|
X-BUILD tutorial |
(30 minutes)
This tutorial details the use of X-BUILD for model building protein structure into electron density. The tutorial works through the list of tools on the main model building palettes, and gives you an idea of the building functionality. The text editor text markers are used to indicate a residue for the action, but after working through the tutorial any part of the RNAse structure (1, 2) can be modeled.
The following files are provided:
(1) Sevcik, J.; Dodson, E.; Dodson, G. G.; Zelinka, J. "The X-ray analysis of ribonuclease Sa.", Metabolism of Nucleic Acids, including Gene Manipulation. Slovak Academy of Science, Bratislava., 33-46 (1987).
(2) Sevcik, J.; Dodson, E.; Dodson, G. G."Determination and restrained
least-squares refinement of the crystal structures of ribonuclease Sa and
its complex with 3'guanylic acid at 1.8 Å resolution", Acta
Cryst, B47, 240-253 (1991).
To complete this tutorial you need a licensed copy of QUANTA that includes this module:
This example requires the file xbuild.tar.Z. <Shift>-Click it to copy it to your area. Then click OK in the next window that comes up. Make a note of the directory into which the file is going to be saved.
Now cd to that directory. Uncompress and untar the file by typing:
Now type:
And start QUANTA:
When QUANTA has finished loading, select Applications / X-BUILD. This starts the application so you can do model building for a partially refined insulin model.
As in the first tutorial, you see the X-AUTOFIT:X-BUILD palette, the Pointer palette, the Ramachandran plot, Object Management Table, and this time a Build atoms palette. You should make sure you can manipulate the view in the molecule window using the virtual trackball option of QUANTA before you continue.
Make sure that the Build Atoms palette is visible. If not, select the Build atoms... tool from the main X-AUTOFIT:X-BUILD palette. The Build Atoms palette appears as the front window.
This tutorial presents the tools and their use. If you have enough time, you can then try using the different tools to carry out your own modeling of the RNAse structure. For each tool, you will be asked to use the 3D text editor tool Goto defined text to go to a specific part of the structure (labelled with a text string) to try the tool. To complete the tutorial as designed, please follow the instructions.
Select Draw / Show from the Map Table pullright. Next select Map / Open a map in the Map Management Table. Choose the map called original.mbk and select Open. It is recommended that the maps be contoured at 1 and 2 sigma (which is 0.211 and 0.422).
Open the coordinates rnase.msf using File / Open. Select the file called rnase.msf. The RNAse appears in the main model window.
Select Applications / X-BUILD to open X-BUILD. The main X-BUILD:X-AUTOFIT palette opens, along with the Pointer palette, an Object Management Table, a Ramachandran Plot, and the display of some of the map. You will also need the Text palette and Building palette. Open these by picking the tools X-BUILD/Text... and X-BUILD/Build atoms....
This tool fits single residues/atoms using gradient refinement. You will use this tool to refine a sulfate ion into some density as it is approximatly 0.5 Å from the center of the density.
Go to the Refine 1 residue text string using the 3D text tool Goto defined text.
Pick the tool Refine 1 residue from the Build atoms palette. The message line at the bottom of the graphics window asks you to Pick a residue/water.
Move the mouse arrow to the sulfate ion (red cross), and click with the left mouse button.
An Accept:Quit palette appears and the sulfate ion begins to refine. The refinement is graphics-limited and will proceed much faster without the map turned on. The message line indicates the progress of the refinement and indicates when the refinement has completed. (If we wanted to stop the refinement, clicking with the left mouse button would do so.) When the message line says Finished, you can either Accept or Quit the solution using the Accept:Quit palette.
Goto the {Geometric conformation} text string using the 3D text tool Goto defined text.
This tool is used to place side chains of amino acids in conformations normally found in proteins when there is no experimental information to define the conformation experimentally. In this case, there is an asparagine residue at this position.
A) Pick the tool /Build atoms/Geometric conformation. The message line at the bottom of the graphics window asks you to Pick atom to set geometry. Pick any atom in the asparagine residue.
B) A Geometry pallete appears on the right-hand side of the screen, containing 6 rotamer options, all trans conformation, and an Accept:Quit pair. You cannot pick any other tool while this palette is active.
The first option (Rota 1 -> 30.3%) is highlighted, and this conformation appears in white on the graphics view of the molecule. Pick different rotamers from the Geometry palette and look at the different conformations, their energy and bumps.
Go to the {Fit side chain by RSR} text string using the 3D text tool Goto defined text.
There you will find a phenylalanine residue in the core of the protein which does not fit the density, in fact it is a very long way from the correct position and too far for gradient-type refinement to work. To fit this residue automatically into the density, pick the tool Fit side chain by RSR. The message line at the bottom of the graphics window indicates that you should pick an atom at which to begin the refinement. This means that the program will ONLY REFINE FROM THE ATOM SELECTED, so you MUST pick the CA atom of this residue to fit the whole side chain.
The phenylalanine immediately jumps into the electron density. The grid refinement is in fact so fast that as you move the CA atom of this residue, the program actively refines the coordinates to completion about 5 times/second (R4000 machine).
Note: Make sure you remember how to use the virtual track ball option (you can practice moving the pointer first) before doing the following:
Pick the tool Move atom + RSR (the 5th tool on the Build atoms palette). The program prompts you pick an atom to move, and RSR simultaneously. Pick the CA of this phenylalanine again.
The Accept:Quit palette appears, a white current solution appears, and the message line at the bottom of the graphics window indicates the current fit to density.
Move the CA atom using the virtual trackball XY motion and watch the side chain trying to stick in the density. Eventually, if you more the CA atom far enough, the side chain will flip to some other density.
The application only applies local non-bonding, so you should also watch the energy value and the non-bond indicators during the refinement. It is interesting to note that although it is usually obvious when you have found the right answer, it is difficult to get to that answer. This tool allows you to move a CA atom (for example), and try different positions, as this atom defines the direction of the side chain.
When you have finished moving the atom, pick either the Accept or Quit option from the Accept:Quit palette, and the white solution disappears. If you choose Accept, the side chain is updated.
Go to the {Edit peptide plane} text string using the 3D text tool Goto defined text.
There are 3 ways to edit peptide planes:
At the labelled peptide bond, you will use all three ofthe above methods. Pick the tool Flip torsion 180 deg. (tool 11 on the Build atoms palette). You are prompted to pick a rotatable bond, in this case the peptide bond marked. (Note that this is not strictly a rotatable bond because this selection results in the rotation of the peptide plane, and not the bond). The peptide plane flips 180 degrees. To change this back again, use the tool Fit main chain by RSR (tool 4). You are prompted to pick a peptide bond. (If you do not pick a peptide bond, the program ignores the tool.) The application fits the atoms into the density by refining the peptide plane AND the omega angle (within limits). The atoms should fit the density again.
To edit the peptide plane and omega manually, you can use the tool Edit backbone tor. This allows you to change the peptide plane with the virtual trackball, but the omega angle can ONLY be changed with the dial box. This is so that the omega angle is not moved by mistake. If you look at the Ramachandran plot you find the two phi-psi pairs of torsion that are changed by this tool displayed, and as you more the peptide plane, the dots move. The message line also shows the actual values of phi and psi + omega.
Go to the {Regularize} text string using the 3D text tool Goto defined text.
You find a glutamic acid residue at this position in the structure, with very little density for the side chain. This residue has rather poor geometry as it has not refined very well. In particular, the COO group is non-planar. To correct the geometry of this residue, select the tool Regularize (tool 16) from the Build atoms palette. The program prompts for the first residue in the regularize zone and the last residue in the regularize zone. Pick any atom in the glutamate residue twice to indicate that this is the only residue to be regularized. The residue regularizes to the expected geometry. The regularization proceeds much faster when the map is turned off, as the display refresh is much faster.
You can now try the interactive editing while the regularizer is active. This time pick the tool Move atoms + reg. res. and select one of the atoms, OE1 or OE2, from this residue. The Accept:Quit palette appears and the message line at the bottom of the QUANTA graphics window shows Residual = 0.0011 (your residual may be slightly different). A green atom representation of the residue appears in the graphics window. Now use the virtual track ball option to move this picked atom. When moving a single residue with the regularizer active, the refresh of the coordinates is limited by the graphics of the machine. You can now drag the side chain around by this atom, where the N and C atoms of the residue are anchored at their initial positions as shown by the crosses. The bonds of the residue change color, depending on the deviation of the geometry at each atom at the ends of this bond.
To change the moving atom, pick a new atom with the left mouse button. The moving atom is marked with a cross. You can free the fixed ends of the residue (O and N) atoms by picking these as moving atoms.
On completion of the editing, select the Quit option from the Accept:Quit palette.
To try out Move atom + reg. zone, you need to be able to view GLU 74, ASP 79 (the residue with the text label {Regularize}), and residue GLN 77. Use the left mouse button to pick and identify the residues. (Although it does not matter on which residue you actually try this, these residues are used for the tutorial.) Select the tool Move atom + reg. zone. You are prompted for:
You need not pick the first/last residues in order, and you can pick any atom of a residue to specify the range to regularized.
Initially the zone regularizes to completion. The message line notes that a new moving atom can be picked at any time (with the left mouse button), and a value for the current residual. A green display of the regularized atoms also appears.
If a residue in the zone of regularization contains a cystine residue as part of a disulfide bond, a text ort comment, Disulphide link added, appears. This is because the last residue in the zone was a cystine residue, so the link + the other cystine residue is added as part of the regularized residues.
You can now move the currrent moving atom with the virtual trackball (or dials), and change the moving atom. Experiment with changing the CA positions, and then either Accept or Quit the editing.
Go to the {Alternate conformation} text string using the 3D text tool Goto defined text.
A GLU residue is at this text marker with two side chain conformations. You can view which conformation is the B-conformer using a coloring option: from the Build atoms palette select the Color atoms... tool. This opens a new palette, Atom color. There are 6 tools on this palette. Pick Color B-alt different. The B-conformer of this residue turns pink.
The coordinates shown for this residue are for the final structure, but since this map is the original map, there is little density to suggest the possibility of a dual conformation.
First, we must delete the B-conformation. Find the tool X-BUILD-X-AUTOFIT/Build atoms/Add-delete/Delete residue. When you pick this tool the application prompts you for a residue to delete. Pick the pink B-conformation. The B-conformation disappears. Now, to place the B-conformation back again, use the tool Add alternate conformation on the same palette. The B-conformation appears, with CA and CB atoms in the same position and the side chain different from the CG atom. This is because the default mode of action for alternate conformation is to place the main chain atoms in the same place, and allow only side chain atoms to have different positions, where CB is the branch point. To change this action you should change the toggle option on the X-BUILD/X-AUTOFIT/Options... dialog box B-conf clamped to A backbone so that it is turned off. Now you can move B-conformation atoms so that the whole can be an alternate conformation, and therefore separate the CB atoms.
You might like to model the alternate conformation residues with various tools to see their actions. Generally, all tools can be used on A/B conformations but there are some specific notes on editing alternate conformation; You should refer to the documentation on this to explain various effects. One typical problem is that fitting the B conformation using the Fit side chain by RSR tool will find the global minima, which is likely to be the A-conformation. If the atoms end up very close, the B-conformation will be merged out of existance. Hence it would be normal to fit the A-conformation to the global minima (so that it has occupancies > = 0.5) and then the B-conformation with the single residue refinement or the tool Move atom + reg res.
Go to the {Add VAL then ASP here} text string using the 3D text tool Goto defined text.
You should find that the display has centered on a region of the map where the density continues without any atoms. You will place a valine residue and an aspartic acid residue in this density.
You should still have the Add/delete palette open; if not, open this from the Build atoms palette using the Add/delete... tool. Select the tool Add res at terminal and select the nitrogen atom at the 3D text marker. (If you do not pick the N-terminal residue, the addition aborts.) A palette appears so you can selecta residue. Select a valine, which then appears within the density. Select this tool again and place a apartic acid. Both residues have been placed into the electron density close to a sensible conformation. This is because the application fits the terminal residues using a tree search algorithm for density fitting to the last/previous residue, so that any new atoms are fitted to the electron dnesity in the best possible conformation. Hence, adding a residue to the termini is extremely easy.
You may want to try to improve the fit to the density using other tools after completing the rest of this set of tutorials.
By this time you should have five palettes open. Close the Color atoms and Add/delete palettes by picking the tool Hide this menu on both palettes (which closes them).
Goto the {Loop fit 1} text string using the 3D text tool Goto defined text.
You can experiment with the the Structure palette for modeling zones of residues. Use the tool Loop fit to fit the loop CA 73 to CA 77 of segment A. Then use the Refine zone tool to improve the loop fitting solution. This loop has been artificially moved from its correct conformation too far from the density; gradient refinement cannot fit the loop to the map correctly. You should use the tool X-BUILD:X-AUTOFIT/Structure.../Loop fit and pick the two labelled atoms ({Loop fit 1} to {Loop fit 2}). The program searches for loop conformations until you click with the left mouse button to abort the search, or the time limit expires (see the Options... dialog bog for the time limit. Default = 10 minutes). You can then view the best 10 solutions (defined by the fit to electron density), accept/quit a solution, or continue the search. If the loop search finds a better solution, use the Refine zone tool to impove the fit by gradient refinement.
To validate the stucture, select the Protein validation tool on the X-Autofit:X-Build palette, and then work through the 3D text list of errors using the Goto next text tool to look
