Option to deprotect `motif` column in universalmotif_df

[snystrom]

I was just messing around with a set of motifs and I wanted to selectively RC one while using a universalmotif_df.

df <- to_df(c(create_motif, create_motif(name = "motif2")))

df %>%
   mutate(motif = ifelse(name == "motif2", motif_rc(motif), motif))

Of course, this doesn't work because of the AsIs class.

So this got me thinking if there could be a way for folks to deprotect the motif for cases like this. Users doing this would know the operation is unsafe. Here's the silly implementation I came up with:

edit_motif <- function(m){
  class(m) <- NULL
  m 
}

So the above becomes:

df <- to_df(c(create_motif, create_motif(name = "motif2")))

df %>%
   mutate(motif = ifelse(name == "motif2", motif_rc(edit_motif(motif)), edit_motif(motif)))

Not sure if this is the perfect solution. For instance, this won't keep the AsIs attribute after the mutate, so maybe a macro for wrapping the whole ifelse operation to deprotect could work? I'll do some more thinking on this, but figured I'd post it now.

[snystrom]

I guess actually, just using I() works:

df <- to_df(c(create_motif, create_motif(name = "motif2")))

df %>%
   mutate(motif = I(ifelse(name == "motif2", motif_rc(edit_motif(motif)), edit_motif(motif))))

Or, of course, calling update_motif() after an operation like this.

[bjmt]

I can see how being able to do some things to motifs while they are still in the universalmotif_df would be useful. But your suggestion in this case would leave the motif out of sync with the consensus column. I think there has to be a way to 'clean up' any changes which need propagating to the rest of the universalmotif_df after applying the desired functions. Maybe there could be some kind of apply-like function provided that would work well with pipes? But how many uses would this sort of functionality have beyond motif_rc()? Maybe we could think of a workaround more specifically for motif_rc() instead?

[snystrom]

Yeah, that's a good point. Perhaps the solution is just to do update_motifs(). I agree with the sentiment that it gets tedious to have to handle all these special cases, when in reality, the tools to solve the problem already exist. update_motifs() and the print out of sync warning do a good job there.

I'll do some more thinking about this. I don't think something like this is a very high priority feature.