 Hello, thank you for viewing this useR 2020 lightning talk recording. My name is Bo Wen. Today I'm bringing to you a brief introduction to an exciting project, MetaShiny, built Shiny apps with a Shiny app from the scientific computing and consulting group at Novartis. MetaShiny is an R package developed to help users with varied R coding experiences to quickly create and deploy Shiny apps using a point-and-click interface. In our effort to integrate Shiny apps into the routine analysis flow at Novartis, we received high-level interests from users across the analytics spectrum. In the meantime, we found it challenging to fulfill the unique demands of clinical trial teams from a wide variety of disease units working with intricate differences in their trial design. Therefore, instead of working on individual Shiny projects, the scientific computing and consulting group has put most focuses on developing Shiny templates and modules that our collaborators can plug and play in their own Shiny projects. However, the R environment and Shiny reactive programming proved to have steep learning curves for non-analytical colleagues such as medical doctors and analytical colleagues with expertise in other software such as SAS and Python. Moreover, for experienced R programmers, writing Shiny code may also be time consuming. MetaShiny strives to close this gap and enable Shiny development without having to write a single line of R code. MetaShiny relies on Shiny modules, which are reusable units of Shiny logic with unique namespaces, thus can be plugged in bigger Shiny projects. What's key in MetaShiny is that the Shiny modules are parameterized. As shown in the diagram below, user requirements for the Shiny app is collected as inputs in a MetaShiny app, and these inputs are passed on to the Shiny module as parameters. For example, the user can specify the plot type argument to be scatter or line, which results in either a scatter plot Shiny app or a line plot Shiny app. This method was inspired by the Eskies package. Now let's see a demo of the MetaShiny app. In step one, user provides some metadata for the app they want to create, such as app name, author, and app layout. In the red box on the right hand side, a live preview of the app is shown to the user as they provide inputs to customize the app. In step two, user adds parameters to the Shiny module. Here, I created a box plot module using the iris data set. I then used the add grouping variable selector in the MetaShiny app on the left hand side to add a group variable selector to the app preview. The preview app is fully interactive, so user can test their app right away. In the final step, user saves code of the app they just created in the zip file or deploy the app to Shiny Server Pro. This project is under active development. Some of the next steps include package dependency, point and click interface upgrading to drag and drop interface, and platform agnostic deployments of the final app. Please reach out to me via email or on GitHub if you have any questions or are interested in collaboration. Thank you for viewing.