Diffbind 'no diff/ly accessible sites' error?

Hello, I’m running a diffbind operation between two conditions from an ATAC-seq experiment with two replicates in each. However, I get an error message resulting in my output files being buggy/red.

Warning message:
In Sys.setlocale(“LC_MESSAGES”, “en_US.UTF-8”) :
OS reports request to set locale to “en_US.UTF-8” cannot be honored
6hVC1_peaks_narrowPeak 6hVC1_peaks_narrowPeak VC NA raw
6hVC2_peaks_narrowPeak 6hVC2_peaks_narrowPeak VC NA raw
6hdTAG2_peaks_narrowPeak 6hdTAG2_peaks_narrowPeak dTAG NA raw
6hdTAG3_peaks_narrowPeak 6hdTAG3_peaks_narrowPeak dTAG NA raw
converting counts to integer mode
gene-wise dispersion estimates
mean-dispersion relationship
final dispersion estimates
Warning message:
No sites above threshold
Error in pv.getPlotData(DBA, attributes = attributes, contrast = contrast, :
Unable to plot – no sites within threshold
Calls: dba.plotHeatmap → pv.getPlotData

Does “No sites above threshold” in the error message just suggest that there were no differentially accessible sites (which I was expecting)? If so, why does the output seem to suggest there was an error in the process, rather than just returning an ‘empty’ interval file or some ‘blank’ result? I’m quite new to using diffbind on Galaxy, so I’m not too familiar with the various error messages that might occur. Thank you!

Yes, that is what the message means.

The tool wrapper is around a Bioconductor tool written in R. Trapping errors for these cases usually relies on the output from R, and the same tool/inputs would likely fail with the same message if run directly.

No problem! Some tools run in Galaxy will produce more meaningful messages than others. Much depends on how error trapping was implemented in the original 3rd party tool. :slight_smile:

The GTN FAQs here can help. Search this page with keywords or error messages:

And this specific FAQ explains how to find all of the failure details:

Hi @jennaj, thank you for your detailed responses to both my questions, and for the FAQ links!

1 Like