Cavities and Pockets

Analysis > Cavities and Pockets... finds the cavities of the structure shown, lists them with their volume, area, depth and lining residues, displays them in the 3D view and follows the volume of a cavity along the frames of an object (MD, scans, NEB, umbrella sampling).

The method is the two-probe grid method of KVFinder (Oliveira et al., 2014; Guerra et al., 2020):

• Probe In (default 1.4 Å, water): the space this probe can reach is cavity space.

• Probe Out (default 4.0 Å): a large probe rolled from outside the molecule defines the bulk solvent; the region it cannot reach is the "hull" of the molecule.

• Cavity points are reachable by Probe In but not by Probe Out, and farther than the Removal distance (default 2.4 Å) from the bulk, so shallow grooves of the surface are not counted. Connected groups of points form the cavities; groups smaller than the Minimum volume (default 5 Å3) are dropped. The Grid step (default 0.6 Å) sets the resolution.

Options

• System / Object / Frame: the structure analyzed.

• Leave out: water, ions (single-atom residues, including metal ions of an active site), ligands (residues that are neither protein nor water: leave them out to see the empty binding pocket) and hydrogens (crystal structures have none).

• Region: the whole molecule, or a box around the selected atoms (for example a ligand or the active-site residues) with a margin; along frames the box follows those atoms.

Click Detect Cavities (Figure 1). The table lists, for each cavity, its volume (Å3), area (Å2), maximum depth (Å), whether it is buried (no mouth to the solvent), and the number and names of the lining residues. Each cavity has its own switch to show it in the 3D view, as a surface or as grid points, with adjustable opacity, in a "Cavities" object under the system (Figure 2). Double-click a cavity to center the view on it; Select Lining Residues selects its residues; Export... writes the table (CSV) and the cavity points (PDB, residue CAV).

Volume along frames

Select a cavity in the table and click Follow Selected Cavity to compute its volume in every frame of the object, with the same parameters. The pocket is recognised in each frame by the atoms that line it.

References: S. H. P. Oliveira et al., BMC Bioinformatics 15, 197 (2014); J. V. S. Guerra et al., SoftwareX 12, 100606 (2020).