DE2 Tutorial

Data Explorer 2 is an entirely new version of our Data Explorer application with additional plot types and functionality.

We've assembled a few brief screen recordings some ofthe key functionality in DE2, but this is not an exhaustive list.

The best place to start when learning about Data Explorer 2 (DE2) is to take a look at the examples are shown when you first open DE2. These give examples of some of the types questions which DE2 was designed to answer and examples of the plots you can generate to answer those questions.

Click on any example to see it live and experiment with the settings.

example 1 with edit.mov

In any plot use the lasso tool to select points. These selections can be saved as "contexts" and used in various other places in the portal. For example, one can use contexts to perform a two class comparison

In the first version of Data Explorer, points always represented a single model. However, in the new Data Explorer, points can represent models or "features" (such as genes). In the next example, each point represents a gene and along the values along the X and Y axis are the means of the gene effect in EMT-high models and EMT-low models. In this plot, one can identify those genes which have a different average gene effect score between those two types of models by looking at the off-diagonal points. Hovering over points show the gene symbol and the respective mean scores.

example 2.mov

To augment to Data Explorer 2, a new tool has been introduced for defining contexts: Context Manager. In this tool, one defines rules which are applied to identify which lines are members of the context.

Note: These are stored as rules and not static lists, so as DepMap profiles more models, the membership of the context will be based on the data at the time you use the context.

See below to see an example of opening Context Manager and defining a context named HER2Amp based on ERRB2 copy number being greater than 1.5.

context manager her2amp.mov

Alternatively, we could define the same context by generating a plot of the distribution for ERBB2 copy number and select those points in the plot. In all plots, you can select points and the contents of the selection is shown on the right "Plot selection" box. From there, you can save the context for later use.

Note, at the end, you can see the resulting context is a rule which lists a set of list of model IDs.

using selection to define context.mov

Data Explorer 2 has much more options for how to use color in plotting. You can group and color by groupings we've defined (such as primary disease) but you can also color models by contexts you've defined.

In the example below, I'm generating a plot to compare ERBB2 expression in lines with ERBB2 amplified vs those which are not, and then defining a context of models whose lineage is "Breast" to only see that comparison only within in that lineage.

color by her2amp.mov

Also, one now has complete control over the exact colors used by selecting the "gear" icon at the top

plot color config.mov