QUANTA's X-Ray crystallography applications include a function for searching a specialized database to locate a suitable set of fragments for modeling undefined regions of a protein.
Coordinates for undefined residues can be defined using the fragment search process. To search for fragments, you use a database for a set of proteins that matches certain specifications. Matched fragments are retrieved, displayed in the viewing area, and listed in the textport. When you accept a fragment, its conformation is copied to the undefined region of your structure.
The fragment database consists of two parts: a distance matrix file and a library directory of complete structures. The distance matrix file contains all the inter-Ca distances for a representative set of protein structures. The library directory contains MSFs with complete sets of atomic coordinates for the structures used in the distance matrix file.
The standard QUANTA installation contains a database of twenty structures derived from the Brookhaven PDB. By default., $HYD_DMF points to the file $QNT_ROOT/dmatrix/dmfile containing the inter-Ca distances for the protein structures. $HYD_LIB points to the library directory $QNT_ROOT/msflib.
If you use the fragment searching facility frequently, you can create a more complete set of structures to be searched. For information on this process, see Creating a Fragment Database
A fragment is chosen using the following criteria:
To run a fragment search, you first must specify the anchor residues on either side of the region to be filled by the fragment. Anchor residues are specified by using one of the tools for picking residues on the Search Fragment Database palette to choose the anchor residues in the modeling window or the Sequence table. When you have completed residue picking, you initiate the search by selecting Search Database from the Search Fragment Database palette.
Fragment searching also can use the bump checking tool, Bump. This tool takes retrieved fragments and fits them over the search template. The inter-atomic distances between the main chain, and the beta carbon atoms of the fragment and neighboring residues are calculated. The fragments with close contacts are rejected. Because this procedure reduces the number of fragments finally selected, the initial database search retrieves extra fragments by relaxing search criteria.
When a search is complete, the structures of matched fragments are read from MSFs and displayed superposed over the template residues. Each fragment is color coded and a color-coded legend is displayed in the lower-right corner of the screen. The legend gives the name of the protein from which each fragment is taken, its distance fit, and the RMS difference in Ca atom position when the fragment is superposed on the template.
Accepting a fragment defines coordinates for unknown residues and redefines coordinates of residues in the anchor regions. In some cases, it is advisable to select adjacent residues having structural homology. These residues act as an anchor, holding the fragment in place.
The following figure illustrates the setup and results of a fragment search:
This section describes the tools on the Search Fragment Database palette.
To access the Search Fragment Database palette, select Fragment Database ... from the X-Structure palette.
Lists the proteins in the currently active Ca distance matrix file in the textport.
Selects template residues by picking the first and last residue in a range. This picking process is additive. Previously selected ranges are not cleared even if a search has been performed. Use either Undo Last or Undo All to clear selections.
Selects template residues by picking each individual residue. This picking process is additive. Previously selected atoms are not cleared even if a search has been performed. Use either Undo Last or Undo All to clear selections:
Deletes the last Ca anchor selection.
Deletes all Ca anchor selections.
Searches the fragment database by Ca-Ca distance for matches to the currently selected anchor residues.
When this tool is active, any database search is followed by bumps checking before the optimal retrieved fragments are displayed. This selection is active by default when Search Database is active.
Displays all the retrieved fragments.
Displays the next fragment on the fragment list and removes all others from the viewing area. The initial fragment to be displayed is the first on the fragment list.
Displays the previous fragment on the list and removes all others from the viewing area.
Opens the Display selected fragments dialog box with all fragments listed. Use the dialog box to select one or more fragments for display.
When only one fragment is displayed, this tool lists names and IDs for the residues in the fragment, and for corresponding residues in the active molecule.
Copies the coordinates of a fragment onto the corresponding residues of the active molecule including the original anchor residues. This tool is grayed unless only one fragment is displayed
Averages the coordinates of the fragment with the corresponding residues of the active molecule. This tool is inactive unless only one fragment is displayed. The tool should not be used if the target sequence has undefined atoms.
Clears all fragments from the display
Opens the Fragment Modeling Options dialog box as illustrated:
From this dialog box, you can specify the number of fragments to be displayed after a search, indicate how fragments are displayed and labeled, change the fragment file specifying the database matrix, and specify the number of residues for annealing.
Exits the Search Fragment Database palette and returns to the X-Structure palette.
This section is an exercise taken from part of the tutorial in the Protein Homology Modeling Tutorial. If you have this book, use it to complete the full exercise.
To do this exercise, you must have the file renin_regularize.msf in your current working directory. This file contains a partially built model of human renin with undefined loops. If you do not have the file, check with your system administrator. The file is also available from the Accelrys Scientific Support group.
The distance matrices (dmfile) for this example are located in the file $HYD_DMF.
To begin this exercise, you make selections from the QUANTA Draw menu to display and focus on only those residues of interest in the renin file (renin_regularize.msf). The undefined regions that are appropriate for fragment searching are residues 49 through 62, residues 103 through 112, and residues 117 through 127. Each of these regions spans a reasonable distance and has residues that can be used as anchors.
1. Open renin_regularize.msf to display the renin structure.
2. From the QUANTA Draw menu, select Display Atoms to open the pull-right menu. From the pull-right menu, choose Selection Tools. The Display Atoms and Display Utilities palettes are displayed.
3. From the Display Atoms palette, choose Type in a selection. The Enter Display Selection Commands dialog box appears.
4. In the text entry fields, enter the text strings:
zone 49 to 62
zone 103 to 112
zone 117 to 127
5. Click Done. Each of the areas of interest are displayed in color 2 (red) in the structure.
6. From the Display Utilities palette, choose Save Selection-Commands. A File Librarian is displayed. In the text entry field enter: Display_frag_area_renin, then click Save.
7. Select Finish from the Display Atoms palette to exit and save the changes.
8. From the Draw menu, select Color Atoms to display the pull-right menu. From the pull-right menu, choose Selection Tools. The Color Atoms and Color Schemes and Utilities palettes are displayed.
9. From the Color Atoms palette, select All Atoms and type in a Selection. The Color Selection Commands dialog box appears.
10. In the text entry fields, enter the text strings
zone 49 to 62 = col 2
zone 103 to 112 = col 3
zone 117 to 127 = col 4
11. Click Done. Each of the areas of interest is colored and displayed uniquely on the structure.
12. From the Color Schemes and Utilities palette, select Save Selection-Commands. A File Librarian is displayed. In the text entry field, enter Col_frag_area_renin. Click Save.
13. Choose Finish from the Color Atoms palette to exit and save the changes.
14. Display the Draw menu. Choose Label Atoms to display the pull-right menu.
15. From the Label Atoms pull-right menu, choose Residue ID and Selection Tools. All labels for the structure are visible, and the Label Atoms palette is displayed.
16. From the Label Atoms palette, select Include, Alpha-Carbon Atoms, and Finish. The Ca atoms of the residues in the structure are labeled by their residue IDs. The changes are saved and the utility is exited.
17. From the Edit menu, choose Bond options. The Specify Bonding Mode dialog box is displayed:
18. Set the options to the following:
Bonding Algorithm: Intra-residue + Named Link Atoms
First link atom: C
Second link atom: N
Local Search
Disable Inter-Segment Bonding
Minimum Distance Criterion: 0.05
Apply to what? Both
19. Click OK. These options result in a display of side chains as they are constructed and prevents spurious bonds from being drawn.
20. From the Applications menu, select X-Structure. The X-Structure palette is displayed.
21. From the Structure palette, select Fragment database... The Search Fragment Database palette is opened.
22. From the Search Fragment Database palette, select Options... The Fragment Modeling Options dialog box is displayed.
23. Set the options in the dialog box to the following values:
Number of Fragments: 5
Fragment Drawing: Main Chain
Fragment Labelling: CA Labels
Anneal over Residues: 1
24. Click OK. The dialog box is removed from the display.
25. From the Search Fragment Database palette, choose Pick Alpha Carbon Range.
26. Rotate the molecule so that the first group of atoms in residues 49 through 62 is visible. Select residues 50 (K) and 51 (C), to mark the beginning of the fragment placement area.
27. Select residues 61 (H) and 62 (K) to mark the end of the fragment placement area. The selected residues are indicated by red crosses on the Ca atoms.
28. From the Search Fragment Database palette, select with Bumps, then choose Search Database.
29. When the search is completed, several best-fit fragments are displayed as illustrated:
30. Rms values for the fragments are listed both in the textport and in the modeling window legend. The fit (dist) column shows the least squares fit between the Ca atoms picked in the current model and the matching Ca atoms for each superposed fragment. The rms (lsq) column shows the least squares fit of the whole peptide backbone of the same selected residues.
31. From the Search Fragment Database palette, choose Display Next and Select Display... to begin the process of studying fragments. The Display selected fragments dialog box is displayed as illustrated, listing all the fragments found in the search:
32. Select fragments 2 and 4 for viewing. Both fragments appear to fit well, but 2 (3app) appears to join better.
33. This fragment is a good candidate for filling the sequence gap because it has one of the lowest values for the least squares fit, it joins well with the surrounding residues, and it has fewer close contacts than other fragments.
34. Again choose Select Display. Select fragment 2, 3app. Only this fragment is displayed superposed on the structure.
35. From the Search Fragment Database palette, choose List Residues. Residues are listed in the textport for both the fragment and the selected Ca range.
36. From the Search Fragment Database palette, choose Accept Fragment. The Copy Fragment dialog box is displayed.
This dialog box lists the fragment and range of coordinates to be copied to renin_regularize.msf.
37. Click Yes. The coordinates from fragment 3app, starting at residue 194, replace this region in renin_regularize.msf.
38. From the Search Fragment Database palette, select Pick Alpha Carbon Range. Select residues 103 (T) and 104 (V), to mark the beginning of the fragment placement area. Then select residues 110 (E) and 111 (V) to mark the end of the fragment placement. The four selected residues are indicated by red crosses on the Ca atoms.
39. Repeat steps 27 through 34 for this residue area.
40. Repeat the searching process for residues 117 through 127.
41. From the Search Fragment Database palette, select Finish. The palette is replaced with the Edit Protein palette, Protein Utilities palette, and the Protein dials emulator.
42. To regularize the fragment regions, choose Select Active Range from the Protein Utilities palette. The Pick Range palette is displayed. Select the residue 50 (K) to mark the beginning and residue 60 (Y) to mark the end. Red bars indicate the selected range.
43. From the Edit Protein palette, select Regularize.... The specified region is regularized using previously selected parameters for CHARMm.
44. Accept the regularization and repeat the process for the other two residue regions (106 through 108 and 117 through 127). Observe energy levels and relaxing of the regions.
45. From the Edit Protein palette, choose Save to MSF.... The Save Options dialog box is displayed.
46. Choose the option Save to a new filename then click OK. A File Librarian is displayed.
47. Enter the filename renin_fragment and then click Save. A new MSF is created and the structure is displayed. The name renin_fragment.msf replaces renin_regularize.msf in the Sequence table.
You have completed this example.
