associated with MMS-induced gene expression changes in fly Kc167 cells, DEGs from 8, 24 and 72 h treatments were combined into a single gene list prior to PEA. In mammalian cells (MEF and MDA-MB231), the DEG list comprised both 8 and 24 h MMS induced gene expression changes. Differentially expressed genes (DEGs) were analyzed using the Ingenuity Systems (IPA, Ingenuity Pathway Analyzer, Qiagen), and enrichment significance was calculated using the right-tailed Fisher Exact Test. Ingenuity analysis was complemented with the publicly available DAVID Bioinformatics Functional Annotation tools set as: i) Pathways; ii) KEGG and Reactome databases. In DAVID, pathway terms in the output were considered as significantly enriched at Trochol chemical information 21092707″ title=View Abstract(s)”>PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/21092707 a p<0.05 and FDR<10 . Uninformative pathway terms such as disease-related or cell-line specific pathways (for example: "Prostate cancer" and "Melanocyte Development and Pigmentation Signaling") were excluded. Both fly and human orthologues were used to determine pathway enrichments in DAVID and Ingenuity tools as appropriate for each database. In some figures, Ingenuity canonical pathway charts were used to facilitate visualization of MMS-induced alterations with NRF2 and ER stress pathways.Gene/Protein Interaction Networks and Landscape AnalysisProtein interaction network was generated using the STRING database [Parameters: i) input: MMS-induced genes and survival hits identified by PEA with each pathway, and converted to human orthologs; ii) Prediction methods: databases, gene fusion, experiments, textmining and co-occurrence; iii) confidence level 0.400 (medium)]. Fly gene/protein interaction networks were built by associating orthologs belonging to NRF2/oxidative stress and UPR-ER stress pathways selected from literature review (Pubmed, Ingenuity and Qiagen databases) as detailed in S3 Table. In summary: i) NRF2-GSH pathway: components of the transcription factor core signaling, NRF2 transcriptional targets, genes involved in GSH synthesis and GSH-mediated detoxification; ii) ER stress/UPR: chaperones and heat shock factors, ER sensors, and ER-toNucleus signaling transducers. Briefly, the network was generated using the STRING database [Parameters: i) input: fly orthologs; ii) Prediction methods: databases, gene fusion, experiments, textmining and co-occurrence; iii) confidence level 0.700 (high)]. Network parameters were saved and the links (interaction strength) were handled by Medusa software [5]. Nodes colors include information on MMS-induced changes in gene expres.