Extending the Reactome Pathway Database for Multi-Omics Data

The increasing availability of high-throughput techniques, such as Transcriptome sequencing (RNA- seq), mass spectrometry (MS) based shotgun proteomics, and microarray studies allow the study of genome- and proteome-wide expression changes. All of these techniques share a common challenge: they result in lists of hundreds of genes or proteins that are differentially expressed between the analysed samples. Deriving biological meaningful conclusions from such data is highly challenging. Pathway analysis techniques have emerged as a solution to this problem. They merge existing knowledge on biological functions with the retrieved lists of expressed genes or proteins to reduce the amount of data. Instead of hundreds of genes or proteins, the researcher is presented with a list of pathways that are all linked to known biological functions. This allows for a much more intuitive interpretation of the data. Recent technological advances have decreased the cost of RNA-seq and proteomics analysis dramatically. This led to an increasing number of studies where samples are characterised using a multitude of omics technologies. Researchers are now faced with the new challenge to integrate this data from different omics approaches into a single biological interpretation. Reactome (http://reactome.org) is a free, open-source, open-data, curated and peer-reviewed knowledge-base of biomolecular pathways. Pathways are arranged in a hierarchical structure that allows the researcher to navigate from high level concepts like immune system to detailed pathway diagrams showing biomolecular events like membrane transport or phosphorylation. Its powerful web interface with its extensive visualization capabilities has made Reactome one of the most popular resources for pathway information. This research project focuses on the development of pathway analysis algorithms that extend Reactome to simultaneously analyse data from different omics technologies. Researchers will then be able to upload their data to the Reactome web application and get a graphical representation of the pathways up- or down-regulated in their analysed samples. The results from the different omics technologies will be visualized separately and differences between, for example, RNA-seq and proteomics analysis will be immediately visible to the user. Additionally, a new statistical approach will be developed to join the data from these different methods into a single pathway-centric view on the analysed samples. These developments are aimed to make Reactome a prime resource for the analysis of multi-omics data. Its simple use makes Reactome accessible to non-bioinformatics experts. Thereby, these new developments will enable, for example, medical doctors to interpret their samples data without additional bioinformatics support and foster translational research projects.