plotScoreHeatmap color parameter

[alanaleah89]

Hi there,

I would like to use my own colour scheme for SingleRs heatmap. I think there may be a bug in the plotScoreHeatmap method.
When I execute the following code I get the same color scheme as the default.
plotScoreHeatmap(p.abx, color = (grDevices::colorRampPalette(c('#f1a340','#f7f7f7','#998ec3')))(100))

However, if I execute the same code with normalize = F, I get the color scheme I want but I would also like my data normalised.
plotScoreHeatmap(p.abx, color = (grDevices::colorRampPalette(c('#f1a340','#f7f7f7','#998ec3')))(100), normalize = F)

[LTLA]

@dtm2451 can you have a look at this?

[dtm2451]

Yes, yes. Sorry for the delay.. back from a long vacation today!

I know where this is from and can definitely address it this week if we want to @LTLA. My only concern is that it doesn't necessarily make sense for a user to use a bipolar color scale for the normalized data. I can't remember if we had kept the colors out of a users hands on purpose when normalize = TRUE. But it would feel like a bug to users anyway. I'll make a PR for this (and any other pheatmap inputs that may be overridden internally, if there are any).

[LTLA]

Right... yeah, I guess we should at least respond to the user-supplied color when normalize=TRUE, and it's on the user to make sure they choose a sensible color scale. In this case, a bidirectional one probably doesn't make a lot of sense, but I suppose someone might prefer something other than viridis (gasp), in which case we should support that.

[IreneP19]

Hi, has anyone fixed this?
I have the same problem. Changing color scale does not work when normalize = T.

Thanks!

[lindaboshans]

Following up on this. I also can not use custom colors when normalize is set to T.

[LTLA]

@dtm2451 did we decide what to do here?

[dtm2451]

Thanks for the bump @lindaboshans.
@LTLA looks to me above like we decided to give users this control, and leave it to them to pick a color scale that makes sense. ...I just forgot to do so. Starting this work now!

[LTLA]

Pushed to 2.4.1, should be available in a few days.