To host datasets, first generate an intial RData object containing the initial processing. This reduces the upfront computational burden to allow the focusing on visualization and exploration.
This Shiny App has been built in to a docker container for easy deployment. You can build the image yourself (and thereby customize any ports you need) after downloading it:
docker build -t RNASeqTemplate .
docker run -d --rm -p 8080:8080 RNASeqTemplate
And it should be hosted at localhost:8080
PROJECTID=$(gcloud config get-value project)
PROJECTNAME="ProjectName"
docker build . -t gcr.io/$PROJECTID/$PROJECTNAME
docker push gcr.io/$PROJECTID/$PROJECTNAME
See Deployment.