Undefined error after running DESeq2

usegalaxy.eu support
, , , ,
1

Hi everyone,
I´m trying to run DESeq2 on my featurecount output.
I use Galaxy Version 2.11.40.7+galaxy2.
Unfortunately, I always get this error message (I tried it 3 times):

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
estimating size factors
estimating dispersions
gene-wise dispersion estimates: 8 workers
mean-dispersion relationship
final dispersion estimates, MLE betas: 8 workers
fitting model and testing: 8 workers
Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'args' in selecting a method for function 'do.call': ignoring SIGPIPE signal
Calls: DESeq ... <Anonymous> -> sendMaster -> .handleSimpleError -> h
Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'args' in selecting a method for function 'do.call': ignoring SIGPIPE signal
Calls: DESeq ... <Anonymous> -> sendMaster -> .handleSimpleError -> h
Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'args' in selecting a method for function 'do.call': ignoring SIGPIPE signal
Calls: DESeq ... <Anonymous> -> sendMaster -> .handleSimpleError -> h
/jetstream2/scratch/main/jobs/51456491/tool_script.sh: line 23: 2239243 Killed                  Rscript '/jetstream2/scratch/main/jobs/51456491/tool_files/deseq2.R' --cores ${GALAXY_SLOTS:-1} -o 'XXXX' -p '/jetstream2/scratch/main/jobs/51456491/outputs/dataset_4afc5bb6-ce6f-4a42-90d3-ae11b562ff53.dat' -A 0.1 -n '/jetstream2/scratch/main/jobs/51456491/outputs/dataset_e1637214-2bfc-46f9-b4a4-7964e68c241c.dat' -H -f '[["infection_status", [{"mock": ["/jetstream2/scratch/main/jobs/51456491/inputs/dataset_abf45c23-7241-40bc-90a0-e1d495944c86.dat", "/jetstream2/scratch/main/jobs/51456491/inputs/dataset_d9467f7d-b3bc-4a61-8d03-44e1d837a3f0.dat", "/jetstream2/scratch/main/jobs/51456491/inputs/dataset_427ca29f-94c0-49dd-babe-518c41c1d2aa.dat"]}, {"E6E7": ["/jetstream2/scratch/main/jobs/51456491/inputs/dataset_2ef9fdc7-c3a5-4c02-91a2-bb777ab58408.dat", "/jetstream2/scratch/main/jobs/51456491/inputs/dataset_a86470cd-d547-4dea-aff6-72792cc0178d.dat", "/jetstream2/scratch/main/jobs/51456491/inputs/dataset_7259eec2-a9c7-4cbb-8000-94d0427ce77b.dat"]}, {"infected": ["/jetstream2/scratch/main/jobs/51456491/inputs/dataset_50696e14-fadf-4419-bf50-03c42dfe4cd1.dat", "/jetstream2/scratch/main/jobs/51456491/inputs/dataset_0380e664-b46c-4122-bc5a-c9ff7def553a.dat", "/jetstream2/scratch/main/jobs/51456491/inputs/dataset_9ecf595e-c510-40db-9de4-85b34413e29a.dat"]}]]]' -l '{"dataset_50696e14-fadf-4419-bf50-03c42dfe4cd1.dat": "Donor_A_HPV31.fastq", "dataset_0380e664-b46c-4122-bc5a-c9ff7def553a.dat": "Donor_B_HPV31.fastq", "dataset_9ecf595e-c510-40db-9de4-85b34413e29a.dat": "Donor_C_HPV31.fastq", "dataset_2ef9fdc7-c3a5-4c02-91a2-bb777ab58408.dat": "Donor_B_HPV31E6.fastq", "dataset_a86470cd-d547-4dea-aff6-72792cc0178d.dat": "Donor_A_HPV31E6E7.fastq", "dataset_7259eec2-a9c7-4cbb-8000-94d0427ce77b.dat": "Donor_C_HPV31E6_7n.fastq", "dataset_abf45c23-7241-40bc-90a0-e1d495944c86.dat": "Donor_A_Mock.fastq", "dataset_d9467f7d-b3bc-4a61-8d03-44e1d837a3f0.dat": "Donor_B_Mock.fastq", "dataset_427ca29f-94c0-49dd-babe-518c41c1d2aa.dat": "Donor_C_Mock.fastq"}' -t 1 -m
sh: error while loading shared libraries: libdl.so.2: cannot open shared object file: No such file or directory
sh: error while loading shared libraries: libdl.so.2: cannot open shared object file: No such file or directory
sh: error while loading shared libraries: libdl.so.2: cannot open shared object file: No such file or directory
Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'args' in selecting a method for function 'do.call': ignoring SIGPIPE signal
Calls: DESeq ... <Anonymous> -> sendMaster -> .handleSimpleError -> h
Error in h(simpleError(msg, call)) : 
  error in evaluating the argument 'args' in selecting a method for function 'do.call': ignoring SIGPIPE signal
Calls: DESeq ... <Anonymous> -> sendMaster -> .handleSimpleError -> h

I started DESeq2 with the following parameters:

  • Select group tags corresponding to levels
  • Factor: Infection_status
  • Factor level 1: infected
  • Factor level 2: transfected with specific viral proteins
  • Factor level 3: mock
  • provide a tabular file with additional batch factors to include in the model: Nothing selected
  • Files have header: Yes
  • Choice of input data: Count data
  • Advanced options: Default
  • Output selector: Generate plots for visualizing the analysis results + Output normalised counts + Output all levels vs all levels of primary factor

When for “output selector”, I deselect “output all levels vs all levels of primary factor”, there is no error message any more. But then, I get only 1 result file with differentially regulated genes. This does not make sense for 3 factor levels.
A workaround would be to run DESeq2 separately three times with only 2 factor levels, e.g. infected vs. mock, infected vs. viral proteins and viral proteins vs. mock.

Do you have any idea why this error appears? Thanks!

2

What happens if you simplify that label? The tool form has instructions: singleWord with no spaces, not starting with a number, and optional underscores only. Shorter is better – I think there is a character limit.

And, this tool only processes a comparison between two factors per level. Limma or EdgeR have more options: contracts, etc. Exact method limits are in the help section.

3

Thank you for your reply!
In fact, the label for factor level 2 was a single word that matches the instructions. I changed the name so that my question gets better understandable. I am sorry that this made my question more confusing.

I want to get three tables with differentially expressed genes:

  • infected vs. mock
  • infected vs. viral proteins
  • viral proteins vs. mock

Should I use Limma or EdgeR instead of DESeq2 for this task?

4

The reasons for using the different tools can involve characteristics of your sample data, so too much to list out here. Both tool forms have a short summary plus a linkout to a User PDF written by the tool authors. You could also just experiment/compare :slight_smile:

Basic examples using Limma in Galaxy are also here: Galaxy Training!