• Download and Installation
• Data Flow and Design
• Converting the EM Maps
• Computing the Difference Density Maps
• Inspecting Data Cross Sections
• Extracting Regions of Interest
• Inspecting the Voxel Histogram
• Extracting Single-Molecule Densities
• Vector Quantization and Docking
• Visualization
• Part II: Advanced Matching with volhist and matchpt

First, follow these registration and download steps (each Situs tutorial is separate and must be downloaded and compiled individually)!

Then, return to this page.

• 0_ncd.pdb: Atomic coordinates of a ncd monomer (as described in Wriggers et al., 1999).
• 0_mt.ascii: ASCII-formatted density map of the undecorated microtubule (from the Milligan lab data site).
• 0_nm.ascii: ASCII-formatted density map of the ncd monomer decorated microtubule (from the Milligan lab data site).
• 0_nd.ascii: ASCII-formatted density map of the ncd dimer decorated microtubule (from the Milligan lab data site).
• run_tutorial.bash: Bash shell script containing all commands of this tutorial.

In the following, we will use these four files to dock ncd to the microtubule, following the steps described in Wriggers et al., 1999. The user can compare all generated files to the files in the solutions directory.

Schematic diagram of EM related routines. Major Situs components (blue) are classified by their functionality. The main data flow is indicated by brown arrows. The visualization (orange) for the rendering of the data requires a molecular graphics viewer (we use here the VMD graphics program, Chimera and Sculptor also support Situs files).

Standard EM formats are supported and are converted to cubic lattices in Situs format. This is done with the map2map utility. Subsequently, the data is inspected and, if necessary, prepared for the docking using a variety of visualization and analysis tools. Situs docking with matchpt requires one volume and one PDB structure for the fitting. Atomic coordinates in PDB format can be transformed to low-resolution maps, if necessary, and vice versa, to allow docking of maps to maps or structures to structures. Here we use the classic approach of vector quantization for generating intermediate coarse grained models of the data. After the vector quantization, which can be generated by external programs quanvol and quanpdb or which is automatically generated, the high-resolution structure is docked to the low-resolution density using matchpt. The resulting docked complex can be inspected using the external graphics program.

We convert EM maps to Situs format with the map2map utility. Enter at the shell prompt:

Enter the file format option 1: ASCII. Enter the axis order option: 1: X Y Z (order is already correct). Now enter the number of columns: 89, the number of rows: 89, and the number of sections: 61. Enter the grid (lattice) spacing in Ångstrom: 6. Finally, enter the origin of the map: 0 0 0.

Repeat this procedure similarly for the other two EM maps to create the files 1_nm.situs and 1_nd.situs (see run_tutorial.bash).

Note that the origin of the coordinate system is set to the first voxel in the map.

Similarly, to compute the difference density map of the second, detached monomer, subtract the monomer-decorated microtubule data from the dimer-decorated microtubule data:

Now we inspect the generated Situs maps with the voledit utility. voledit requires a fixed-width font in the shell window.

Here is an example voledit session for file 1_nd.situs:

Similarly, (see the run_tutorial.bash script) we create the files 3_nd.situs, 3_nm.situs, 3_s1.situs, and 3_s2.situs from the corresponding files 1_nd.situs, 1_nm.situs, 2_s1.situs, and 2_s2.situs.

The above inspection shows that the background peak is already at the origin, so there is nothing to do here. However, if necessary volhist could be run with additional file names as arguments, in which case the program would provide options for rescaling or shifting the densities, or for matching the density histogram of two maps.  The more advanced functionality is described in part II of this tutorial.

Now do the same for the second, detached monomer (see the run_tutorial.bash script). Similarly, one finds that the voxel at position x=19, y=12, z=13 is located near the center of the density of the detached monomer in file 3_s2.situs at a threshold level of 12. The threshold cutoff values were found by trial and error to prevent "leaking" of the density into neighboring monomers. Here we also shrink the maps about the found extracted volumes.

Now we can perform the vector quantization and fast docking of the attached monomer with the matchpt utility. At the shell prompt, enter

The program returns a list of solutions for (N, M=7), ranked by their codebook vector rms deviation. The correlation coefficient is also shown. The fourth column in the output below shows the internal permutation of the vectors that determine the superposition. Note that the correlation coefficient is within a very narrow numeric range, despite the significant differences in orientation between these fits, whereas the vector rmsd is more discriminatory.

Now do the same for the second, detached monomer. Is there a unique best fit? What is the lowest rms deviation and the gap to the subsequent fits? Extract the best fit to the file 5_s2_1.dock.pdb and the second best to 5_s2_2.dock.pdb. See also the run_tutorial.bash script for details on automating these steps.

Note: As described in (Birmanns & Wriggers, 2007), the fast matching by codebook vectors can be refined by a post-processing with the correlation-based tool collage. We have explained this refinement step in the separate multi-fragment docking tutorial.

The following sequence of commands in the VMD text console (cf. VMD user guide) will load the two docked structures, 5_s1_1.dock.pdb and 5_s2_1.dock.pdb, and render them in cartoon representation, coded by color. The script then instructs VMD to render the density maps 3_mt.situs in green, 4_s1.situs in blue, and 4_s2.situs in purple:

Don't forget to hit "enter" after the last line! 

The result should look like this:

The matchpt tool supports the registration of point clouds even in cases where they are of unequal size, such as in oligomeric assemblies. Also, as mentioned above, volhist provides options for rescaling or shifting the densities, or for matching the density histogram of two maps. The more advanced usage of these routines is demonstrated on the above examples in part II of this tutorial.