The GENESIS Model Explorer App is a Shiny application developed by the Genetic Analysis Center at the University of Washington in collaboration with Seven Bridges Genomics. It enables users to visualize and explore the results of the GENESIS Null Model workflow.
It is meant to provide an intuitive interface for researchers to easily select, visualize, and explore phenotype variables, genotypes, and GENESIS model results interactively with no prior R programming knowledge.
Researchers can use the app to explore their own data or use example data provided with the app.
If comparisons with genotypes are desired, users can use the GDS Genotype Extractor tool, which enables researchers to export genotype dosages for a small number of variants from a GDS file into the format required by this app.
You can install the development version from GitHub with:
# install.packages("devtools")
devtools::install_github("UW-GAC/GENESISModelExplorer")To launch the app, run the following:
GENESISModelExplorer::run_app()This app contains three tabs, which will be described in their own sections below.
- Overview
- Load data
- Plot
This page gives a brief overview of the app’s purpose and what files are required to run the app.
This tab allows you to select which data you would like to plot. You can select either the example data provided by the app or provide your own data.
To use the example data, click on the Use example data? checkbox. The other file selections will disappear.
To use your own data, you will need to select both a null model file
and a phenotype file. They are required to be in RData format. Both
of these files are produced by the GENESIS Null Model app. For the
null model file input, we suggest using the file(s) listed in the
GENESIS Null Model null_model_report_files output field because they
are smaller. You can optionally modify the phenotype file to add
additional columns to use when plotting, as long as the modified version
is stored in an Annotated Data Frame and saved as an RData file.
You may optionally provide a genotype file to generate plots by the genotype of selected variants. This file should be generated by the GDS Genotype Extractor app.
Once you have selected your data, click the “Load Data” button at the bottom of the page. This will load your data into the app and combine it into a single data frame to use in plotting. The app will display a progress bar while the data are being loaded. Once the data have been loaded, the app will automatically load the next tab, Plot.
This tab is where you will generate plots of your data in the GENESIS Model Explorer app. You can generate:
- Histogram, density, or bar plots of a single variable.
- Hexbin, scatter, box, or violin plots of two variables.
- Plots grouped (colored) by a categorical variable.
- Plots faceted by a categorical variable.
To interact with this tab, you will select plot options in the panel on the left side of the page. Once you are happy with your selections, you will click on the Generate plot button at the top of the panel. The plot will then be displayed on the right side of the page.
When selecting which variable to include in plots, variable names are prefixed by the file where they originated in order to differentiate between variables with the same name in different files. For example, “Model: resid.marginal” indicates that the app will plot the “resid.marginal” variable from the null model file, and “Phenotype: bmi” indicates that the app will plot the variable named “bmi” in the phenotype file.
To generate plots of a single variable, select a variable for the x-axis from the dropdown menu and set the y-axis dropdown set to “—”.
To generate plots of two variables, select a variable in both the x-axis and y-axis dropdown menus.
After selecting variables, a default plot type is chosen automatically, but you may change the type using the plot type dropdown menu. The default and allowed types are chosen based on the data types of the variable(s) that you have selected. For example, the “histogram” plot type is only available when you have selected an x-axis variable and no y-axis variable, while the “hexbin” plot type is only available when you have selected quantitative variables for both the x-axis and y-axis. Finally, you are not able to select categorical variables for both the x-axis and the y-axis.
You may further customize your plot with the following:
- The group by dropdown menu will select a variable to group your plot by. This displays different categories in different colors. Note that only categorical variables appear in this dropdown. If you have selected the “hexbin” plot type, the plot will not be grouped even if you have selected a variable to group by.
- The facet by dropdown will select a variable to facet your plot by. This displays different categories as different panels. Note that only categorical variables appear in this dropdown.
Additional options are also available depending on the plot type you have selected.
Once you have made your variable selections, click on the Generate plot button at the top of the panel. The plot will appear to the right of the panel. This shows an example histogram of the model outcome.
You may then make new variable selections in the panel on the left. To update the plot with the new variable selections, click on the Generate plot button again. This shows an example of plotting the model outcome as a histogram, grouped by the “sex” variable in the phenotype file.
When you are done using the GENESIS Model Explorer App, just close the app window.
Please note that the GENESISModelExplorer project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.