Menu

Home

Teresa Paramo

trj_cavity finds protein cavities througout Molecular Dyanamics (MD) simulation trjectories. The program works with PDB and GROMACS compatible formats such as XTC or TRJ.

Project Members:


Discussion

  • Teresa Paramo

    Teresa Paramo - 2014-03-11

    INSTALLATION INSTRUCTIONS

    trj_cavity can be used with GROMACS or without it. The version using GROMACS takes advatange of GROMACS atom index files and the use of compressed trajectories, which reduces drastically disc usage. Such version needs GROMACS to be installed in your computer.

    However, if you don't want to install GROMACS, there is a stand-alone version of trj_cavity which works with PDB format. The trajectory is then introduced as a text file containing a list with the sequence of PDBs, one PDB path per line.

    Both versions are essentially the same, and in both cases you will need to download the package trj_cavity.tar.gz and decompress it:

    tar -zxf trj_cavity.tar.gz

    Then you need to move to the decompressed folder:

    cd trj_cavity

    If you want the GROMACS version, you should type:

    make

    And an executable called trj_cavity will be generated. If you want the stand alone version instead, type:

    make no_gromacs

    The excutable is now called trj_cavity_ng. An overview of the options available for each version can be found typing:

    ./trj_cavity -h
    or
    ./trj_cavity_ng -h

     

    Last edit: Teresa Paramo 2014-07-01
    • manikya

      manikya - 2019-04-02

      Hello,

      I am trying to install trj_cavity v2.0 with version of Gromacs (version 5.1.2). Unfortunately, following error has happened.
      bioinfo2@bioinfo2-Vostro-3470:~/manikya/tunnel_volume/trj_cavity_v2.0$ make
      Makefile:25: *** "GMXLDLIB not found, please source GMXRC. If you want to build the stand-alone version, type: make no_gromacs". Stop.

       
  • Teresa Paramo

    Teresa Paramo - 2014-03-11

    USING trj_cavity

    The ouptut of trj_cavity basically depends on two parameters, the the degree of buriedness (-dim) of the target cavities and the size of the grid voxel (-spacing). The most important thing you should bear in mind when changing the dim/spacing parameters is: use the same values when you intend to compare the results of diffrent trajectories or proteins!

    degree of buriedness

    The degree of buriedness is basically the minimum number of sides where the protein must surround the cavity. As we are working with voxels, the maximun number of dimensions is 6. It should be noted that the higher the value is, the more restrictive is the search. Therefore, if you just want to find a completely buried inner cavity you should use -dim 6; if you want to allow your cavities to be connected to the bulk in one side you should use -dim 5 instead; if you seek a tunnel you will need a -dim 4, while very exposed surface pockets might require -dim <=3.

    If you don't know which dim value would be the best fit for your cavity, we recommend you to try the different values with your first frame/PDB and then apply the optimal value to the rest of the trajectory. By default the program uses dim=5, which is normally a good starting to point to explore the cavity topology.

    grid spacing

    The size of the grid voxel is related to the precission: the smaller, the more precise (but the program will be slower!). However, as discussed in the paper, the default grid spacing of 1.4 Ansgtroms is normally a good choice, as it corresponds to the effective van der Waals radius of a water molecule, helping to identify solvent-accesible cavities. More importantly, it also makes the results comparable with the rest of the tools. However, it might be some cases were a smaller/bigger grid voxel might be neccesary to define differently sized cavities.

    standard output files

    -o Provides a structure file (for example, a PDB file) of the cavity/ies found when working with a single structure, or the topology file to visualyze the trajectory of the cavity/ies when working with trajectories.

    -ov Provides a file with the volume of the cavity/ies for each trajectory frame.

    -ot Provides a trajctory of the cavity/ies found trhoughout the trajectory provided.

    -ostat PDB file representing all the positions where a cavity was found with its frequency along the trajecotry as percentage in the b-factor column.

    trajectory fitting

    Althought in priciple the search of cavities itself is independent of the coordinates (i.e the cavity should be found even if the protein changes orientation), when using a starting seed point coordinate with -seed, producing statististics or outputting more than one cavity (-mode all), a proper fitting of the tractory coordinates is required.

    While the coordinate dependency of the -seed option is obvious, the identification of a cavity as the same one than in previous frames requires consistent coordinates, which is key in the calulation of the statistics and the simulataneus analysis of several cavities. Therefore, it is recommended to fit the trajectory to a reasonable group of atoms of the starting structure before analysing it. Normally, you should aim to fit the area of the protein surrounding the cavity/ies of interest.

     
  • Teresa Paramo

    Teresa Paramo - 2014-03-11

    SOLVENT ANALYSIS

    trj_cavity can analyse the solvent occupancy of the cavities identified. If you are using the GROMACS version of the program you can specify any atom group, for example a ligand molecule.

    This command line:

    ./trj_cavity -s topology.pdb -f trajecotry.xtc -os solvent_count.xvg -osstat solvent_stats.pdb -n index.ndx

    Produces solvent_count.xvg and solvent_stats.pdb files. The first file provides the number of solvent molecules overlaping the cavities found for each frame, while the second is a PDB representing the positions where solvent atoms were found to overlap the cavities, with their frequency as a precentage in the b-factor column.

    The non-gromacs version doesn't use index files, but will identify as "solvent" any atom with residue names SOL,TIP*,NA or CL.

    ./trj_cavity_ng -f pdb_list.txt -os solvent_count.xvg -osstat solvent_stats.pdb

     

    Last edit: Teresa Paramo 2014-07-01
  • Teresa Paramo

    Teresa Paramo - 2014-03-11

    TUNNEL ANALYSIS

    trj_cavity can characterize tunnels in pores or channel proteins. As mentioned before, you will need to set -dim to 4 (or less), and the alignment axis of the tunnel if is different than z.

    ./trj_cavity -s topology.pdb -f trajecotry.xtc -dim 4 -ob tunnel.pdb -obt tunnel_trajctory.xtc -obs sector_profile.xvg -oba bottleneck.xvg

    or

    ./trj_cavity_ng -f pdb_list.txt -dim 4 -ob tunnel.pdb -obt tunnel_trajctory.xtc -obs sector_profile.xvg -oba bottleneck.xvg

    The analysis is perfomed by "slicing" the cavity found into planes along the alinment axis ensuring connectivity. The average area of those slices along the trajectory is provided in the file sector_profile.xvg. Assuming a circular shape, these results can be output in terms of radius (-sector radius option).

    The minimum of those sectors, i.e. the tunnel bottleneck, can be inspected for each frame in the file bottleneck.xvg. However, you can specify a fixed value of the alignment axis inorder to inspect a specific part of the tunnel (-axis option).

    Finally, the "slices" used to calculate the tunnel properties can be inspected as a trajecotry with tunnel.pdb topology and tunnel_trajectory.xtc.

     

    Last edit: Teresa Paramo 2014-07-01

Log in to post a comment.