error in mean.var.plot

[roryk]

Hi everyone,

I have an error when I run mean.var.plot:

nbt = mean.var.plot(nbt, y.cutoff=2, x.low.cutoff=2, fxn.x=expMean,  fxn.y=logVarDivMean)

Error in seq.int(rx[1L], rx[2L], length.out = nb) : 'to' must be finite

I stuck a link up on Dropbox to a nbt object that is causing the problem.

https://dl.dropboxusercontent.com/u/2822886/nbt.RData

I'm rubbish at debugging problems in S4 objects-- any ideas about what is wrong?

[roryk]

Figured out my issue. Thanks!

[carmensandoval]

HI Roryk,

I'm having the same problem... what solved it for you in the end?

[roryk]

Sorry @carmensandoval, I forget exactly what it was but it was some problem with my data though not the function. I think I had not logged the data before making the Seurat object.

[azimmunivar]

I'm also having an issue with this. Did anyone get to the bottom of it?

[roryk]

@azimmunivar I forgot to log the data which caused the issue. Did you do that?

[marong0511]

@roryk I am also having an issue with this. Did you deal with UMI data? I did normalize data before running this.
I found when I scaledata with a linear model (default setting), everything goes well. But when I used 'negbinom' model to deal with UMI data, it goes wrong in Findvariablegenes with the same error you mentioned before.

[lihong1github]

I found the solution to this issue. If you normalize the data first using the "NormalizeData" function as shown below, it will fix this error.

seurat.object <- NormalizeData(
object = seurat.object,
normalization.method = "LogNormalize",
scale.factor = 10000)