nextgenusfs
SignalP v6?
Hi John,
I've got some SingalP6 output I'd like to include in my annotate stage. I've tried the --signalp method but it threw an error. I'm using funannotate v1.8.7. I'll pop in the error then a head of the singalp output file. Maybe you can advise? I suspect it's something to do with the format but can't quite figure it out. Ideally I wouldve liked funannotate to run it natively but it didnt want to pickup the installation.
Existing SignalP results found: Predict/annotate_misc/signalp.results.txt
Traceback (most recent call last):
File "/home/chris/miniconda3/envs/Funannotate/bin/funannotate", line 10, in
head signalp.results.txt
SignalP-6.0 Organism: Eukarya Timestamp: 20211008163009
ID Prediction OTHER SP(Sec/SPI) CS Position
FUN_000001-T1_FUN_000001 NO_SP 1.000034 0.000000 FUN_000002-T1_FUN_000002 NO_SP 1.000061 0.000000 FUN_000003-T1_FUN_000003 NO_SP 1.000047 0.000000 FUN_000004-T1_FUN_000004 NO_SP 1.000045 0.000000 FUN_000005-T1_FUN_000005 NO_SP 0.999985 0.000069 FUN_000006-T1_FUN_000006 NO_SP 1.000068 0.000000 FUN_000007-T1_FUN_000007 NO_SP 1.000054 0.000000 FUN_000008-T1_FUN_000008 NO_SP 1.000050 0.000001
Cheers, Chris
Looks like another new format.... is it tab delimited or space? And what does a "has signal peptide" look like?
It's tab delimited. Here are the first two identified with signal peptides.
SignalP-6.0 Organism: Eukarya Timestamp: 20211008163009
ID Prediction OTHER SP(Sec/SPI) CS Position
FUN_000010-T1_FUN_000010 SP 0.000237 0.999737 CS pos: 21-22. Pr: 0.9766 FUN_000017-T1_FUN_000017 SP 0.000320 0.999666 CS pos: 25-26. Pr: 0.4228
Thanks can you attach file so I can make sure the parser gets the last column correctly? I do t want to make any assumptions based on if GitHub stripped any formatting, etc.
Thanks. How about the call to signalP v6, this is what is being run on v5:
cmd = ['signalp', '-stdout', '-org', 'euk', '-format', 'short', '-fasta']
Does that still work with v6?
Parser should be working (I hope) in the master branch, can install with pip to check.
Hiya,
I've found time again to work on this, I just pulled the new master branch and had a go, I've hit a new error.
: Existing SignalP results found: Predict/annotate_misc/signalp.results.txt
Traceback (most recent call last):
File "/home/chris/miniconda3/envs/Funannotate/bin/funannotate", line 8, in
Oh, they must not write the last empty column if there is no hit. That's not very nice of them.... should just be able to put a try/except on that line.
Okay, can you try with latest in master again and see if it prints any lines to stderr that it cannot parse?
Heya, looks like it's sort of worked. There appear to be a handful of entries it cannot parse. See below for the output.
Existing SignalP results found: Predict/annotate_misc/signalp.results.txt signalP parse error: FUN_001618-T1_FUN_001618 SP 0.379709 0.620269 signalP parse error: FUN_001686-T1_FUN_001686 SP 0.491810 0.508172 signalP parse error: FUN_003357-T1_FUN_003357 SP 0.449099 0.550900 signalP parse error: FUN_003817-T1_FUN_003817 SP 0.398018 0.601982 signalP parse error: FUN_004349-T1_FUN_004349 SP 0.432366 0.567619 signalP parse error: FUN_005570-T1_FUN_005570 SP 0.471067 0.528935 signalP parse error: FUN_005972-T1_FUN_005972 SP 0.409434 0.590562 signalP parse error: FUN_008891-T1_FUN_008891 SP 0.363779 0.636212 [Oct 21 02:55 PM]: 697 secretome and 1,949 transmembane annotations added
Oh, also, I get this when I try to include phobius results. Seems to be some sort of clash. I've run this directly after my last test using --phobius phobius.results.txt
Existing SignalP results found: Predict/annotate_misc/signalp.results.txt
Traceback (most recent call last):
File "/home/chris/miniconda3/envs/Funannotate/bin/funannotate", line 8, in
Okay thanks. The ones that it can't parse don't have the last column, that must be a bug or something. It is predicting a signal peptide but not the cleavage site?
Second error makes sense I think, let me see what I need to do to make new results work.
Hey @Dikaryotic , saw this issue by chance, could you provide me with the sequences that show no CS being predicted? This is definitely unexpected behaviour.
Hello,
I recently installed funannotate and it appear the signalp6 pursing error is still there "ValueError: not enough values to unpack (expected 5, got 4)" when passing signalp6 results to funannotate annotate.
Have anybody managed a solution for it ?
I am also having the same issue. What is the exact input file type or path supposed to be for --signalp ?
--signalp predict_results/signalp/prediction_results.txt \
Existing SignalP results found: /mnt/ufs18/rs-008/Hausbeck_group/Lukasko/BotrytisDNASeq/CCR7/Predict_Annotate/AI7_fun/annotate_misc/signalp.results.t
xt
-------------------------------------------------------
Traceback (most recent call last):
File "/mnt/home/lukaskon/anaconda3/envs/funannotate/bin/funannotate", line 10, in <module>
sys.exit(main())
File "/mnt/home/lukaskon/anaconda3/envs/funannotate/lib/python3.8/site-packages/funannotate/funannotate.py", line 716, in main
mod.main(arguments)
File "/mnt/home/lukaskon/anaconda3/envs/funannotate/lib/python3.8/site-packages/funannotate/annotate.py", line 1341, in main
lib.parseSignalP(signalp_out, secreted_out)
File "/mnt/home/lukaskon/anaconda3/envs/funannotate/lib/python3.8/site-packages/funannotate/library.py", line 7299, in parseSignalP
pos = components[2].split("-")[0]
IndexError: list index out of range
would help to have a snapshot of your failed file and what version of funannotate you are running. are you running signalp outside of funannotate then?
I just ran an annotation with signalp6 installed but I think I had to tweak something in the latest code release so you might need to grab code from github to make sure it is all working.
I am running funannotate v1.8.15 and yes, signalp6 outside of funannotate.
Here is the annotate log file:
B5_signalp_annotate_slurm.txt
Here are the signalp results from annotate_misc:
signalp.results.txt
If the issue is the signalp file format, maybe I can reformat to make this compatible for the time being.
Let me know if anything else would be helpful. Thank you!