LAB-AID (Laboratory Automated Interrogation of Data) is an interactive web application for visualization of multi-level data from biological experiments.
A key step in understanding the results of biological experiments is visualization of the data, particularly when measurements exist within a hierarchy of interdependence. LAB-AID is a simple tool specifically designed to automatically visualize and query data resulting from such experiments. It aims to:
- lead to improved understanding of results
- help to avoid misleading interpretation of statistics
- easily identify outliers and sources of batch and confounding effects
For more details of LAB-AID's functionality, please see our paper.
LAB-AID is an R Shiny application. It can be run through R Studio locally, hosted via Shiny Server or run from a Docker image. It is also possible to deploy the application to shinyapps.io. In the latter three cases, you will access the LAB-AID application through a web browser.
If you have any problems installing LAB-AID via the methods described below, please let us know by raising an issue here.
The LAB-AID application can be run on your computer locally via RStudio. First, please follow the instructions here to download and install R and RStudio Desktop, if you do not already have these (and if you already have RStudio installed, we would recommend updating to the latest version). Then download all files from the LAB-AID repository (for example, by clicking the green 'Code' button at the top of this page, selecting 'Download ZIP', and uncompressing the downloaded ZIP file). Before running LAB-AID, you will need to make sure you have installed all the packages that LAB-AID requires (see below for the full list of required packages, and package installation instructions).
Finally, execute the following command in the RStudio Console pane, replacing PATH with the path to the folder the repository files have been saved in:
shiny::runApp(appDir = 'PATH')
Alternatively, you can open the files ui.R or server.R (or both) in RStudio, and press the 'Run App' button in the upper-right corner of the RStudio Source pane.
Note that it is possible that if you have an older version of R installed, you may encounter problems installing the packages that LAB-AID needs; if so, you can run the application through one of the alternative methods described below.
Docker is a platform for developing, shipping, and running applications which may have complex dependencies (such as the set of R packages on which LAB-AID relies). The computing environment needed to run a particular application is encapsulated in a Docker "container image". The LAB-AID docker container image is hosted on Docker Hub (n.b. a Docker Hub account is required to see this page, but is not required to download and run the LAB-AID image).
To run LAB-AID via the Docker container image, first please refer to the Docker documentation for instructions on how to setup Docker Desktop on your system. Then the container image can be downloaded and run by executing the following command line (for example by opening the 'Terminal' application on Mac OS, pasting the command, and pressing return):
docker run --rm -itd -p 80:80 zkozic/lab-aid:latest
Once this has finished, open your web browser and navigate to the address localhost:80 (for example by clicking this link).
If preferred, you can also use the shinyapps.io web hosting service to deploy the LAB-AID application. You will need to create a shinyapps.io account and to configure your local RStudio client. To do this, please follow the detailed instructions in the shinyapps.io documentation on how to deploy applications from RStudio Desktop to shinyapps.io. After deployment, LAB-AID can then be accessed through the web browser on the shinyapps.io website.
(n.b. advanced usage)
If you have access to, or manage, a Shiny Server installation, all that is required is to download the LAB-AID repository into the directory containing Shiny applications. Make sure that the R installation that the Shiny Server is using has all the required packages installed, and that the Shiny Server has sufficient privileges to read and write files within the application directory.
LAB-AID has been tested with R versions 4.1, 4.2 and 4.3. It requires the following R packages to be installed:
- shiny
- shinyjs
- shinyWidgets
- shinycssloaders
- tidyverse
- magrittr
- readxl
- jsonlite
- reshape2
- patchwork
- pheatmap
- plotly
- WriteXLS
- lme4
- car
- DT
- ComplexHeatmap (Bioconductor)
These packages can be installed by hand, or via the setup.R script provided in this repository (recommended). To use the latter, simply open the file in RStudio Desktop, and click the 'Source' button in the upper-right corner of the RStudio Source pane; all required packages will then be installed. If you are asked whether to Update all/some/none old packages, answer 'none' by typing n, then pressing return. If you are asked Do you want to install from sources the package which needs compilation? (Yes/no/cancel), answer 'no' by typing no, then pressing return.
Note that the setup.R script works best in a fresh installation of R. If you have a pre-existing R installation, in which you have already installed packages, it is possible that you might encounter problems installing the packages required by LAB-AID, as some of these may be incompatible with the versions of packages you already have. Unfortunately, these package version incompatibilites are beyond our control; in this case, we would recommend installing LAB-AID via the Docker image, or on shinyapps.io.
LAB-AID is configured through the config.json file. The file has the following structure:
- Title - application title.
- About - application and/or data set description.
- Data sets - entries for each data set (an entry is required for every data set used).
- Name - name of the data set.
- Path - path to the data set .xlsx or .csv file.
- nFactors - number of experimental factors.
- Description - descripton of the data set.
In general, we recommend adding and/or removing data sets via the Configuration tab within the running LAB-AID application. When adding new data sets, it is necessary to fill in the 'Dataset name' and 'Number of factors' fields; the optional 'Description' field provides text to appear on LAB-AID's About tab. Alternatively, the config.json file can be edited manually, but in this case, the correct JSON structure must be adhered to; please refer to the JSON documentation for further details on this format.
Note that when first run, LAB-AID is pre-installed with the example dataset described in our preprint.
LAB-AID input files can be Excel spreadsheets (.xls or .xlsx) or comma-separated value (.csv) files. Tables have to be in long format, with experimental factor columns first, followed by columns containing measured variables.