The Challenge
The client’s internal system was limited to looking at individual expression studies only. The client needed a way for their team of biologists to visualize gene expression data generated from in-house and published RNA-seq and microarray studies. More specifically, they were looking for ways to compare aggregated gene expression data across tissues, cell lines or cell types, and to be able to answer queries such as "what genes are most highly expressed in tissue X?", "in what cell line is gene Y most highly expressed?", and "what genes are most specific to cell type Z?".
Our Solution
Genestack built the "Expression Repository" - an application providing a visual, interactive exploration interface of gene expression with the following key features:
-
display of aggregated data from all available studies (grouped by tissue/cell type/cell line)
-
single-study view, with sample-level or group-level data
-
sorting of rows and columns by expression level
-
expansion of rows and columns into boxplots to show distribution of gene expression values
-
ontology browsing feature, allowing to zoom in and out of a tissue or cell type expression data using a hierarchy defined by an ontology
-
search across genes, pathways, ontology terms and study attributes to filter the heatmap
-
time-plot display for time course gene expression data
-
access to thousands of datasets from public repositories (ArrayExpress, GEO, etc.)
-
viewing data from public, shared and private sources simultaneously
Impact
Empowering biologists to explore and interrogate the gene expression data available autonomously, without the need to ask bioinformaticians for help, thus freeing up the time of bioinformaticians for more complex tasks and increasing research efficiency. Also the possibility of looking at data across multiple studies can provide a better understanding of gene function, thus facilitating the discovery of potential drug targets.