PT  - JOURNAL ARTICLE
AU  - Steven Xijin Ge
TI  - iDEP: An integrated web application for differential expression and pathway analysis
AID  - 10.1101/148411
DP  - 2017 Jan 01
TA  - bioRxiv
PG  - 148411
4099  - http://biorxiv.org/content/early/2017/06/09/148411.short
4100  - http://biorxiv.org/content/early/2017/06/09/148411.full
AB  - iDEP (integrated Differential Expression and Pathway analysis) is a web application that reads in gene expression data from DNA microarray or RNA-Seq and performs exploratory data analysis (EDA), differential expression, and pathway analysis. The key idea of iDEP is to make many powerful R/Bioconductor packages easily accessible by wrapping them under a graphical interface, alongside annotation databases. For EDA, it performs hierarchical clustering, k-means clustering, and principal component analysis (PCA). iDEP detects differentially expressed genes using the limma and DESeq2 packages. For a group of co-expressed genes, it identifies enriched gene ontology (GO) terms as well as transcription factor binding motifs in promoters. Pathway analysis can be performed using several packages like GAGE, GSEA, PGSEA, or ReactomePA. iDEP can also detect chromosomal gain or loss using the PREDA package. iDEP uses annotation of 69 metazoa and 44 plant genomes in Ensembl for ID mapping and functional categorization. Pathway information was also compiled from databases like KEGG, Reactome, MSigDB, GSKB, and araPath. As an example, we extensively analyzed an RNA-Seq dataset involving siRNA-mediated Hoxa1 knockdown in lung fibroblasts, and identified the down-regulation of cell-cycle genes, in agreement with previous findings. Our analyses also reveal the possible roles of E2F1 and its target genes, including microRNAs, in blocking G1/S transition, and the upregulation of genes related to cytokines, lysosome, neuronal parts. By integrating many R and Bioconductor packages with comprehensive annotation databases, iDEP (http://ge-lab.org/idep/) enables users to conduct in-depth bioinformatics analysis of transcriptomic data through a graphical interface.