platexpress_shiny

Transforming platexpress data for visualization in shiny dashboards.

validate and need server functionality

Shiny best practice for error handling involves utilization of the validate(...,errorClass= character(0)) and need() for output rendering functions. With the keyword argument errorClass, special CSS classes can be applied to the error message. Objects of the "try-error" class can also serve as input for validate.

starting rstudio for developing

Clone the git-repo into an desired directory. Edit the PASSWORD environment variable. Start docker desktop on windows or the docker client on ubuntu. Open a shell (e.g. the windows powershell) and change your current directory to the downloaded git-repo directory with the command cd path-to-git-repo. Start the docker container by using the docker-compose up command. Visit the URL localhost:8787 in an browser of your choice, google chrome or firefox are recommended. Use rstudio as user and the variable PASSWORD as your password. Load the required_packages.R script into your working directory (WD) and submit the code. This will install all necessary packages.

some notes on docker

Sometimes you need to remove docker containers or images. If you do this, keep in mind, that everything that has been installed inside the container is also removed by removing the container. Thus, all files that are inside the container and not on a shared directory will be removed to. The shared directory of this docker container is the whole directory in which you place the docker-compose.yml (default: the git-repo), which is set with the dot (".") in the volume section of the docker-compose.yml.

If you need to re-install the docker container, first remove it. Use following commands to identify the container-id: docker ps (if the container has started) or docker ps --all. Identify the container-id of the rocker/rstudio container and use the command: docker rm <id> to remove the container. If you realy want to clean up everything you can also delete the image. This can be done with the docker rmi <image-id> cmd, identify the image-id with docker images.

INFO

• use the required_packages.R script for installing additional packages
• change the PASSWORD environment variable in the docker-compose.yml file before you submit the docker-compose up cmd
• Folder description:
• data:
  • biolector and synergy test datasets
• src:
  • platexpress_module: functions for interactions with platexpress
  • shiny: shiny R file with ui and server specifications

TODO

• Layout file creation with shinyMatrix
• add validate and need shiny functions in server
• add boxData functionality and refactor functions
• documentation
• change and refactor code in the platexpress_interactions.R tryCatch blocks:
  • edit the warning block
  • add error handling
  • add final statements
• change ui of groups tab
• develop a plan where to go with this application
• check Multiple conditions section of: error handling

implementation of correctBlanks and skipWells plateXpress functions

implemented this functionality:

raw <- skipWells(raw, skip = "A9")
plate <- skipWells(plate, skip = "A9")
data <- correctBlanks(data=raw, plate=plate)
viewPlate(data, rows = c("A","B","C"), cols = 1:9)