|Python-based Hierarchical ENvironment for Integrated Xtallography|
Tutorial 7: Fitting a flexible ligand into a difference electron density map
This tutorial will start with experimental data and a model (1J4R.pdb) with ligand removed (1J4R_no_ligand.pdb) and a randomized ligand conformation (1J4R_random.pdb), and will fit the ligand into difference density calculated from the experimental data and partial model. The tutorial is designed to be read all the way through, giving pointers for you along the way. Once you have read it all and run the example data and looked at the output files, you will be in a good position to run your own data through LigandFit.
Setting up to run PHENIX
echo $PHENIXthen you should get back something like this:
/xtal//phenix-1.3If instead you get:
PHENIX: undefined variablethen you need to set up your PHENIX environment. See the PHENIX installation page for details of how to do this. If you are using the C-shell environment (csh) then all you will need to do is add one line to your .cshrc (or equivalent) file that looks like this:
source /xtal/phenix-1.3/phenix_env(except that the path in this statement will be where your PHENIX is installed). Then the next time you log in $PHENIX will be defined.
Running the demo 1J4R-ligand data with LigandFit
mkdir tutorials cd tutorialsNow type the phenix command:
phenix.run_example --helpto list the available examples. Choosing 1J4R-ligand for this tutorial, you can now use the phenix command:
phenix.run_example 1J4R-ligandto build the 1J4R-ligand structure with LigandFit. This command will copy the directory $PHENIX/examples/1J4R-ligand to your current directory (tutorials) and call it tutorials/1J4R-ligand/ . Then it will run LigandFit using the command file run.sh that is present in this tutorials/1J4R-ligand/ directory. This command file run.sh is simple. It says:
#!/bin/sh echo "Running LigandFit on 1J4R data..." phenix.ligandfit data=1J4R.mtz model=1J4R_no_ligand.pdb ligand=1J4R_random.pdbThe first line (#!/bin/sh) tells the system to interpret the remainder of the text in the file using the sh (or bash) -shell (sh). The command phenix.ligandfit runs the command-line version of LigandFit (see Automated Structure Solution using LigandFit for all the details about LigandFit including a full list of keywords). The arguments on the command line tell LigandFit about the data file (data=1J4R.mtz), the model without ligand (model=1J4R_no_ligand.pdb), and the ligand (ligand=1J4R_random.pdb). (Note that each of these is specified with an = sign, and that there are no spaces around the = sign.) The structure factor amplitudes are in the datafile 1J4R.mtz. This is an mtz file which is a binary file that contains summary information about the dataset as well as the reflection data. Although the phenix.run_example 1J4R-ligand command has just run LigandFit from a script (run.sh), you can run LigandFit yourself from the command line with the same phenix.ligandfit data= ... command. You can also run LigandFit from a GUI, or by putting commands in another type of script file. All these possibilities are described in Using the PHENIX Wizards.
Where are my files?
Once you have started LigandFit or another Wizard, an output directory will be created in your current (working) directory. The first time you run LigandFit in this directory, this output directory will be called LigandFit_run_1_ (or LigandFit_run_1_/, where the slash at the end just indicates that this is a directory). All of the output from run 1 of LigandFit will be in this directory. If you run LigandFit again, a new subdirectory called LigandFit_run_2_ will be created. Inside the directory LigandFit_run_1_ there will be one or more temporary directories such as TEMP0 created while the Wizard is running. The files in this temporary directory may be useful sometimes in figuring out what the Wizard is doing (or not doing!). By default these directories are emptied when the Wizard finishes (but you can keep their contents with the command clean_up=False if you want.)
What parameters did I use?
Once the LigandFit wizard has started (when run from the command line), a parameters file called ligandfit.eff will be created in your output directory (e.g., LigandFit_run_1_/ligandfit.eff). This parameters file has a header that says what command you used to run LigandFit, and it contains all the starting values of all parameters for this run (including the defaults for all the parameters that you did not set). The ligandfit.eff file is good for more than just looking at the values of parameters, though. If you copy this file to a new one (for example ligandfit_lores.eff) and edit it to change the values of some of the parameters (resolution=3.0) then you can re-run LigandFit with the new values of your parameters like this:
phenix.ligandfit ligandfit_lores.effThis command will do everything just the same as in your first run but use only the data to 3.0 A.
Reading the log files for your LigandFit run file
While the LigandFit wizard is running, there are several places you can look to see what is going on. The most important one is the overall log file for the LigandFit run. This log file is located in:
LigandFit_run_1_/LigandFit_run_1_1.logfor run 1 of LigandFit. (The second 1 in this log file name will be incremented if you stop this run in the middle and restart it with a command like phenix.ligandfit run=1). The LigandFit_run_1_1.log file is a running summary of what the LigandFit Wizard is doing. Here are a few of the key sections of the log files produced for the 1J4R-ligand SAD dataset.
Summary of the command-line arguments
------------------------------------------------------------ Starting LigandFit with the command: phenix.ligandfit data=1J4R.mtz model=1J4R_no_ligand.pdb \ ligand=1J4R_random.pdbThis is just a repeat of how you ran LigandFit; you can copy it and paste it into the command line to repeat this run.
Fitting the ligand into difference density with RESOLVE ligand fitting
The LigandFit Wizard will use the partial model 1J4R_no_ligand.pdb to calculate FC and PHIC (model amplitudes and phases ) for the structure factors in 1J4R.mtz. Then these are used to calculate a simple (FO-FC) exp(iPHIC) difference map which is written to resolve_map.mtz. Then the RESOLVE algorithm for ligand fitting is used to fit your ligand 1J4R_random.pdb into this map. To make the fitting process suitable for parallelization, the fitting is done in a series of tries, each considering more possible rotations of the ligand and each carrying out a more exhaustive search than the previous one. Then depending on how many processors are available, they are run one at a time or in parallel. Here is the summary of the five attempts at fitting the 1J4R difference map:
Fitting ligand #1 from /net/cci-filer1/vol1/tmp/terwill/phenix_examples/1J4R-ligand/1J4R_random.pdb Ligand file has 45 atomsSearching for ligand number 1 Estimated ligand volume: 185.0 A**3 Starting CC of ligand as input to map: 0.31 Starting local CC of ligand as input to map: 0.18USING LOCAL CC: 0.18 Setting background=False as nproc=1 Try: 1 quickTrying fit with n_indiv_tries_min = 5 n_indiv_tries_max = 10 n_group_search = 3 ligand_file = 1J4R_random.pdb Try: 2 thorough Trying fit with n_indiv_tries_min = 100 n_indiv_tries_max = 100 n_group_search = 4 ligand_file = 1J4R_random.pdb Try: 3 randomize Randomizing ligand conformation before starting fit Randomized version of /net/cci-filer1/vol1/tmp/terwill/phenix_examples/1J4R-liga nd/1J4R_random.pdb placed in random.pdb Trying fit with n_indiv_tries_min = 100 n_indiv_tries_max = 100 n_group_search = 4 ligand_file = random.pdb Try: 4 extra_thorough Trying fit with n_indiv_tries_min = 300 n_indiv_tries_max = 300 n_group_search = 6 ligand_file = 1J4R_random.pdb Try: 5 extra_thorough_randomize Randomizing ligand conformation before starting fit Randomized version of /net/cci-filer1/vol1/tmp/terwill/phenix_examples/1J4R-ligand/1J4R_random.pdb placed in random.pdb Trying fit with n_indiv_tries_min = 300 n_indiv_tries_max = 300 n_group_search = 6 ligand_file = random.pdb Running 5 jobs on 1 processors Splitting work into 5 jobs and running 1 at a time with sh in /net/cci-filer1/vol1/tmp/terwill/phenix_examples/1J4R-ligand/LigandFit_run_1_/TEMP0 Starting job 1... Starting job 2... Starting job 3... Starting job 4... Starting job 5... Solution for try : 1 SCORE= 123.868 CC= 0.650 LIGANDS= 1 LIG= ligand_fit_1_1.pdb TEMPLATE=1J4R_random.pdb Solution for try : 2 SCORE= 128.528 CC= 0.700 LIGANDS= 1 LIG= ligand_fit_1_1.pdb TEMPLATE=1J4R_random.pdb Solution for try : 3 SCORE= 128.170 CC= 0.710 LIGANDS= 1 LIG= ligand_fit_1_1.pdb TEMPLATE=random.pdb Solution for try : 4 SCORE= 129.750 CC= 0.720 LIGANDS= 1 LIG= ligand_fit_1_1.pdb TEMPLATE=1J4R_random.pdb Solution for try : 5 SCORE= 129.574 CC= 0.720 LIGANDS= 1 LIG= ligand_fit_1_1.pdb TEMPLATE=random.pdbIn this case the best fits were found on tries 4 and 5 (the ones with the most thorough fitting attempts). The log file for the best fit is in ligand_fit_1_1.pdb (the first 1 refers to the solution number, and the second 1 refers to the ligand number, as you have the option of finding more than one ligand in the map with number_of_ligands=n). This log file will describe the analysis of the ligand and the rigid parts of the ligand, the search for locations of rigid parts of the ligand, and the extension of those rigid parts into density to complete the ligand. In these runs the parameters n_indiv_tries_min=5 and n_indiv_tries_max=10 are the minimum and maximum number of placements of each rigid part of the ligand to test for completing the ligand. The parameter n_group_search=3 is the number of different rigid parts of the ligand to try. On subsequent tries at fitting these will all be increased to search more thoroughly. This fitting of the ligand was able to place all 45 atoms in the ligand, with a correlation to the map in the region of the ligand of 0.72. The "LOCAL CC" is the correlation of the model density with map density in the region occupied by the ligand itself, plus any contiguous points with high difference electron density that are connected to the region occupied by the ligand. This local CC is more useful at discriminating between correct and incorrect models than the standard correlation (which only includes the region occupied by the model.) Note that the fit to the map has improved only slightly in this case over the initial fit found on the first try. Sometimes this is the case and sometimes a considerably better fit may be found by searching more thoroughly.
The LigandFit_summary.dat summary file
A quick summary of the results of your LigandFit run is in the LigandFit_summary.dat file in your output directory. This file lists the key files that were produced in your run of LigandFit (all these are in the output directory) and some of the key statistics for the run, including the overall correlation between the model and the map and the number of copies of the ligand placed. Here is the summary for this 1J4R-ligand ligand-fitting run:
LIGAND SOLUTIONS FOR RUN 1 SORTED BY OVERALL CC **** FILES ARE IN THE DIRECTORY: LigandFit_run_1_ **** RANK SCORE CC COPIES FITTED/TOTAL NAME TEMPLATE RMSD CC_START LIG_VOLUME 1 129.750 0.720 1 45 / 45 ligand_fit_1_1.pdb 1J4R_random.pdb 0.00 0.18 185.00
How do I know if ligand fitting worked?
What to do next
Once you have run LigandFit and have obtained a good ligand model, you will want to refine the model. Add the model to your PDB file containing the remainder of the structure and refine the whole structure. You may need to define the geometry of your ligand. In that case the tool phenix.elbow in the PHENIX package well by most useful (it is quite automatic and generates all the files you need). Once you have fitted your ligand(s), the next thing you might want to do is to re-run the AutoBuild Wizard, including your ligands with input_lig_file_list=myligand.pdb. If you do not obtain a good fit of your ligand to the map, there are a few things you should check: