You are here

Questions for "modeling_instructions.pdf" in "Homology or Loop Modelling"

11 posts / 0 new
Last post
Questions for "modeling_instructions.pdf" in "Homology or Loop Modelling"
#1

Dear Sir or Madam,
I am trying to following "modeling_instructions.pdf" in "Homology or Loop Modelling".

http://www.meilerlab.org/index.php/jobs/resources

The first thing I need help is
Section "2. Run ROSETTA Comparative Model protocol".
After typing the COMMAND FOR COMPARATIVE MODELING PROTOCOL (attached) into terminal, I got "Exit 255" in the prompt and "ERROR" in the comparative_model.log (attached). The options file is also attached.

As far as I understand, I actually replace $ with my local directory, and both "#" and "-" should be properly used. So what is the problem?

Thank you very much.

Yours sincerely
Cheng

1) COMMAND FOR COMPARATIVE MODELING PROTOCOL & corresponding prompt

lanselibai@ubuntu:~$ ~/Cheng/rosetta_2014.30.57114_bundle/main/source/bin/minirosetta.linuxgccrelease @/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/comparative_model.options -database ~/Cheng/rosetta_2014.30.57114_bundle/main/database >& /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_output_model/comparative_model.log &
[7] 5215
[6] Exit 255 ~/Cheng/rosetta_2014.30.57114_bundle/main/source/bin/minirosetta.linuxgccrelease @/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/comparative_model.options -database ~/Cheng/rosetta_2014.30.57114_bundle/main/database &> /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_output_model/comparative_model.log
lanselibai@ubuntu:~$

2) comparative_model.log

ERROR: ERROR: Comments in an option file must begin with '#', options must begin with '-' the line:

is incorrectly formatted
caught exception ERROR: ERROR: Comments in an option file must begin with '#', options must begin with '-' the line:

is incorrectly formatted

3) comparative_model.options

#before running rosetta with this options file, change $WORKSHOP_ROOT to the path of the tutorials directory on your local machine

-run:protocol threading #call threading protocol
-in:file:fasta /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/2foxA.fasta #target sequence
-in:file:template_pdb /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/1f4pA.pdb #template structure
-in:file:fullatom #input will be fullatom
-in:file:psipred_ss2 /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/2foxA.psipred_ss2 #optional: psipred secondary structure
-in:file:alignment /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/2foxA.1f4pA.aln #input alignment
-loops:frag_sizes 9 3 1 #sizes of fragments
-loops:frag_files /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/aa2foxA09_05.200_v1_3 /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/aa2foxA03_05.200_v1_3 none #fragment files
-loops:remodel quick_ccd #use ccd to remodel loops
-idealize_after_loop_close #idealize structure after closing loops (structure will have ideal rosetta bond lengths and angles)
-loops:extended true #force extended on loops (phi-psi angles set to 180 degrees) independent of loop input file. for rebuilding loops entirely.
-loops:build_initial true #precede loop modeling with initial round of removing missing densities and building simple loops
-loops:relax fastrelax #fastrelax loops (5 cycles at default)
-cm:min_loop_size 4 #minimum size of loops to consider for rebuild
-out:nstruct 2 # number of models to build
-out:file:silent_struct_type binary #output file type
-out:file:silent /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_output_model/2foxA_models.out #output file name
-out:file:fullatom #output file will be fullatom

Post Situation: 
Sun, 2014-09-21 09:02
lanselibai

I'm guessing there's an issue with non-printing characters, likely from bouncing the file through Windows. (The Windows line ending is the two character sequence CR-LF, whereas the Unix line ending is just the single charachter LF. That extra "CR" charachter may be the cause of Rosetta's confusion.)

There's an application on Unix called "dos2unix" which will fix such formatting issues. (You may need to install it, as it doesn't always get installed by default.) Just run "dos2unix comparative_model.options" before re-running your command and see if that helps.

Mon, 2014-09-22 08:50
rmoretti

Hi R Moretti,
After installing dos2unix and converting the option file into Unix format, the error of "Exit 255" disappeared. However, prompt like "[1] 3891" still came out. In the meantime, there is only one file of "comparative_model.log" in the "my_output_model" folder, which is without anything.

I copy the whole command into terminal and press "Enter" to invoke executable, option file, database, etc. Do you think this is a proper way to do?

Thank you very much.

Yours sincerely
Cheng

lanselibai@ubuntu:/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model$ dos2unix comparative_model.options
dos2unix: converting file comparative_model.options to Unix format ...
lanselibai@ubuntu:/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model$ ~/Cheng/rosetta_2014.30.57114_bundle/main/source/bin/minirosetta.linuxgccrelease @/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/comparative_model.options -database ~/Cheng/rosetta_2014.30.57114_bundle/main/database >& /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_output_model/comparative_model.log &
[1] 3891
lanselibai@ubuntu:/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model$

Mon, 2014-09-22 10:03
lanselibai

That's expected. That comes from the Unix shell (I believe "bash" in your case) because your commandline has a "&" at the end of it. That tells the shell to run the program in the background, instead of waiting until it's completed. The "[1] 3891" means that the program running in the background is shell job 1 with a Unix PID of 3891. These numbers are useful if you want to do job control later (e.g. if you wanted to cancel running the job.).

You'll want to inspect the "/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_output_model/comparative_model.log" file to see the output which Rosetta is making (the ">&" means to redirect the output to the specified file instead of being output to the terminal.). Also, if you wait until the program is done running, the shell will eventually print a "job done" message. (Although you may need to hit enter at the commandline to see it.)

I'd recommend looking at some of the many Unix commandline tutorials that are on the internet. If you're going to be working with Rosetta frequently, it's well worth the investment in time. Some tutorials I might recommend are listed here: http://structbio.vanderbilt.edu/bsdf/workshop/workshop_prep.php

Tue, 2014-09-23 08:20
rmoretti

Hi R Moretti,
Many thanks. The UNIX commandline tutorials are very useful. I really like them.

Yours sincerely
Cheng

Tue, 2014-09-30 04:44
lanselibai

(Please ignore this post)

Tue, 2014-09-30 07:10
lanselibai

Dear friends,
I have another question for "modeling_instructions.pdf" in "Homology or Loop Modelling".

In the "2. Run ROSETTA Comparative Model protocol", I got the following message after submitting the command line

lanselibai@ubuntu:/mnt/hgfs/Downloads$ ~/Cheng/rosetta_2014.30.57114_bundle/main/source/bin/minirosetta.linuxgccrelease @/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/comparative_model.options -database ~/Cheng/rosetta_2014.30.57114_bundle/main/database >& /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_output_model/comparative_model.log &
[1] 3151

...(omit previous prompt)
protocols.comparative_modeling.ThreadingJobInputter: Warning: no template pdb provided for alignment
protocols.evaluation.ChiWellRmsdEvaluatorCreator: Evaluation Creator active ...
protocols.jd2.JobDistributor: no more batches to process...
protocols.jd2.JobDistributor: 0 jobs considered, 0 jobs attempted in 0 seconds
protocols.jd2.JobDistributor: no jobs were attempted, did you forget to pass -overwrite?

The attachments are the log file and options file.

I could not understand why "no template pdb provided" as I really put a 1f4pA.pdb file in the directory:

lanselibai@ubuntu:/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model$ ls
1f4pA.fasta 2foxA.fasta aa2foxA09_05.200_v1_3 PrepByMyOwn
1f4pA.pdb 2foxA.psipred_ss2 COMMAND_FOR_COMPARATIVE_MODELING_PROTOCOL.txt S_0001.pdb
2foxA.1f4pA.aln aa2foxA03_05.200_v1_3 comparative_model.options score.sc

I also could not understand why it said "forget to pass -overwrite?"

Can I ask how to solve this?

Thank you very much.

Yours sincerely
Cheng

Tue, 2014-09-30 07:13
lanselibai

You'll actually get that message if the name of the template in your alignment doesn't match the name of the PDB. (Technically its only the first five characters of each that need to match, and it's safest if it's all uppercase in the alignment file.) Keep in mind that it's the second sequence in the alignment that's considered as the template - the first should be the sequence you're trying to thread.

The "forget to pass -overwrite?" is actually referring to another common reason for no jobs being attempted - there being existing outputs already in the directory, indicating that the run has already been done once before, meaning you need to pass -overwrite to overwrite those structures.

Tue, 2014-09-30 13:40
rmoretti

Hi R Moretti,
Thank you for your help.

1) I have checked my .aln file. Just as you said, the first row is target sequence and the second row is template sequence (attached).

Does the name of the .aln file matter? Should it be target.template.aln or template.target.aln? I tried both but got the same problem.

I also tried to use upper letters to name the PDB file and alignment file as well as corresponding positions in the options file. But the same problem (i.e.Warning: no template pdb provided for alignment) happened.

2) For "-overwrite" issue, do you mean there is already an old output file in the output directory, which prevents the new output file to be generated? However, in my case, no any old output file exists in the output directory.

I also tried to add "-overwrite" in my command line. But "forget to pass -overwrite?" still exists.

Yours sincerely
Cheng

File attachments: 
Wed, 2014-10-01 03:29
lanselibai

Sorry, for the capitol letters, I was specifically referring to the naming of the sequence within the alignment file. SO it should be

1F4PA 1 PKALI...

For the -overwrite, the message being printed is incorrect in your case. The rationale for it is another common case (which doesn't apply for you), where the "no jobs run" condition is met is where someone is re-running an already (successfully) run protocol with slightly different settings. In that case -overwrite is the correct solution. Unfortunately, Rosetta can't tell the difference between that case and your case, so it prints out the help message which doesn't apply in your particular situation.

Mon, 2014-10-06 16:06
rmoretti

Hi R Moretti,
Many thanks for your help. However, even though I have changed all the lower case names into upper case ones, the same problem (i.e. Warning: no template pdb provided for alignment) still exists.

The attachments are
1) log file containing Warning message
2) Options file
3) Alignment file

(I literarily follow all the instructions in "modeling_instructions.pdf" (Meiler_Lab) and tried both lower case and upper case to name the target and template.)

The command line I am using is:

lanselibai@ubuntu:~$ ~/Cheng/rosetta_2014.30.57114_bundle/main/source/bin/minirosetta.linuxgccrelease @/mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_input_model/comparative_model.options -database ~/Cheng/rosetta_2014.30.57114_bundle/main/database >& /mnt/hgfs/ROSETTA/Tutorial/Meiler_Lab/Rosetta_Tutorials/tutorials/modeling/my_output_model/comparative_model.log &

Do you think it is another reason that causes the Warning message?

Thank you very much.

Yours sincerely
Cheng

Tue, 2014-10-07 02:58
lanselibai