You are here

"sparks failed" error when running "make_fragments.pl"?

7 posts / 0 new
Last post
"sparks failed" error when running "make_fragments.pl"?
#1
-Precision-7920-Tower:make_fragments.pl -verbose myfasta.fasta 
no id specified. parsing filename instead.
cannot parse id from filename so using 't001_'
ID: t001 CHAIN: _
File for psipred not found! Generating from scratch instead.
picking fragments with options:
                       DEBUG: 1
            add_pdbs_to_vall: 
                       chain: _
                     cleanup: 1
             csbuild_profile: 0
exclude_homologs_by_pdb_date: 0
                           f: myfasta.fasta
                   fastafile: t001_.fasta
                       frags: 1
                        homs: 1
                          id: t001
                n_candidates: 1000
                     n_frags: 200
             old_name_format: 0
                      porter: 0
                  porterfile: 
                     psipred: 1
                 psipredfile: 
                      rundir: /rosetta/rosetta-abintio-prediction
                       runid: t001_
                         sam: 0
                     samfile: 
                 torsion_bin: 0
                     verbose: 1
--------------------------------------------------------------------------------

FILENAME: t001_.fasta
Sequence: MAVGEIQISALPQAALPIDLSDIFHLKQGIEDKRCTLEQLLAPHSSLKNNPHGVTKTQIGLDNVINALQLVAANNLSDIVNVAEARANLQIMSSEEVNSLIQQHINDKSNPHNTTKAQVGLSNVQNWTTSNLYNEDADKYATARAVNNLYKAVQASYPVGTIHLSMNPANPSTYLICGGTWELVSKGRALVGYDSDSRPVGSTFGSQTTALTVNNIPAHTHSVYVTGGGHTHGATASISAFDYGTKGTTTFDYGTKTVSTYDYGNRTTSAAGLTQTTFINRGYADGGWRGLSVEGASYDPRATLTTANNHTHSVAIGAHNHTVGIGAHSHTVAIGSHNHSATVTVTGAGEHSHSGTTGPTGSGQAFSVEQPSFVLYVWQRTA
Running sparks for phi, psi, and solvent accessibility predictions
t001_.fasta.pssm exists
At line 239 of file phipsi_ss0.f (unit = 77, file = '/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/SPINE-X/weights/ss')
Fortran runtime error: End of file
Aborting: Can't run first SS0 predictor
Error in file: t001_.fasta.phipsi
Traceback (most recent call last):
  File "/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/bin/buildinp.py", line 255, in run1
    buildinp(fphipsiss, fmat, finp)
  File "/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/bin/buildinp.py", line 238, in buildinp
    seq1, ssec1, phipsi1, Fphipsi = rdphipsi(fphipsiss)
  File "/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/bin/buildinp.py", line 9, in rdphipsi
    for line in file(fn):
IOError: [Errno 2] No such file or directory: 't001_.fasta.phipsi'
sparks path length > 100 characters, which may cause internal errors. Try moving to a shorter path prefix.
sparks failed!

Dear Friends, 

 

the above error comes when running "make_fragments.pl". Can you please let me know how this can be solved? 

 

Thanks!

Post Situation: 
Tue, 2019-09-03 03:43
Danielsebas

Hello,

as the error says:

sparks path length > 100 characters, which may cause internal errors. Try moving to a shorter path prefix.
sparks failed!

the poblem is that sparksX is hardcoded to look at only the first 100 or so characters of a file path.

/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/SPINE-X/weights/ss

so if you can, make a smylink to the tools folder much higher up in directory space.

ie something like:

ln -s /media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/ /media/Deepak/rosetta_tools

or just move the tools folder 'higher up' and it will work.

 

make sure you run from /media/Deepka/rosetta_tools if you use that method though.it has to 'think' it's a shorter path.

 

Tue, 2019-09-03 10:24
danpf

thanks! I did this:

 

ln -s /media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/ /media/Deepak/NPRP10/rosetta_tools

 

and then:

 

Precision-7920-Tower:/media/Deepak/NPRP10/rosetta_tools/tools/fragment_tools# perl make_fragments.pl -verbose myfasta.fasta 
Use of uninitialized value $ENV{"FRAGMENT_PICKER_NUM_CPUS"} in int at make_fragments.pl line 35.
Use of uninitialized value $ENV{"BLAST_NUM_CPUS"} in int at make_fragments.pl line 41.
Use of uninitialized value $ENV{"PSIPRED_USE_weights_dat4"} in int at make_fragments.pl line 52.
Use of uninitialized value $ENV{"SLAVE_LAUNCHER_MAX_JOBS"} in int at make_fragments.pl line 71.
Use of uninitialized value $ENV{"SLAVE_MAX_WAIT"} in int at make_fragments.pl line 73.
Use of uninitialized value $ENV{"SLAVE_MAX_ATTEMPTS"} in int at make_fragments.pl line 74.
Use of uninitialized value $ENV{"PDB2VALL_IGNORE_ERRORS"} in int at make_fragments.pl line 80.
no id specified. parsing filename instead.
cannot parse id from filename so using 't001_'
ID: t001 CHAIN: _
File for psipred not found! Generating from scratch instead.
picking fragments with options:
                       DEBUG: 1
            add_pdbs_to_vall: 
                       chain: _
                     cleanup: 1
             csbuild_profile: 0
exclude_homologs_by_pdb_date: 0
                           f: myfasta.fasta
                   fastafile: t001_.fasta
                       frags: 1
                        homs: 1
                          id: t001
                n_candidates: 1000
                     n_frags: 200
             old_name_format: 0
                      porter: 0
                  porterfile: 
                     psipred: 1
                 psipredfile: 
                      rundir: /media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools
                       runid: t001_
                         sam: 0
                     samfile: 
                 torsion_bin: 0
                     verbose: 1
--------------------------------------------------------------------------------

FILENAME: t001_.fasta
Sequence: MAVGEIQISALPQAALPIDLSDIFHLKQGIEDKRCTLEQLLAPHSSLKNNPHGVTKTQIGLDNVINALQLVAANNLSDIVNVAEARANLQIMSSEEVNSLIQQHINDKSNPHNTTKAQVGLSNVQNWTTSNLYNEDADKYATARAVNNLYKAVQASYPVGTIHLSMNPANPSTYLICGGTWELVSKGRALVGYDSDSRPVGSTFGSQTTALTVNNIPAHTHSVYVTGGGHTHGATASISAFDYGTKGTTTFDYGTKTVSTYDYGNRTTSAAGLTQTTFINRGYADGGWRGLSVEGASYDPRATLTTANNHTHSVAIGAHNHTVGIGAHSHTVAIGSHNHSATVTVTGAGEHSHSGTTGPTGSGQAFSVEQPSFVLYVWQRTA
Running sparks for phi, psi, and solvent accessibility predictions
running psiblast for sequence: t001_.fasta
At line 239 of file phipsi_ss0.f (unit = 77, file = '/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/SPINE-X/weights/ss')
Fortran runtime error: End of file
Aborting: Can't run first SS0 predictor
Error in file: t001_.fasta.phipsi
Traceback (most recent call last):
  File "/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/bin/buildinp.py", line 255, in run1
    buildinp(fphipsiss, fmat, finp)
  File "/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/bin/buildinp.py", line 238, in buildinp
    seq1, ssec1, phipsi1, Fphipsi = rdphipsi(fphipsiss)
  File "/media/Deepak/NPRP10/softwares/rosetta/rosetta_src_2019.31.60840_bundle/tools/fragment_tools/sparks-x/bin/buildinp.py", line 9, in rdphipsi
    for line in file(fn):
IOError: [Errno 2] No such file or directory: 't001_.fasta.phipsi'
sparks path length > 100 characters, which may cause internal errors. Try moving to a shorter path prefix.
sparks failed!

 

It still takes the old path for some reason. Can you please let me know what am I doing wrong? thanks!

Wed, 2019-09-04 03:19
Danielsebas

Hrm that's odd... unfortunately it seems like you will have to remove the symlink and mv the folder there completely.  sorry : (

Wed, 2019-09-04 07:08
danpf

thanks! do I need to move only the "spaks-x" folder up the path or all folders; database (657 GB), psipred, csblast, blast. I mean all the folders generated by "install_dependencies.pl"? thanks for your suggestions!

Thu, 2019-09-05 00:39
Danielsebas

Ok, it works now, I just played around with sym linking appropriate folders and finally it works. thanks!

Thu, 2019-09-05 03:09
Danielsebas

Hello,

I am currently facing the exact problem you did. Would you be able to share exactly how you set up your symlinks? I tried the initial one suggested by danpf and it also failed for me as well. It would be extremley helpful and appreciated. Thanks!

Tue, 2020-03-10 14:39
Jacob_Verburgt