Fatal error: heatmap2 tool

I am attempting to follow the Galaxy tutorial for reference-based RNA-sequencing (https://galaxyproject.github.io/training-material/topics/transcriptomics/tutorials/ref-based/tutorial.html#visualization-of-the-differentially-expressed-genes).

Using Galaxy, I have successfully (1) extracted the most DE genes from the normalized counts table and (2) extracted the normalized counts into a table. Currently, I am attempting to visualize my differential expressed genes. I have been unable to plot the heatmap with my normalized counts using the tool heatmap2. I could not find others in the community with a similar issue. Any help would be greatly appreciated!

I am using the tool heatmap2 (toolshed.g2.bx.psu.edu/repos/iuc/ggplot2_heatmap2/ggplot2_heatmap2/2.2.1+galaxy1).

Parameters used were:
Data transformation - Log2(value) transform my data
Enable data clustering - True
Labeling columns and rows - Label columns and not rows
Coloring groups - Blue to white to red
Data scaling - Do not scale my data

Error Details
Tool generated the following standard error:
Fatal error: Exit code 1 ()
Attaching package: ‘gplots’
The following object is masked from ‘package:stats’:
lowess
Error in hclust(x, method = “complete”) :
NA/NaN/Inf in foreign function call (arg 11)
Calls: heatmap.2 -> hclustfun -> hclust

1 Like

I met the same error and changed the parameter “Data transformation” to - Log2(value+1) transform my data. Finally it plotted the heatmap successively. I think there should be “zero” in your data too. So after “Log2(value)” it got the Inf value in the data.
(Sorry for my poor English.)

2 Likes

Bookmarking.

This might be a bug with the tutorial. I made an issue for it and we will review/test and fix it if needed. Or, that term works with the tutorial data but not for some class of other data.

1 Like

Sounds reasonable, I had the same problem, used value+1 and it got through, thanks!

1 Like

I am still challenging with this error even though I am trying the “Log2(value+1) transform my data” transformation! I am actually wonder why and how to figure it out ?!