Empty InterproScan output

Cheers,
I recently received two independent “errors” from interproscan after running it in two different datasets. Both nucleotides fasta with more than 100k seqs, ran in nucleotide mode. I quoted “errors” because I did not receive actual bug/error report, instead, the jobs finished and the outputs (tsv) were empty; no error message and no std error output (see run info below).
I’ll try run in the other instances (.org, .au) and see if it is a specific .eu problem. In the meantime I hope to get some helpful opinion this.
Thanks!

###tool version
toolshed.g2.bx.psu.edu/repos/bgruening/interproscan5/interproscan/5.0.0

### Command Line
interproscan.sh -dp --input /data/dnb03/galaxy_db/files/a/a/9/dataset_aa95a83b-926e-4ed7-8561-2e33759e001a.dat --seqtype n -f TSV --applications TIGRFAM,PIRSF,ProDom,Panther,SMART,PrositeProfiles,PrositePatterns,HAMAP,PfamA,PRINTS,SuperFamily,Coils,Gene3d,SignalP-EUK,Phobius,TMHMM --tempdir $TEMP --pathways --goterms --iprlookup -o /data/dnb03/galaxy_db/job_working_directory/011/505/11505935/outputs/galaxy_dataset_fa4479c9-10de-42d7-9035-96b0fefa9d68.dat 2>&1


###stdout
/data/dnb03/galaxy_db/job_working_directory/011/505/11505935/working
/opt/galaxy

fooooo
/data/0/interproscan/interproscan-5.36-75.0/
/bin/java
/bin/java
23/09/2020 23:44:14:531 Welcome to InterProScan-5.36-75.0
23/09/2020 23:44:24:260 Running InterProScan v5 in STANDALONE mode... on Linux
23/09/2020 23:44:58:842 Loading file /data/dnb03/galaxy_db/files/a/a/9/dataset_aa95a83b-926e-4ed7-8561-2e33759e001a.dat
23/09/2020 23:44:58:852 Running the following analyses:
[Coils-2.2.1,Gene3D-4.2.0,Hamap-2019_01,PANTHER-14.1,Pfam-32.0,Phobius-1.01,PIRSF-3.02,PRINTS-42.0,ProDom-2006.1,ProSitePatterns-2019_01,ProSiteProfiles-2019_01,SignalP_EUK-4.1,SMART-7.1,SUPERFAMILY-1.75,TIGRFAM-15.0,TMHMM-2.0c]
2020-09-24 00:39:26,642 [amqEmbeddedWorkerJmsContainer-2] [uk.ac.ebi.interpro.scan.business.sequence.NucleotideSequenceLoader:132] WARN - You are analysing more than 2000 nucleotide sequences.  Either use an external tool to translate the sequences or Chunk the input and then send the chunks to InterProScan. Refer to  https://github.com/ebi-pf-team/interproscan/wiki/ScanNucleicAcidSeqs#improving-performance
24/09/2020 00:47:20:215 Loading file /data/dnb03/galaxy_db/job_working_directory/011/505/11505935/tmp/vgcnbwc-compute-c20me-8483.novalocal_20200923_234424257_c6n2//jobLoadNucleicAcidSequence/orfs.filtered.fasta
24/09/2020 00:47:20:218 Running the following analyses:
[Coils-2.2.1,Gene3D-4.2.0,Hamap-2019_01,PANTHER-14.1,Pfam-32.0,Phobius-1.01,PIRSF-3.02,PRINTS-42.0,ProDom-2006.1,ProSitePatterns-2019_01,ProSiteProfiles-2019_01,SignalP_EUK-4.1,SMART-7.1,SUPERFAMILY-1.75,TIGRFAM-15.0,TMHMM-2.0c]
Pre-calculated match lookup service DISABLED.  Please wait for match calculations to complete...

UPDATE!!!
I’ve just tested the same job with a single nucleotide sequence in nucleotide mode and got the same empty output…
Right after this I ran the same single sequence in protein mode AND IT WORKED
So the problem is that the “Type of the input sequences” parameter is inverted?!
Please leave some reply
Thanks

Can anyone tell me what line 26 on InterProScan is? I’m searching for that because my job ended almost immediately. I had no error message, but the tool standard output says:

/data/dnb03/galaxy_db/job_working_directory/033/754/33754599/tool_script.sh: line 26: interproscan.sh: command not found

It seems like there is no shell for InterProScan – is that right?

Any advice would be appreciated!

Hi @jaredbernard
I am trying to figure out the error’s origin.

1 Like