ylim error - Unable to plot and save data

[ruzy1]

Hello! I'm getting an error when I am saving the SAM data, and it crashes when I try to save. It is also unable to plot the graph in the SampleSize tab, so I think the crashing and the plot are related. Please help... :')

This is the error:
Warning: Error in plot.window: need finite 'ylim' values

This is my code:

library(shiny)
library(impute)
runGitHub("SAM", "MikeJSeo")
Downloading https://github.com/MikeJSeo/SAM/archive/HEAD.tar.gz

Listening on http://127.0.0.1:5696
Estimating sequencing depths...
Resampling to get new data matrices...
perm= 1
perm= 2
perm= 3
perm= 4
perm= 5
perm= 6
perm= 7
perm= 8
perm= 9
perm= 10
perm= 11
perm= 12
perm= 13
perm= 14
perm= 15
perm= 16
perm= 17
perm= 18
perm= 19
perm= 20
perm= 21
perm= 22
perm= 23
perm= 24
perm= 25
perm= 26
perm= 27
perm= 28
perm= 29
perm= 30
perm= 31
perm= 32
perm= 33
perm= 34
perm= 35
perm= 36
perm= 37
perm= 38
perm= 39
perm= 40
perm= 41
perm= 42
perm= 43
perm= 44
perm= 45
perm= 46
perm= 47
perm= 48
perm= 49
perm= 50
perm= 51
perm= 52
perm= 53
perm= 54
perm= 55
perm= 56
perm= 57
perm= 58
perm= 59
perm= 60
perm= 61
perm= 62
perm= 63
perm= 64
perm= 65
perm= 66
perm= 67
perm= 68
perm= 69
perm= 70
perm= 71
perm= 72
perm= 73
perm= 74
perm= 75
perm= 76
perm= 77
perm= 78
perm= 79
perm= 80
perm= 81
perm= 82
perm= 83
perm= 84
perm= 85
perm= 86
perm= 87
perm= 88
perm= 89
perm= 90
perm= 91
perm= 92
perm= 93
perm= 94
perm= 95
perm= 96
perm= 97
perm= 98
perm= 99
perm= 100
Number of thresholds chosen (all possible thresholds) = 142
Getting all the cutoffs for the thresholds...
Getting number of false positives in the permutation...
Number of thresholds chosen (all possible thresholds) = 144
Getting all the cutoffs for the thresholds...
Getting number of false positives in the permutation...
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Number of thresholds chosen (all possible thresholds) = 1
Getting all the cutoffs for the thresholds...
Getting number of false positives in the permutation...
Warning: Error in plot.window: need finite 'ylim' values
64: plot.window
63: localWindow
62: plot.default
60: samr.assess.samplesize.plot
47: observe [C:\Users\rache\AppData\Local\Temp\RtmpuWIFEa\shinyappfa465b3b15\SAM-d998a4ed8cba4b7f071ec7f3b5b9606f4c192eed/server.R#916]
46:
3: runApp
2: runUrl
1: runGitHub

At line #916 in the server.R file, the code is as follows:

isolate({

dir = input$dir
file = input$fname

siggenes.table = getSiggenesTable()
missrate.table = getMissRateTable()
delta.table = getDeltaTable()
samr.assess.samplesize.obj =  getSampleSize()
Allgenes = getAllgenesTable()

samr.obj = getSamrObj()
delta = findDelta()
min.foldchange = input$min.foldchange

inputParameters = getInputParameters()
inputParameters = inputParameters[,-1, drop = FALSE]
  
computedValues = getComputedValues()
computedValues = computedValues[,-1, drop = FALSE]

eigengene = getEigengene()
imputedX = getImputedX()
originalX = getOriginalX()

titleStyle = createStyle(fontSize = 14, fontColour = "#FFFFFF", halign = "center", fgFill = "#4F81BD")

wb = createWorkbook()  

[MikeJSeo]

Is this still not working.. maybe did you try the example that is provided (i.e. Two class.xlsx)?

[ruzy1]

Unfortunately, it's still not working. I did try the Two Class example, and it worked perfectly. But my own data is not working. I don't see any cells that are missing values, so I don't think missing values would be an issue. Do you have any ideas of what may be creating the ylim issue?

[MikeJSeo]

Hmm yea I have no clue..
Maybe if you haven't already you can subset (delete) some of your data and try to narrow down data points that is causing the issue. Or does it not work even for the smallest data you can run?
You can also send me some permuted dataset of yours that doesn't work and I can take a look.

[SamGG]

As @MikeJSeo said, without some data and the parameters to reproduce the problem, it is difficult to solve the error. The ylim depends on ymax which is computed from FDRs. Probably one of those values is NA or Inf although the initial data are OK.

	temp = samr.assess.samplesize.obj$results
	ymax = max(c(temp[, "FDR,1-power", ], temp[, "FDR90", ],
		temp[, "FDR10", ], temp[, "FNR,type1 error", ], temp[,
			"FDR90", ], temp[, "FDR10", ]))

https://github.com/cran/samr/blob/bf0d63f2bc4d93ba7baad0dabdce3bf9d22c5ccf/R/samr.morefuns.R#L1334C3-L1334C3