Expression Atlas at EMBL-EBI

One example of a value-added database is the Expression Atlas (the Atlas). It contains pre-computed (analysed) gene expression values from RNA-sequencing, microarray expression and proteomics experiments across different tissues, cell types, developmental stages and many other experimental conditions.

Queries can be either in a:

• baseline context, e.g. find genes expressed in the macaque brain
• differential context, e.g. find genes that are up- or down-regulated in response to auxin in Arabidopsis (Figure 7)

Explore the Expression Atlas home page by clicking on the :

Figure 7 The Expression Atlas home page. You can browse all differential and baseline experiments or search for data by gene, gene property, species or biological condition.

All data in Expression Atlas are free to browse, download, reuse and are selected from the ArrayExpress and functional genomics archives. All datasets are manually curated to a high standard by in-house curators, using terms from the Experimental Factor Ontology (EFO) where possible, and processed using standardised analysis methods.

Searching the Expression Atlas

You can search the Atlas by gene attributes (gene symbols, Ensembl gene ID or keywords), as well as by sample attributes and experimental factors, taking advantage of ontology-driven query expansion. For example, searching for the disease ‘leukemia’ will automatically return expression data from samples of ‘leukemia’ itself and from its subtypes and closely related diseases (‘chronic lymphocytic leukemia’, ‘promyelocytic leukemia,’ etc), since the relationship between “leukemia” and related diseases is defined in the EFO.

An example of how differentially expressed genes are displayed in Expression Atlas is shown below in Figure 8.

The gene expression values are displayed as a heatmap where rows represent genes and columns are the experimental comparisons. In the tabs at the top of the page you will see the results of the gene set enrichment analysis (see the Gene set enrichment section of the Functional genomics II course). Gene set enrichment analysis is performed using the Piano package from Bioconductor. For each comparison, the overlap between the set of differentially expressed genes and terms from GO, InterPro, and Reactome is performed, using Fisher’s exact test with multiple testing correction (FDR < 0.1).

Explore an Expression Atlas differential experiment page by clicking on the :

Figure 8 Example differential expression page in Expression Atlas.