I am a scientist researching infectious diseases and how they affect various genes. I need to save my data and visualizations in a searchable centralized place so that I can compare my findings to other researchers performing similar analyses. Let's see how SEMOSS helps me find these researchers, assemble their data, and use their findings to augment research of my own.
One of my experiments revolves around exposing human immune system cells to the Middle East Respiratory Syndrome (also known as MERS). I take two measurements – pre and post exposure – in order to explore how the exposure changes gene expression levels.
I use the interactive and customizable SEMOSS heat map shown on the right to then visualize these changes. Gene symbols are along the X axis and the two experiment measurements are along the Y axis. Each square contains a gene expression measurement (darker meaning more expressed).
Use the filters in the top corner to find genes of interest by seeing which have large variations.
I see that a lot of genes have undergone large changes, some have more than doubled their gene expression with exposure to MERS. Let’s now focus only on those that have changed the most and see what else we can learn.
I’ve now mapped the amount of change that each of the genes underwent from the previous visualization (labeled here as Fold Change) and added an additional coordinate of Binding Strength. The scatter plot visualization allows me to view trends in the data while also being able to view the properties of individual points in the graph.
In the scatter plot we see how Fold Change relates to Binding strength. We are specifically interested in the genes that have had the largest measurable change (Fold change), when we hover over the right-most points in the scatter plot we see that SERPINE1, PLAT, MEFV, and MC4R are the gene symbols with the largest fold change in expression levels. These genes also seem to be similar in that they all have large Binding Strength.
I have now reached the end of my own data, this is where I can take advantage of the open-data stored within SEMOSS to augment my own research. Let’s see what open-data exists around these genes.
Here we see a simple network graph with our genes of interest. By extending this graph we see what relationships these genes have in common as it might explain their large change in expression.
Select ‘geneSymbol’ in the top left key and then select a data type from the traversal dropdown in the upper right. The graph will populate with related data types from each of the different gene symbols (hint: try CellularComponent or MolecularFunction).
The graph shows us that some properties are specific to one gene or the other, and other properties are shared between them. These shared characteristics provide a jumping point for further research and insight.
Finally we want to see who are the other researchers working on these genes so that we can reach out discuss collaboration opportunities. Let’s see how we can use the citation networks contained within SEMOSS.
In this parallel coordinates visualization, we can see the different diseases that have been linked to the genes that we have been exploring as well as that papers have been published around those diseases and the authors of those papers.
The two genes that we have been tracking all along, SERPINE1 and PLAT, have a shared publication! This is a great place to start. Let’s select that publication by dragging along the publication axis to highlight the authors of that paper.