Refactor panels organization to follow Shiny modules approach

[jaeddy]

See slides here for more details: https://github.com/daattali/shiny-training-rstudioconf-2018/blob/master/slides/05-modules.pdf

Basically, instead of

iatlas-demo/
├── global.R
├── panels
│   ├── panel-server-cellcontent.R
│   ├── panel-server-clonaldiversity.R
│   ├── panel-server-corrheatmap.R
│   ├── panel-server-survivalcurve.R
│   ├── panel-ui-cellcontent.R
│   ├── panel-ui-clonaldiversity.R
│   ├── panel-ui-corrheatmap.R
│   └── panel-ui-survivalcurve.R
├── server.R
└── ui.R

...we'd want something more like this:

iatlas-demo/
├── global.R
├── modules
│   ├── cellcontentmodule.R
│   ├── clonaldiversitymodule.R
│   ├── corrheatmapmodule.R
│   └── survivalcurvemodule.R
├── server.R
└── ui.R

Each module script would include a UI and server function — e.g., cellcontent_module_UI() and cellcontent_module() in cellcontentmodule.R. Module scripts would then be sourced in global.R and functions used in ui.R and server.R

[Gibbsdavidl]

Thanks for the link to the training slides. I'm reviewing them now. Very helpful. Template for modules: https://gist.github.com/jcheng5/c09a2449c0a94c8f498c4a68a416bc3b

[Gibbsdavidl]

Would we make the modules to follow the figure taxonomy? We can reuse the modules and have multiple links (with different analysis aims and titles) to each one.

iatlas-demo/ ├── global.R ├── modules │ ├── Scatter plot module.R │ ├── Boxplot family module.R │ ├── Bargraphs module.R │ └── Kaplan-Meier module.R | |-------etc.R ├── server.R └── ui.R

[jaeddy]

Thanks, @Gibbsdavidl — the module template also looks handy.

I was thinking that we'd try to follow the figure taxonomy, but with the modules themselves more organized by "bioapplication" as @vthorsson describes it. Something along these lines:

iatlas-demo/
├── global.R
├── modules
│   ├── samplegroupsmodule.R
│   ├── featurecorrelationmodule.R
│   ├── geneexplorermodule.R
│   └── survivalmodule.R
├── widgets
│   ├── scatterplot.R
│   ├── boxplot.R
│   ├── barplot.R
│   ├── mosaicplot.R
│   └── kmplot.R
├── server.R
└── ui.R

Such that functions for creating different widgets can be used by multiple modules. Not sure if this is the best approach, but trying to minimize duplicated code.

[jaeddy]

Modifying the structure a bit, based on @vthorsson's question about loading data...

Instead of "widgets", we'd have a place where we store re-usable "functions" (I think I've also seen these named as "helpers" — either would be fine with me).

iatlas-demo/
├── global.R
├── modules
│   ├── samplegroupsmodule.R
│   ├── featurecorrelationmodule.R
│   ├── geneexplorermodule.R
│   └── survivalmodule.R
├── functions
│   ├── scatterplot.R
│   ├── boxplot.R
│   ├── barplot.R
│   ├── mosaicplot.R
│   ├── kmplot.R
│   ├── load_data.R
│   └── utils.R
├── server.R
└── ui.R

[jaeddy]

@andrewelamb: below is a more detailed plan for mapping our existing content into the proposed reorganization. For this pass, the focus should be more on setting up the overall file/folder structure, not on consolidating code (which we'll try to tackle with #2). In other words, it's fine for now if code is duplicated in multiple files, as long as we can get the overall app to work.

This can probably be a single branch/PR, but might require multiple commits for any debugging that comes up...

iatlas-demo/
├── global.R
├── modules
│   ├── cellcontentmodule.R
│   ├── immuneinterfacemodule.R
│   ├── featurecorrelationmodule.R
│   └── survivalmodule.R
├── functions
│   ├── heatmap.R
│   ├── boxplot.R
│   ├── barplot.R
│   ├── kmplot.R
│   ├── load_data.R
│   └── utils.R
├── server.R
└── ui.R

modules

• cellcontentmodule.R:
  • panel-ui-cellcontent.R -> cellcontent_UI(...)
  • panel-server-cellcontent.R -> cellcontent(...)
• immuneinterfacemodule.R:
  • panel-ui-clonaldiversity.R -> immuneinterface_UI(...)
  • panel-server-clonaldiversity.R -> immuneinterface(...)
• featurecorrelationmodule.R:
  • panel-ui-corrheatmap.R -> featurecorrelation_UI(...)
  • panel-server-corrheatmap.R -> featurecorrelation(...)
• survivalmodule.R:
  • panel-ui-survivalcurve.R -> survival_UI(...)
  • panel-server-survivalcurve.R -> survival_UI(...)

functions

• load_data.R (copy/paste for now):
  • global.R::load_<x>_data(...) for each 'x'
• utils.R (copy/paste for now):
  • global.R::getVars(...)
  • global.R::getNiceName(...)
  • global.R::buildDataFrame_corr(...)
  • global.R::buildDataFrame_surv(...)

[andrewelamb]

The functions are still in globals for now, but can be removed at any point.

[andrewelamb]

This is finished