Comment[ArrayExpressAccession]	E-GEOD-36243
MAGE-TAB Version	1.1							
Public Release Date	2012-11-16
Investigation Title	Transcriptomic response to benzo[a]pyrene treatment in HepG2 cells (Affymetrix)							
Comment[Submitted Name]	Transcriptomic response to benzo[a]pyrene treatment in HepG2 cells (Affymetrix)
Experiment Description	Whole-genome transcriptome measurements are pivotal for characterizing carcinogenic mechanisms of chemicals and predicting toxic classes, such as genotoxicity, from in vitro and in vivo assays. DNA microarrays have evolved as the gold standard for this purpose. In recent years deep sequencing technologies have been developed that hold the promise of measuring the transcriptome with RNA-seq in a more accurate and unbiased manner than microarrays. So far, however, few applications have been published that assess the performance of RNA-seq within a toxicogenomics context. Here, we applied RNA-seq for the characterization of the in vitro transcriptomic responses in HepG2 cells upon exposure to benzo[a]pyrene (BaP), a well-known DNA damaging carcinogen. We demonstrate the performance of RNA-seq with respect to the identification of differentially expressed genes and associated pathways, in comparison with microarray technology. RNA-seq data generates more complete and thus accurate data on differentially expressed genes and affected pathways than microarrays. Additionally, we highlight the potential of RNA-seq for characterizing mechanisms related to alternative splicing and thereby gathering new information. Exposure to BaP alters the isoform distribution for many genes, including regulators of cell death and DNA repair such as TP53, BCL2 and XPA, which are relevant for genotoxic responses. Finally, we demonstrate that RNA-seq enables to investigate allele-specific gene expression, although no changes for that could be observed. Our results provide evidence that RNA-seq is a powerful tool for toxicology which, compared to microarrays, is capable of adding valuable information at the transcriptome level for characterizing toxic effects caused by chemicals. Examination of 2 biological replicates at 2 different timepoints							
Term Source Name	ArrayExpress	EFO
Term Source File	http://www.ebi.ac.uk/arrayexpress/	http://www.ebi.ac.uk/efo/efo.owl						
Person Last Name	Gaj	Gaj	vanDelft	Kleinjans				
Person First Name	Stan	Stan	Joost	Jos				
Person Email	geo@ncbi.nlm.nih.gov							
Person Affiliation	Maastricht University							
Person Address	Department of Toxicogenomics, Maastricht University, Universiteitssingel 50, Maastricht, Limburg, Netherlands							
Person Roles	submitter							
Protocol Name	P-GSE36243-1	P-GSE36243-6	P-GSE36243-3	P-GSE36243-8	P-GSE36243-7	P-GSE36243-2	P-GSE36243-4	P-GSE36243-5
Protocol Description	ID_REF =  VALUE = Log2 of RMA intensity from MBNI custom annotation	12.5M-5g of cRNA were hybridized for 16 hr on 45M-0C on GeneChip Human Genome U133 Plus 2.0 Array, using AffymetrixM-. GeneChipM-. Fluidics Station 450 and the GeneChip Hybridization, Wash and Stain Kit	HepG2 cells were maintained as a monolayer culture in 95% humidity, atmosphere with 5% of CO2 and 37 M-:C. HepG2 cells were passaged at pre-confluent densities with the use of a trypsin-EDTA solution. Cells were cultured and passaged in a Minimal Essential Medium (MEM) with 10% of Foetal Bovine serum, 1% penicillin/streptomycin, 1% sodium-pyruvate and 1% non-essential amino acids.	The Affymetrix CEL files were imported in R 2.12.2 under BioConductor v2.7 (Gentleman et al., 2004). Probe sets were reannotated using the BrainArray custom CDF v13.0 annotations for both EnsEMBL genes and transcripts, generating two separate datasets. These custom CDFs were based on EnsEMBL v58. Next, both datasets were separately log transformed, RMA normalised and further evaluated through an extensive quality control pipeline consisting of box plots, fitPLM, NUSE, RLE, clustering/heat maps, PCA and correlation plots. There were no technically deviating arrays. The limma package (Smyth, 2005) was applied for computing a linear model using time and dose as main components. The resulting p-values were FWER-corrected using the False Discovery Rate (FDR) approach. Genes were considered to be differentially expressed (DE) when their (i) absolute fold change M-bM-^IM-% 1.5; (ii) q-value (FDR) < than 0.05; (iii) average expression in at least one of the two experimental groups > 6 (log2-scale).	Arrays were scanned using genechip scanner 3000 7G	When the HepG2 cells were 80% confluent, the medium was replaced with fresh medium containing either compound or solvent (0.5% DMSO, 0.5% EtOH or PBS)	Trizol extraction was performed using the miRNAeasy kit from qiagen accoding to the manufacturerM-bM-^@M-^Ys protocol	Biotinylated cRNA were prepared according to the standard Affymetrix protocol from 250ng total RNA using the 3M-bM-^@M-^Y IVT express kit
Protocol Type	bioassay_data_transformation	hybridization	grow	feature_extraction	image_aquisition	specified_biomaterial_action	nucleic_acid_extraction	labeling
Experimental Factor Name	TIME	COMPOUND						
Experimental Factor Type	time	compound						
Publication Title	RNA-Seq Provides New Insights in the Transcriptome Responses Induced by the Carcinogen Benzo[a]pyrene.							
Publication Author List	van Delft J, Gaj S, Lienhard M, Albrecht MW, Kirpiy A, Brauers K, Claessen S, Lizarraga D, Lehrach H, Herwig R, Kleinjans J							
PubMed ID	22889811							
Publication DOI	10.1093/toxsci/kfs250							
Comment[SecondaryAccession]	GSE36243							
Comment[GEOReleaseDate]	2012-11-16							
Comment[ArrayExpressSubmissionDate]	2012-03-02							
Comment[GEOLastUpdateDate]	2012-11-18							
Comment[AEExperimentType]	transcription profiling by array							
Comment[AdditionalFile:Data1]	GSE36243_Full_Affy_Table_RMA_Normalisation_LIMMA.txt							
SDRF File	E-GEOD-36243.sdrf.txt