CCP4 Tutorial: Session 5 - Molecular Replacement Tutorial, alternative 5c

See also the accompanying document giving background information.

In the following instructions, when you need to type something, or click on something, it will be shown in red. Output from the programs or text from the interface is given in green.

Alternative 5c) Run AMoRe Molecular Replacement Program

Running AMoRe is not as simple as running MolRep; it is a good idea to read the documentation to understand what is happening. Click on the Help button at the top of the AMoRe window to open the documentation.

AMoRe will not find all the solutions in one run of the program. We must run the program three times to find the three molecules.

Exercise

  1. From the Molecular Replacement module select AMoRe.

  2. There are two windows. In the Amore Model Database window enter the file name of the model coordinate file. For Trial model coords select the directory DATA and the file model.pdb. The model is given the name model. You can now Close this window.

  3. In the AMoRe window, enter a job title such as:

    Job title AMoRe run1 (mr tutorial step 1)

  4. The default mode is Auto-Amore which will run through all of the AMoRe functions to find the solutions for one molecule. The input trial model has automatically been set to model.

  5. In the Files folder enter the name of the experimental data file:

    MTZ in DATA cardiotoxin.mtz

  6. In the folder Key Parameters, select and enter:

    Resolution range from minimum 50.0 to maximum 3.0

  7. In the folder Sort and Tabling Parameters, the second line, put:

    add -1.8 to input Bfactors

  8. Click on Run -> Run Now.

  9. When finished, look at the output files with View Files from Job. You will see that there are three output files called ...rot_model.mr (the rotation solution), ...tran_model.mr (the translation solution) and ...fit_model.mr (the fitting solution).

    Hints

    The "..." in the output file names above will be something like TEST_3_ (so you will have TEST_3_fit_model.mr, for example).

    The naming convention is for the first word to be the Project alias (in this case TEST), and the number to correspond to the number of the job in the main window (in this it would have been job number 3).

    In the steps below, each run of AMoRe will generate new rot, tran and fit files, and at each stage you will need to use solutions from the last (i.e. most recent) run of AMoRe. So to distinguish output from each run and get the correct file, check the number in the file name.

    Look at the 'fit mr' file (that is the one called ...fit_model.mr). This file contains the best solutions for finding one molecule - the rotation angles are in columns 3 to 5 and the translations in columns 6 to 8. The 'quality' of the solution is indicated by the R-factors and correlation coefficients in columns 9 to 11. You see there are several solutions with very similar correlation coefficients (column 9).

    #CCP4I VERSION CCP4Interface 1.3.13
    #CCP4I SCRIPT MR fit model
    #CCP4I DATE 19 Sep 2003  10:45:24
    #CCP4I USER mgwt
    #CCP4I PROJECT TEST
    #CCP4I JOB_ID 3
    #CCP4I SYMMETRY C2
    #CCP4I CELL 78.7000 40.4000 56.0000 90.0000 117.1000 90.0000
    #CCP4I RESOLUTION 3.0 50.0
    SOLUTIONF1_1    1   58.11   75.86  145.70  0.3917  0.0000  0.2774 35.3 55.6 21.4   1
    SOLUTIONF2_1    1   58.06   75.94  146.08  0.3916  0.0000  0.2775 35.3 55.6 21.5   2
    SOLUTIONF3_1    1   58.30   75.67  146.26  0.3916  0.0000  0.2772 35.2 55.6 21.5   3
    SOLUTIONF4_1    1   58.14   75.88  146.15  0.3916  0.0000  0.2773 35.3 55.6 21.5   4
    SOLUTIONF5_1    1  162.94   67.56  149.34  0.0863  0.0000  0.2886 34.7 55.9 24.5   5
    SOLUTIONF6_1    1  163.21   68.31  149.33  0.0861  0.0000  0.2891 34.7 56.0 24.4   6

    Click the Quit button.

In order to find the second molecule, run just a translation function, with known solutions for the first molecule.

  1. Go back to the AMoRe window and change from running auto-AMoRe to running just a translation function to find the second molecule in the asymmetric unit. Adapt the job title to reflect the change.

    Run translation function

    To run the translation function we need to know the rotation solutions - so:

    Rotation solution file TEST ...rot_model.mr

    Click on the View button for this file. You will see the file contains many solutions - to be quicker we will use just five. If we put a # at the beginning of a line in this file then the solution will not be used. Click on the button Change All - there will be # at the beginning of every line. Now click on the first five lines - the # will be removed from these lines. Make sure to select five distinctly different rotation function solutions. From the following example solutions, note which ones should be selected:

    SOLUTIONRCD   1  293.73   67.53  149.69  0.0000  0.0000  0.0000 11.9 61.1  3.7 20.7   1
    	SOLUTIONRCD   1   56.20   72.99  147.08  0.0000  0.0000  0.0000 11.3 61.3  4.6 14.4   5
    	#SOLUTIONRCD   1   56.67   73.30  148.13  0.0000  0.0000  0.0000 11.2 61.3  4.5  4.9   6
    	SOLUTIONRCD   1  108.10   81.12  343.87  0.0000  0.0000  0.0000 11.2 61.3  4.9 12.7   7
    	#SOLUTIONRCD   1   57.60   73.08  146.18  0.0000  0.0000  0.0000 11.1 61.3  4.5 11.2   8
    	#SOLUTIONRCD   1   59.61   73.25  146.92  0.0000  0.0000  0.0000 11.0 61.2  4.4 15.0   9
    	SOLUTIONRCD   1  162.46   68.98  148.85  0.0000  0.0000  0.0000 10.6 61.5  4.2  6.8  10
    	#SOLUTIONRCD   1  160.59   68.66  148.70  0.0000  0.0000  0.0000 10.5 61.5  4.1 16.9  11
    	#SOLUTIONRCD   1  162.14   67.21  149.35  0.0000  0.0000  0.0000 10.5 61.5  4.2 15.2  12
    	SOLUTIONRCD   1   33.62   66.42  151.90  0.0000  0.0000  0.0000 10.3 61.8  4.2  5.0  14

    Now click on Save and Exit.

  2. We want AMoRe to look for another solution assuming that we already have one solution. We must tell AMoRe where to find the 'known' solutions. Click on the button Add known solutions - this makes an extra line in the window, select:

    Solution file TEST ...fit_model.mr

    This file contains many solutions and we want to try just the best five. Click on the View button for this file. Click on the button Change All - there will be # at the beginning of every line. Now click on the first five lines - the # will be removed from these lines. Again, make sure to use five distinctly different solutions, for example:

    SOLUTIONF1_1    1   58.11   75.86  145.70  0.3917  0.0000  0.2774 35.3 55.6 21.4   1
    	#SOLUTIONF2_1    1   58.06   75.94  146.08  0.3916  0.0000  0.2775 35.3 55.6 21.5   2
    	#SOLUTIONF3_1    1   58.30   75.67  146.26  0.3916  0.0000  0.2772 35.2 55.6 21.5   3
    	#SOLUTIONF4_1    1   58.14   75.88  146.15  0.3916  0.0000  0.2773 35.3 55.6 21.5   4
    	SOLUTIONF5_1    1  162.94   67.56  149.34  0.0863  0.0000  0.2886 34.7 55.9 24.5   5
    	#SOLUTIONF6_1    1  163.21   68.31  149.33  0.0861  0.0000  0.2891 34.7 56.0 24.4   6
    	SOLUTIONF7_1    1  294.50   69.27  150.39  0.1836  0.0000  0.3921 33.9 56.4 22.5   7
    	SOLUTIONF8_1    1  110.92   83.51  343.13  0.4164  0.0000  0.1736 33.5 56.9 25.7   8
    	SOLUTIONF9_1    1  162.05   68.98  148.53  0.1092  0.0000  0.3383 32.6 57.3 22.3   9

    Now click on Save and Exit.

  3. Click on Run -> Run Now.

  4. The output from this job is then put into another run of AMoRe. Go back to the AMoRe window and change the name of the input solution file to the output file from the last job:

    Solution file TEST ...tran_model.mr

  5. Look at this solution file - click on the View button. The start of this file should look like this:

    SOLUTIONTF1_7   1   58.15   75.99  145.81  0.3902  0.0000  0.2768 34.0 54.0  0.0   1   23.0
    SOLUTIONTF2_7   1  293.59   69.46  149.57  0.3272  0.4298  0.2711 35.9 51.8 32.4   1   27.6
    SOLUTIONTF1_17  1   58.22   75.88  145.60  0.3901  0.0000  0.2768 34.1 54.0  0.0   2   23.0
    SOLUTIONTF2_17  1  293.59   69.46  149.57 -0.1728 -0.0702  0.2711 35.9 51.8 32.5   2   27.6
    SOLUTIONTF1_6   1   58.15   75.99  145.81  0.3902  0.0000  0.2768 34.0 54.0  0.0   1   23.0
    SOLUTIONTF2_6   1  293.67   67.46  149.71  0.3277  0.4283  0.2711 35.6 52.1 32.4   1   27.6

    The first two lines are one solution for the position of two molecules. The correlation score for the solution is the sum of the score on both lines (here it is 34.0 + 35.9). To save time we do not want to try every solution in this file so edit the file like before. Click on the button Change All - there will be # at the beginning of every line. Now click on the first ten lines - the # will be removed from these lines. Now click on Save and Exit.

  6. Click on Run -> Run Now.

  7. When the job has finished there is an output file ..tran_model.mr with groups of three lines which are three positions for three molecules. It is a good idea to refine these solutions using the AMoRe fitting function. So at the top of the AMoRe window:

    run refine fitting

    and select the last tran mr file:

    Translation solution file TEST ...tran_model.mr

    Again it is a good idea to use only a few of the solutions in this file to save time. Click on the View button for this file. Click on the button Change All - there will be # at the beginning of every line. Now click on the first 12 lines - the # will be removed from these lines. Now click on Save and Exit.

  8. Click on Run -> Run Now.

    Unlike the output from MolRep, which automatically checks the packing and generates sets of coordinates in the correct orientation, AMoRe only outputs sets of rotations and translations. A separate task is provided to apply the transformations and check the packing of the potential solutions.

  9. Select the Build AMoRe Output Model task from the Molecular Replacement module.

  10. Use the refined fitting solutions from the last AMoRe run as input:

    Fitting solution file TEST ...fit_model.mr

    The window will then redraw. From the fitting file it knows that there are three solutions, one for each expected molecule in the asymmetric unit, and that the coordinates are generated in each case by applying the rotations and translations from AMoRe to the original search model.

    It is also necessary to supply an MTZ file which will give the task the correct cell parameters. Expand the "Cell parameters for coordinate file" folder, select "Extract symmetry and cell info from MTZ file", then select the MTZ file:

    MTZ in DATA cardiotoxin.mtz

    Finally, enter a name for the output file which will contain the transformed coordinates:

    Output coords TEST cardio_model.pdb

  11. Click on Run -> Run Now. The output consists of a number of pbd files with the transformed coordinates - the different 'molecules' in the structure are identified by different chain names: A,B,C and so on.

    The task also runs the DISTANG program to check for problems with packing - so you should examine the log file to see the output of this step.

    The next step is to Refine the Structure.


Valid CSS! Valid XHTML 1.0!