The Ontario Genomics Institute (OGI) and the Ontario Institute for Cancer Research (OICR) co-hosted a one hour webinar about the Reactome Pathway Database (http://www.reactome.org) - a freely available, manually-curated resource of core biological pathways. The Reactome database offers pathway data encapsulating areas of human biology ranging from basic pathways of metabolism to complex events such as GPCR signaling and apoptosis.  This follow up webinar introduced the updated website and included case studies from various research groups.
 
The presentation was given by Dr. Robin Haw, Manager of Reactome Outreach, OICR, and covered how to use the updated Reactome resource for:

• Browsing and searching pathway knowledge,

• Integrating network and pathway data,

• Using Pathway and Expression Analysis tools to analyze experimental datasets,

• Annotating experimental datasets with Reactome BioMart,

• Discovering network patterns related to cancer and other diseases using the Reactome Functional Interaction Network Cytoscape plugin,

• Introducing use cases for Reactome data and analysis tools.

Click here to download the slide presentation

Click here to access a recording of the webinar

Click here to take part in a Reactome User Survey

Q&A

Can you represent protein expression data from multiple sites of modification to a particular protein?

You can query Reactome with protein expression data using the Expression Analysis tools. Unfortunately, the Expression tool does not distinguish between multiple sites of modification to a particular protein.

Can you request a Reactome map to a specified list of targets/proteins, limiting the map to the specified list?

The pathway analysis tools will allow users to identify the pathways associated with a list of targets/proteins. However, at this time it is not possible to limit a map to the specified lists.

Can you tease apart the graphical image of the Skypainter results to provide images for each category for better resolution/detail?

No, however we are working on new tools to assist with pathway visualization and offer more detail.

If a figure shows a protein in a particular pathway located in a particular cellular compartment and performing a particular reaction, how can a user trace those assertions back to the original paper(s)?

The details panel of the Pathway browser (below the pathway diagram) will provide information about the pathway, reaction and molecule. Selecting the reaction in the pathway diagram will show the details for the reaction. As you scroll down through the details, you will see the References that contains a link to PubMed that support the reactions.

Are metabolites (compounds) supported by either the over-representation analysis or expression analysis facilities, or are these analyses limited to genes/proteins?

Yes, the pathway and expression analysis tools support the use of KEGG Compound and ChEBI IDs.

Can you provide some guides to go further in the analysis using the Cytoscape plug-in?

If you would like to learn more about the Reactome Functional Interaction (FI) Cytoscape plugin, please go to the FI Cytoscape Plugin link under 'Tools' of the Navigation bar on the Reactome homepage. You can also refer to publication "A human functional protein interaction network and its application to cancer data analysis" that describes the construction of the FI network.

Given a set of genes derived from a GWAS study of a clinical phenotype, how do you interpret Reactome results? I am thinking about the possibility of obtaining confusing results due to genetic heterogeneity (involvement of multiple, apparently unrelated pathways)?

You could simply submit a list of gene names or Entrez Gene IDs to the pathway analysis to map the IDs to the pathways. Alternatively, if you have a numerical value associated with each gene (e.g. statistic) you could superimpose the results onto the pathway diagrams using the expression analysis tools.

Is there a way to influence the p score value by the strength/confidence of a particular quantitative measurements (or set of measurements)?

Not in the context of the Reactome tools. You would have to do any statistical analysis first and then submit the list of genes/proteins to the pathway or expression analysis.

Are there canonical pathways built in?

The pathways in Reactome are canonical.

How closely do the pathways reflect GO categories?

All molecules, reactions and pathways have a GO cellular component annotation. Wherever the GO annotations are available the reaction and pathway will have a GO molecular function and biological process annotation, respectively.

What are your thoughts on GeneGO and other commercially available resources?

Both academic and commercial pathway databases do their best to cover the vast space of biological knowledge. The tools they provide are complimentary to one another. However, the open source pathway resources like Reactome provide access not only to the data but also the source code of the website and tools. Commercial resources don't allow you that same level of access.

Is there a way of contrasting data between databases (discrepancies between database data) to get a general feel of the ''consensus'' answer?

You could take a look at the website Pathguide. This website provides a way of contrasting the databases and there data.

InnateDB says that it takes the data of Reactome. Any advantage to using Reactome directly instead of it?

If you are an immunologist, it might be worthwhile looking at Reactome data in the context of immunological data stored in InnateDB. However, with the new pathway visualization system and analysis tools it would be better to refer to the Reactome website. Furthermore, you will always find more up-to-date pathway information at Reactome. Some external databases that use our pathway data are sometimes not using the most recent Reactome release.