Pathways

How does gene expression in a particular study relate to disease mechanisms and gene regulatory networks?  Pathway search and visualization features in Single Cell Portal give insight into these biological questions.

Background

Pathway diagrams are key to understanding the molecular processes that cause biological conditions.  These diagrams are directed graphs of nodes – e.g. genes, metabolites, pathways, and phenotypes – and the interactions among them.  Interactions include stimulation, inhibition, binding, catalysis, and more.  

Nodes can interact with other nodes individually -- e.g. PCSK9 inhibits LDLR -- or as sets.  Gene sets in these pathway diagrams generally represent gene families, like RAS GTPases, or complexes, like mTORC1.  These nodes and interactions are also often shown in cellular context, with interactions shown depicted in particular cellular compartments like the nucleus, cytosol, or extracellular matrix.  Cells are often labeled by cell type – hepatocyte, glial cell, etc. – and supplemented with labels for their organ or tissue context.  Pathways diagrams are sourced from WikiPathways, and ultimately from scientific literature.

Single Cell Portal's new pathway diagrams show expression for each gene in a pathway, for the current study.  This enables you to see if expression is enriched, and how, with high resolution into gene modules and nodal biology.
 

Search

The search box within a study now suggests pathways, in addition to genes as before.  Pathway are suggested if the text query matches A) part of the pathway title, or B) a gene in the pathway.  For example, typing “lac” in the search box returns genes that have names matching that query – e.g. “LACC1” – and pathways that have titles including that string -- e.g. "“Mammary gland development: pregnancy and lactation - stage 3 of 4”.

Visualization

Click on a pathway suggestion to show the diagram and expression overlay.  

Gene nodes in the biological pathway diagram are colored by gene expression.  As conveyed in the legend, red means high average gene expression, and blue is low.  Bold color means a high percent of cells express the gene, and faint color means a low percent of cells express it.  These metrics are also used in dot plots.  Hovering over a gene shows the expression metrics.

You can also click on a pathway node -- green nodes like "Involution" in the example above -- to see expression in a related pathway.

The panel at right has a description that explains the pathway diagram.  The description sometimes contains links with more detail, or to references.  The legend above the diagram shows how the "scaled mean expression" and "percent expressing" metrics are visually conveys.

You can also get a convenient list of genes in the pathway by clicking the "Genes" button at bottom right.  Similarly, if the pathway diagram has "Cell type" or "Disease" annotations from WikiPathways, then you will see buttons you can click to see those.

 

Try it, give feedback!

You can try the above example in https://singlecell.broadinstitute.org/single_cell/study/SCP1671.  Pathway search and visualization works in any human study.

Tell us what you think!  Get in touch with us at [e-mail], on Bluesky, or X.