Multipanel plot return type

[dan-knight]

It seems that create.multipanelplot does not return a plot object if a filename is specified (it just writes the plot to the > file). Is that expected behavior? Also, when an object is returned, it seems to be of class "multipanel" - is that something >that was made up for BPG or is that an actual lattice-related class?

Originally posted by @alkaZeltser in uclahs-cds/package-ApplyPolygenicScore#24 (comment)

Off the top of my head, I don't think this should be the case. It would be inconsistent with the other plotting functions. It's possible that there's some technical limitation specific to multipanel plots, but I'd hope that wouldn't be the case. You can see here that it's returning the grob rather than returning the result of write.plot. This is inconsistent with other plotting functions..

The class name is set here. I think that should be changed to include the other trellis/grid classnames (adding multipanel) rather than setting it to only multipanel.

[jarbet]

I think that should be changed to include the other trellis/grid classnames (adding multipanel) rather than setting it to only multipanel.

Agreed! There is a bug where you can't immediately input a plot made by create.multipanelplot into another create.multipanelplot call.

The temporary solution @stefaneng came up with is to just redefine the class:

class(multipanel.plot) <- c('multipanel', 'frame', 'gTree', 'grob', 'gDesc');

Then the above multipanel.plot can be input into another create.multipanelplot call.

[pboutros]

I have no memory of this either way, but suggest tracing the initial commit history of create.multipanelplot on svn to see if there's anything useful in there