SSD

The SSD extension of Chimera provides a user interface to the Structure Superposition Database (SSD). The database contains structural superposition data from a large number of pairwise structure comparisons using the program MinRMS. Each pairwise comparison yields a series of superpositions ranging from few to many pairs of alpha-carbons matched. For each number of alpha-carbons paired, the superposition with the lowest root-mean-square deviation (RMSD) is kept.

The SSD extension is not distributed with Chimera, but it is freely available from the Structure Superposition Database web site. Using the SSD extension, users can view all or a subset of the superpositions for a given pair of structures, view multiple structure superpositions (when pairwise comparisons have been done for a given structure versus multiple other structures), access RMSD values and other quantities describing the superpositions, and view pairwise and multiple sequence alignments derived from the structural superpositions.

Beginning users of Chimera should consult the tips.

SSD should be listed in the Chimera menu under Tools... Structure Comparison. If SSD is not listed:

1. Make sure that the module has been downloaded and unzipped; this will make a directory (folder) named ssd Folder that contains the directories ssd and AlignPlot.
2. Make sure that Chimera can find the module. This can be done in either of two ways:
   • Move or copy these two directories (ssd and AlignPlot) into the share directory of your Chimera installation.
   • Add the directory containing these two directories to the list of places where Chimera should look:
     1. in Chimera, choose Favorites... Add to Favorites/Toolbar.
     2. Add the location. For example, add /Users/meng/Desktop/ssd Folder if it contains the ssd and AlignPlot directories.
     3. Save this setting so that SSD will automatically appear in the menu in later uses of Chimera.

• Browse superfamilies
• Search structures
• Search superpositions
• Search multiple structures

The buttons Back and Forward can be used to move among previous browser contents. Mainchain shows the mainchain atoms only of any displayed structures. Close closes the SSD Browser, and Help opens this manual page in a web browser.

At least two structures must be chosen before superpositions can be viewed. Even when several structures are chosen, there is another step (choosing superpositions) to indicate which of the following will be examined:

• all or a subset of the superpositions between two structures
• a single superposition of multiple structures

The Browse superfamilies button Go brings up a list of structures in the SSD. It may take a few seconds for the entire list to appear. Structures are listed by 4-character PDB identifier and arranged by superfamily as defined in SCOP. Clicking a PDB identifier displays the structure in the graphics window and information on the structure in the SSD Browser. The Back button can be used to return to the list. Activating a checkbox next to an individual PDB identifier chooses that structure; activating a checkbox next to a superfamily name chooses all structures in the superfamily. Clear all deactivates all checkboxes. After at least two structures have been chosen, pressing View superpositions brings up the dialog for choosing superpositions.

individual structure information

The Search structures section allows searching by a single PDB identifier. If the structure is not in the SSD, a message that no proteins were found will appear. When the PDB identifier (uppercase or lowercase) of a structure in the SSD is entered and Search is pressed, the structure is displayed in the graphics window and information on the structure is shown in the SSD Browser. The individual structure information includes protein name and source species. To the right of the PDB identifier is a checkbox that toggles model display and a menu for changing the model color. Pressing Select superposition partner(s) brings up a list of structures in the SSD. After at least one other structure has been chosen, pressing View superpositions brings up the dialog for choosing superpositions.

The Search superpositions section provides the most direct route to viewing pairwise superpositions. Entering the PDB identifiers of two structures in the SSD that have been superimposed with each other and pressing Search bypasses the additional step for choosing superpositions and goes straight to the pairwise superposition view. If the SSD does not contain superpositions between the two structures, a message that no superpositions were found will appear.

The Search multiple structures section allows a list of PDB identifiers to be entered. After Search is pressed, PDB identifiers not found in the database are reported; if none of the structures are in the SSD, an additional error message will appear. If some of the structures are in the SSD, the dialog for choosing superpositions will appear.

CHOOSING SUPERPOSITIONS

After structures have been chosen, it is usually still necessary to choose superpositions; this is done within the N vs N superpositions dialog.

To choose all or a subset of the superpositions between two structures (a pairwise superposition), click on the white or gray button in the row and column corresponding to the desired pair of structures. These buttons are in the lower part of the N vs N superpositions dialog. None of the other settings in the dialog affect what is obtained by clicking in this checkerboard area. The Display Options section of the SSD Browser controls whether all or a subset of the pairwise superpositions will be shown. To view only a subset based on user-specified ranges for RMSD and percent alpha-carbons matched, specify the ranges before clicking the white or gray button.

To choose a single superposition of multiple structures (a multiple superposition):

1. specify which structure should be treated as the template; the set of superpositions of each additional structure with the template will be examined
2. specify a criterion for picking a single superposition from each set of superpositions:
   • % a-C matched - for example, if 75% and the shorter protein has 400 alpha-carbons, indicates the superposition with 300 alpha-carbons matched
     - OR -
   • RMSD - for example, if 2.0 angstroms, indicates the superposition with the set of alpha-carbons matched at 2.0 angstroms RMSD

When superpositions between two structures have been chosen, the two structures are displayed in the Chimera window and a graph of the number of alpha-carbons matched versus RMSD is shown in the SSD Browser. Above the graph are the two structures' PDB identifiers, checkboxes that toggle model display, and menus for changing model color.

The graph depicts all (or a specified subset) of the superpositions calculated for a pair of structures. The number of paired alpha-carbons, or matches, is plotted versus RMSD (green) and the longest distance between paired alpha-carbons (blue). Clicking with the mouse within this graph determines which superposition is displayed in the graphics window. In the figure, the graph has been clicked to choose the superposition with 239 matches, an RMSD value of 3.56 angstroms, and a longest distance between matched alpha-carbons of 7.30 angstroms. The superposition is marked by a vertical red line on the graph.

Zoom In expands the X axis so that the graph shows a smaller range of match values in the same amount of space; Zoom Out reverses this. When the graph is zoomed in, the entire range of match values can still be viewed using the slider above the graph.

The Display Options section of the SSD Browser only pertains to pairwise superpositions. One can:

• view all superpositions - view the entire set of superpositions calculated for the pair of structures
  - OR -
• view superpositions in range - view the subset of superpositions calculated for the pair of structures that fall within the specified ranges of both:
  • RMSD - RMSD of the sets of matched alpha-carbons
  • % a-C matched - percent of alpha-carbons matched relative to the number of residues in the shorter protein

The Color by options are:

• structure - make each structure a single color, as shown by its color setting
• matched/unmatched - color the matched parts of a structure according to its color setting, and color the unmatched parts gray

Clicking the Transform Matrix button next to the graph opens a small window showing the transformation matrix used for the current superposition. The window title is of the form

pdb1 vs pdb2 - N residues matched

Clicking Sequence Align opens a structure-based sequence alignment corresponding to the current superposition, using the Chimera extension Multalign Viewer. In the structure-based sequence alignment, only the residues matched in the superposition are aligned in the sequence; unmatched residues are treated as insertions (gaps in the other sequence). For details on the sequence display and how it can be customized, see the Multalign Viewer documentation. Briefly, the name of each sequence is shown in bold over a rectangle matching the model color. This indicates the sequence is associated with the structure shown in the Chimera window.

When a superposition of multiple structures has been chosen, the superpositions (plural since a multiple superposition is actually a combination of pairwise superpositions specified by some criterion) are displayed in the Chimera window and information about them is shown in the SSD Browser: The information includes PDB identifiers, checkboxes that toggle model display, and menus for changing model color. N is the number of alpha-carbon pairs, or matches, in the superposition, RMSD is the root-mean-square deviation of the sets of alpha-carbons matched, and Long Dist is the longest distance between matched alpha-carbons. Ger-Lev is the Levitt-Gerstein probability score, described in:

"A unified statistical framework for sequence comparison and structure comparison," M. Levitt and M. Gerstein, Proc. Natl. Acad. Sci. USA, 95,5913 (1998).

Clicking one of the buttons to the left of a PDB identifier opens the pairwise superpositions of that structure with the template structure.

Clicking Sequence Align opens a structure-based sequence alignment corresponding to the current superpositions, using the Chimera extension Multalign Viewer. In the structure-based sequence alignment, only the residues matched in the superpositions are aligned in the sequence; unmatched residues are treated as insertions (gaps in the other sequences). The name of each sequence is shown in bold over a rectangle matching the model color. This indicates the sequence is associated with the structure shown in the Chimera window.

For new users, here are a few basic tips on moving structures, showing the command line, and displaying ribbons in Chimera.

The Chimera User's Guide is a comprehensive manual in HTML format (viewed with a web browser) that comes with your Chimera download. Along with manual pages for each tool and command, it contains tutorials for beginning and intermediate users. Help... User's Guide in the Chimera menu opens the copy that came with your local installation. This copy is the most appropriate to use since it will match your installed Chimera software.