How to correctly incorporate constraint information into ab initio structure prediction?

Member Site Forums Rosetta 3 Rosetta 3 – Applications How to correctly incorporate constraint information into ab initio structure prediction?

Viewing 1 reply thread
  • Author
    Posts
    • #3081
      Anonymous

        Hi, I am now trying to do an ab initio structure prediction with the constraint information generated from Gremlin.

        My script of ab initio structure prediction with Rosetta is as follows, with the part related to constraint data bolded:

        #!/bin/bash

        #$ -S /bin/bash

        #$ -j y

        #$ -cwd

        #$ -q all.q

        #$ -pe mpi 128

        #$ -hard -l h=”node19|node20″

        ROSETTA_TOP=/home/xtal/rosetta3.9/rosetta_bin_linux_2018.09.60072_bundle/main/source/bin

        ROSETTA_DB=/home/xtal/rosetta3.9/rosetta_bin_linux_2018.09.60072_bundle/main/database

        /opt/openmpi/3.1.0/bin/mpirun $ROSETTA_TOP/AbinitioRelax.mpi.linuxgccrelease -database $ROSETTA_DB

        -in:file:fasta ./input_files/f22.fasta

        -in:file:frag3 ./input_files/aat000_03_05.200_v1_3

        -in:file:frag9 ./input_files/aat000_09_05.200_v1_3

        -nstruct 100000

        -out:pdb

        -abinitio:relax

        -psipred_ss2 ./input_files/t000_.psipred_ss2

        -abinitio::increase_cycles 10

        -abinitio::rg_reweight 0.5

        -abinitio::rg_reweight 0.5

        -abinitio::rsd_wt_helix 0.5

        -abinitio::rsd_wt_loop 0.5

        -relax::fast

        -constraints::cst_file ./input_files/1549559600_cb.cst

        -constraints::cst_weight 10.0

        -constraints::cst_fa_file ./input_files/1549559600_cb.cst

        -constraints::cst_fa_weight 10.0

         

        1549559600_cb.cst is the constraint file from Gremlin, which reads partially like:

        AtomPair CB 48 CB 59 SCALARWEIGHTEDFUNC 3.327 SUMFUNC 2 SIGMOID 8.008 2.786 CONSTANTFUNC -0.5

        AtomPair CB 49 CB 124 SCALARWEIGHTEDFUNC 2.323 SUMFUNC 2 SIGMOID 6.801 5.525 CONSTANTFUNC -0.5

        AtomPair CB 129 CB 133 SCALARWEIGHTEDFUNC 2.276 SUMFUNC 2 SIGMOID 10.713 2.755 CONSTANTFUNC -0.5

         

        However, when I search for constraint-related information in the log file (e.g. the file name of constraint file “1549559600_cb.cst“), 

        I could not find any signs that Rosetta was using my constraint file to do ab initio structure prediction.

        Does anyone discover there is something wrong with my script? May I know how to correct the script?

        Thank you very much.

         

      • #14579
        Anonymous

          I’m not sure if this is a result of the formatting from posting this here, but is it looks like you forgot a backslash after ‘relax:fast’

           

          That will cause the ‘line’ to not continue on the next one, and will treat ‘-constraints::cst_file ./input_files/1549559600_cb.cst’ as a separate command.

           

          you can check this by running your script with ‘./myscript &> out.log’ and the third line should contain the rosetta command, and the arguments should match the ones in your script.

           

          • #14589
            Anonymous

              danpf, you are correct..

              After correcting the script it is running with the constraint input.

              Thanks for your help!!!

        Viewing 1 reply thread
        • You must be logged in to reply to this topic.