User Guide

Version 2.0 beta

HCI Laboratory, Faculty of Informatics, Masaryk University

Loschmidt Laboratories, Faculty of Science, Masaryk University
Department of Computer Science and Engineering, Faculty of Applied
Sciences, University of West Bohemia
The software package is provided "as is" without warranty of any kind. In no event shall the author or his
employer be held responsible for any damage resulting from the use of this software. The program pack-
age, including source codes, example data sets, executables, and this documentation, is distributed free of
charge for academic use only. Permission is granted to copy and use programs in the package provided no
fee is charged for it and provided that this copyright notice is not removed.

Recommended Citation:

Chovancová, E., Pavelka, A., Beneš, P., Strnad, O., Brezovský, J., Kozlíková, B., Gora, A., Šustr, V., Klvaňa, M.,
Medek, P., Biedermannová, L., Sochor, J. Damborský, J. (2012). CAVER 3.0: A Tool for the Analysis of
Transport Pathways in Dynamic Protein Structures. PLoS Computational Biology, 8: e1002708.

Contacts:

Human Computer Interaction Laboratory

Contact person: Barbora Kozlikova ([email protected])
Faculty of Informatics, Masaryk University
Botanická 68a,
602 00 Brno,
Czech Republic
Webpage: http://decibel.fi.muni.cz
Phone: +420 549 496 939, Fax: +420 549 491 820

Loschmidt Laboratories, Department of Experimental Biology

Contact person: Jiri Damborsky ([email protected])
Research Centre for Toxic Compounds in the Environment
Faculty of Science, Masaryk University
Kamenice 5, Bld. A13, 625 00 Brno, Czech Republic
Webpage: http://loschmidt.chemi.muni.cz
Phone: +420 549 493 467, Fax: +420 549 492 556

Department of Computer Science and Engineering

Contact person: Martin Manak ([email protected])
University of West Bohemia
Univerzitni 8, 306 14 Plzen, Czech Republic
Webpage: http://www.kiv.zcu.cz/
Phone: +420-377-63-2401, Fax: +420-377-63-2402

Project webpage: www.caver.cz

E-mail: [email protected]
© Copyright 2005-2018
HCI Laboratory, Faculty of Informatics, Masaryk University, Brno, Czech Republic
Loschmidt Laboratories, Institute of Experimental Biology, Faculty of Science, Masaryk University, Brno,
Czech Republic
Department of Computer Science and Engineering, Faculty of Applied Sciences, University of West Bohe-
mia, Pilsen, Czech Republic
Contents
1 Introduction …………………………………………………………………………………………………………………………………………… 1
1.1 About CAVER Analyst ………………………………………………………………………………………………………………….. 1
1.2 Hardware and Software Requirements ………………………………………………………………………………………. 1
1.3 User Support ………………………………………………………………………………………………………………………………. 2

2 Software Features …………………………………………………………………………………………………………………………………. 3

2.1 Workspace…...……………………………………………………………………………………………………………………………… 3
2.2 Main Menu ....................................................................................................................................... 4
2.3 Top Toolbar …………………………………………………………………………………………………………………………………. 9

3 Windows and their Features .................................................................................................................... 12

3.1 Structures Overview Window ……………………………………………………………………………………………………… 12
3.2 Visualization Window …………………………………………………………………………………………………………………… 16
3.2.1 Manipulation with the Scene ………………………………………………………………………………………….. 16
3.3 Tunnel Computation Window ………………………………………………………………………………………………………. 17
3.4 Tunnel Advanced Settings Window ………………………………………………………………………………………………. 21
3.5 Tunnel Statistics Window ……………………………………………………………………………………………………………… 21
3.6 Cavity Computation Window ………………………………………………………………………………………………………… 27
3.7 Cavities Overview Window …………………………………………………………………………………………………………… 29
3.8 Locked Probes Window …………………………………………………………………………………………………………………. 32
3.9 Structure Sequence Window ………………………………………………………………………………………………………… 33
3.10 Structure Selections Window ………………………………………………………………………………………………………. 33
3.11 Structure Dynamics Window ………………………………………………………………………………………………………… 35
3.12 Structure Properties Window ……………………………………………………………………………………………………….. 35
3.13 Structure Statistics Window ………………………………………………………………………………………………………….. 36
3.14 Tunnel Graph Window ………………………………………………………………………………………………………………….. 36
3.14.1 Static case ……………………………………………………………………………………………………………………….. 37
3.14.2 Dynamic case ………………………………………………………………………………………………………………….. 38
3.15 Coloring Window ………………………………………………………………………………………………………………………….. 40
3.16 Alignment Window ……………………………………………………………………………………………………………………….. 41
3.17 Structure Clip Planes Window ….………………………………………………………………………………………………….. 42
3.18 Surface Configuration Window …………………………………………………………………………………………………….. 43
3.19 Search in Structure ……………………………………………………………………………………………………………………….. 43
3.20 Delete from Structure …………………………………………………………………………………………………………………… 44
3.21 Measurement Window …………………………………………………………………………………………………………………. 44
3.22 Mutagenesis Window ……………………………………………………………………………………………………………………. 45
3.23 Asymmetric Tunnels Window ……………………………………………………………………………………………………….. 46
3.24 Residue Graph Window ………………………………………………………………………………………………………………… 47
3.25 Contours Window …………………………………………………………………………………………………………………………. 49
3.26 Console Window …………………………………………………………………………………………………………………………… 50
1 Introduction
1.1 About CAVER Analyst
CAVER Analyst 2.0 is a software program for the analysis of access pathways to buried active sites in
proteins in long molecular dynamics (MD) simulations. Many proteins possess deeply buried binding sites in
the protein core. The size, shape, physico-chemical properties and dynamics of pathways leading from the
protein surface to the core are of the utmost importance for the accessibility and biological activity.
Accessibility of the pathways for the ligands is as important as the ligand’s fit to the binding site itself.
CAVER Analyst 2.0 assists with calculation, analysis, visualization and rational re-design of the ligand access
pathways in MD.

CAVER Analyst 2.0 offers a number of features assisting analysis of the access pathways in protein struc-
tures:

 Visualization of all identified access pathways.

 Frequency and average overall cost of individual pathways.
 Average radius, length, volume and curvature of a given pathway.
 Maximum and average bottleneck of a given pathway.
 Residues lining a given pathway and bottleneck.
 Detailed pathway profiles.
 Pathway dynamics by animation of long MD simulations.
 Specialized visualization techniques for exploration of tunnel profile, tunnel bottleneck, and
surrounding residues over time.
 Measurement of distances and angles between atoms.
 Mutation of individual residues, using rotamer libraries.
 Multiple clip planes for exploration of protein inner structure.

1.2 Hardware and Software Requirements

Minimal configuration:
 Java 1.8
 2 GB RAM
 32-bit architecture
 Intel HD family graphics adapter

Recommended configuration:
 Java 1.8 or later
 8 to 16 GB RAM
 64-bit architecture
 AMD Radeon or NVIDIA GeForce family graphics card

1
Regardless of the recommended configuration, CAVER Analyst is platform-independent; therefore the
following platforms are supported:
 Windows 7, 8, 10
 Mac OS X 10.7.5 or later
 Linux major distributions including Fedora Core, Red Hat and Ubuntu

1.3 User Support

Almost 6,000 users world-wide have downloaded some version of CAVER (almost 80,000 individual
downloads) and many of them have provided invaluable feedback. Users’ comments and recommendations
have always been instrumental for the development of the software thus we are grateful for any kind of
feedback from your experience with the CAVER Analyst.

To obtain more information, provide recommendation, request new features or report bugs in CAVER
Analyst, please write to us at [email protected].

2
2 Software Features
2.1 Workspace
The workspace of the CAVER Analyst application consists of the main menu, toolbar and four main panels
(see figure 2.1). The top part of the left panel (control panel) is used to control loaded structures
(Structures Overview window), to manage existing selections (Structure Selections window) and to manage
computed cavities (Cavities Overview window). The bottom of the left panel is reserved for working with
various tools, e.g., structural and sequence alignment and controlling probe size and transparency of
protein and tunnel surface.

The main menu and toolbar provide access to all functions of the application. The toolbar has a shortcut to
the most often used functions, such as loading the molecule from a disc, downloading it online from the
PDB database, saving and loading the current session, changing the visualization style of active structures,
tunnels, cavities, and others. Full descriptions will be provided later in this document.
The bottom panel (explorer panel) allows users to explore primary protein structure and make residue
selections (Structure Sequence window), to control the animation playback when a dynamic simulation is
loaded (Structure Dynamics window) and to explore various tunnel statistics – lining residues, bottlenecks,
etc. (Tunnel Computation Statistics window). It contains also the Console window for controlling the main
functions of the application using the command line. It also offers a new visualization feature for
exploration of tunnel profile over time (Residue Graph window).
The right panel (computation panel) allows users to choose input parameters for tunnel computation and
to launch their calculation (Tunnel Computation window). Furthermore, it contains the Cavity Computation
window for setting input parameters and calculation of cavities. This area also contains the Color window
for changing color schemes for individual atoms, residues or chains for visualization purposes. When you
want to see more information about loaded structures, you can use the Structure Properties window,
which will also be displayed in the right panel. More detailed information about given structure is present
in the Structure Statistics window. This panel also contains the Mutagenesis window for designing
mutations of residues on static structures.
Finally, the central visualization panel serves for visualization and view control (Visualization window,
visualizing and setting of graph statistics of computed tunnels (Tunnel Graph window), for displaying log
messages (Log window), and for novel representation of tunnel cross cut and its evolution over time
(Contours window).

IMPORTANT NOTE
Windows in one panel can be switched by clicking their title header. Each panel can be closed or reopened
via the Main Menu. Their size can be deliberately altered by dragging. Users can also design their own
layout by dragging the windows to the desired panel (drag the title header). They can return to the default
layout and size in the main menu: View/Reset Windows. Windows can also be undocked and separated
from the workspace.

3
 Figure 2.1: CAVER Analyst work area

2.2 Main Menu

Top main menu contains the following categories:

 File
o Open Structure(s)…
Allows the user to open a structure from the PDB file stored on the hard drive. CAVER
Analyst creates its own caver folder (in Windows in C:/Documents and
Settings/username/caver). This is where all downloaded PDB files are stored.
o Open Molecular Dynamics…
Allows the user to open a set of PDB files representing molecular dynamics stored on the
hard drive.
o Download Structure…
Opens the dialog window enabling the user to enter the PDB ID of the molecule, which will
then be downloaded from the PDB database.
o
o Convert Dynamics…
Allows the user to convert the molecular dynamics from the AMBER, GROMACS, or
CHARMM file format to the set of PDB files.
o Load Workspace…
Allows the user to load sessions previously saved on the hard drive (in .cws format).
o Save Workspace…
4
 Saves the current CAVER Analyst session (all loaded structures, computed tunnels and their
settings) to the hard drive (.cws format).
o Application Log
Opens the center panel with information about performing actions.
o Application Settings
Opens the dialog window with general settings of the application.
o Console
Opens the Console window where the user can control the main functions of the
application using the command line.
o Exit
Closes the CAVER Analyst application.

 View
o Visualization Window
Opens the center panel containing a visualization of the scene. More details in section 3.2.
o Highlight Selection
Enables/disables the highlighting of atoms, residues or chains (according to the settings in
the Selection window) when passing over them with the cursor.
o Surface Configuration…
Opens the Surface Configuration window in the bottom left panel. It allows the user to
change the transparency of the molecular surface, tunnel surface, and selection surface,
and define the probe size, which controls the preciseness of the surface. It has no effect on
other visualization styles.
o Scene Fog
Enables/disables visualization of fog, enhancing the perception of depth in the scene.
o Scene Drag Rotation
When enabled, users can launch automatic rotation of the whole scene. The direction and
speed of rotation is determined by dragging the mouse.
o Orthographic Projection
Allows the user to change the projection type. Orthographic projection does not distort the
scene.
o Reset Camera
Resets the scene so the loaded structures are centered to the viewport.
o Reset Structures
Resets all the loaded structures to their original position after loading to the application.
o Toolbars
Here users can customize the Top Toolbar section – show/hide groups of related icons,
change icon size or customize the toolbar.
 File – shows/hides icons related to opening or downloading structures.
 Workspace – shows/hides icons related to loading or saving current session.
 Computation – shows/hides icon related to tunnel computation.
 Structure Visualization – shows/hides icons related to changing visualization
modes of structures.
 Tunnel Visualization – shows/hides icons related to changing visualization modes
of tunnels.
 Cavity Visualization – shows/hides icons related to visualization of cavities.
 Tools – shows/hides icons related to additional tools (currently color chooser and
5
 making screenshots).
 Small Toolbar Icons – changes the icon size.
 Reset Toolbars – sets all toolbars to their default settings.
 Customize – add arbitrary icons to the Top Toolbar section.
o Reset Windows
All windows are returned to their default position and size.

 Visualization
Allows the user to change the visualization style of activated protein structures, computed
tunnels, or cavities. If no tunnels or cavities for a given structure have been computed,
changing their visualization style is disabled.
Structure Visualization Styles:
o Points
Visualizes molecule as crosses positioned into the center of each atom.
o Dots
Displays atoms as dotted spheres of van der Waals radii.
o Wireframe
Represents a molecule with its bonds visualized as lines.
o Alpha Trace
Visualizes the polypeptidic chain of a protein by connecting the C-alpha atoms of amino
acids.
o Sticks
Displays molecular bonds as three-dimensional sticks.
o Balls & Sticks
Shows both atoms and bonds of the protein.
o Van der Waals Radii
Molecules are represented by a set of spheres with van der Waals radii.
o Cartoon
Representation of secondary structures.
o Surface
Visualization of the protein surface. Its transparency can be set using the
Visualization/Transparency function available from the main menu.

Tunnel Visualization Styles:

o Points
Displays tunnels using Tunnel Subdivision Surface, where the surface is represented by
dots. This method can also be used for visualizing tunnels in molecular dynamics.
o Centerline
Shows a line representing the centerline of a tunnel. This method can also be used for
visualizing tunnels in molecular dynamics.
o Spheres
Shows the tunnel as a set of intersecting spheres. This method can also be used for
visualizing tunnels in molecular dynamics.

o Detailed Surface
This is the widest and most precise tunnel representation. This method takes more
computation time, thus it is used only in a static case.
6
 o Clusters (all snapshots in one)
This method is available only when computed tunnels in dynamics are available. It displays
the centerlines for all tunnels computed for all snapshots at once. Centerlines are colored
according to their related clusters.
o Tunnel Asymmetric Surface
Shows the asymmetric shape of the tunnel surface.

Cavity Visualization Styles:

o Surface
Shows the surface of detected cavities.
o Locked Probes
Shows the locked probes, i.e., sites, where a probe of given size cannot move, it is locked in
its position. More details in section 3.8.

 Structure
o Overview
Opens the top left panel containing a list of all loaded structures. More details in section
3.1.
o Selections
Opens the top left panel containing a list of all created selections. More details in section
3.10.
o Alignment
Opens the Alignment window in the bottom left panel. Details in section 3.16.
o Search in Structure…
Opens the dialog window where the user can search for specific parts of the protein by
stating their ID. Details will be described in section 3.19.
o Delete from Structure…
Opens the dialog window where the user can delete specific parts of the protein. Details
will be described in section 3.20.
o Clip Planes
Opens the Structure Clip Planes window in the top right panel. Details will be described in
section 3.17.
o Sequence
Opens the bottom panel containing a sequential representation of all loaded proteins.
More details in section 3.9.
o Dynamics
Opens the bottom panel containing control buttons for animating molecular dynamics.
More details in section 3.11.
o Properties
Opens the top right panel, which contains the summarized information about the active
structure. More details in section 3.12.
o Statistics
Opens the top right panel containing information about the constitution of active structure.
More details in section 3.13.
o Measurement
Opens the bottom left panel enabling the user to perform distance and angle
measurements between atoms. More details in section 3.21.
7
 o Mutagenesis
Opens the bottom right panel enabling to design mutations of selected residues for static
structures. More details in section 3.22.
 Tunnel
o Computation
Opens the top right panel, enabling the users to enter parameters for tunnel computation.
More details in section 3.3.
o Asymmetric Tunnels
Opens the top right panel, enabling the users to define parameters for computation of
asymmetric tunnels. More details in section 3.23.
o Residue Graph
Opens the bottom panel enabling the users to explore the tunnel profile over time, along
with the surrounding residues and their properties. More details in section 3.24.
o Contours
Opens the central panel enabling the user to explore the contour shape of a selected cross
section of a tunnel (i.e., its bottleneck) over time, along with the surrounding residues and
their properties. More details in section 3.25.
o Statistics
Opens the bottom panel with detailed statistics about computed tunnels – clusters, lining
residues, bottlenecks, etc. More details in section 3.13.
o Graph
Opens the center panel containing a window which displays various tunnel statistics. More
details in section 3.14.
o Advanced Settings
Opens the top right panel, enabling the user to view and edit all parameters influencing the
CAVER algorithm for tunnel computation. More details in section 3.4.
 Cavity
o Computation
Opens the top right panel, enabling to set the input parameters for cavities computation.
More details in section 3.6.
o Overview
Opens the top left panel containing the list of detected cavities. More details in section 3.7.
o Locked Probes
Opens the top right panel enabling the users to set the parameters of locked probes. More
details in section 3.8.
 Help
o Automatic Updates
Enables the user to turn on and off automatic updates of the application.
o User Guide
Opens this user guide.
o Report Bug
Opens the dialog window where the user can report a bug or desired feature.
o About CAVER Analyst
Information about product version, authors and copyright.

8
2.3 Top Toolbar
The top toolbar contains buttons with shortcuts to the most frequently used functions (see figure 2.3).

Figure 2.3: Toolbar icons

Functions of icons (from left to right):

 Open Structure(s)…
Opens a dialog window for selecting a PDB file from the hard disc.
 Open Molecular Dynamics…
Opens a dialog window for selecting a set of PDB files the from hard disc which represent
molecular dynamics.
 Download Structure…
Allows the user to enter a 4-character PDB ID for a molecule, which will then be
downloaded from the PDB database.
 Convert Dynamics…
Opens a dialog window for converting the molecular dynamics in the AMBER, GROMACS, or
CHARMM format to the sequence of PDBs.
 Search in Structure…
Opens the dialog window for selecting parts of the protein by stating their ID.
 Delete from Structure…
Opens the dialog window for deleting parts of the protein by stating their ID.
 Load Workspace…
Opens a dialog window for selecting a previously stored working session in .cws format.
 Save Workspace…
Opens a dialog window where the user selects the destination for saving the current
working session in .cws format.
 Tunnel Computation
Opens the Tunnel Computation panel.
 Asymmetric Tunnels
Opens the panel for setting the parameters for computation of asymmetric tunnels.
 Residue Graph
Opens the panel containing the visualization of the tunnel profile and its surrounding
residues in molecular dynamics.
 Contours
Opens the panel containing the visualization of the tunnel cross cut contour shape and its
surrounding residues in molecular dynamics.
 Cavity Computation
Opens the Cavity Computation panel.

Changing visualization styles of molecules and tunnels (they influence only the “active” structures and
tunnels – highlighted in the Structure window and more styles can be activated simultaneously):

 Structure Points
Molecules are represented by crosses in the centers of all atoms.
 Structure Dots
9
 Atoms are represented by dotted spheres with van der Waals radii.
 Structure Wireframe
Molecules are rendered by lines representing bonds between atoms.
 Structure Alpha Trace
C-alpha atoms of all amino acids are connected and interpolated – it displays the
polypeptidic chain in a simple manner.
 Structure Sticks
Molecules are represented by 3D sticks showing bonds between atoms.
 Structure Balls & Sticks
The most detailed representation of a molecule. It displays all atoms and bonds.
 Structure Van der Waals Radii
Atoms are represented by spheres of van der Waals radii.
 Structure Cartoon
Molecules are displayed using secondary structures – helices, sheets, coils.
 Structure Surface
Displays the molecular surface of the protein. Surface preciseness can subsequently be
refined – see section 3.18.

 Tunnel Points
Tunnels are visualized using a dotted subdivision surface. It can also be animated.
 Tunnel Centerline
Tunnels are displayed as a line representing their centerline. Also suitable for moving
tunnels in molecular dynamics.
 Tunnel Spheres
Tunnels are represented by intersecting spheres positioned on the tunnel centerline. Also
suitable for visualizing tunnel dynamic changes.
 Tunnel Detailed Surface
The most exact representation of tunnel boundaries in surface form. It is also possible to
use this representation in molecular dynamics. Moreover, the user can change the
preciseness of the tunnel surface by changing the probe size – see section 3.18.
 Tunnel Clusters (all snapshots in one)
This representation is suitable only for tunnels computed in a trajectory. It displays tunnels
as centerlines and visualizes tunnels from all snapshots at once. The centerline color
determines individual clusters detected in the computation phase.
 Tunnel Asymmetric Surface
Tunnels are represented as a surface describing the tunnel void space more precisely,
based on the user-defined parameters.
 Cavity Surface
Enables the visualization of detected cavities.
 Locked Probes
Enables to define the parameters of the locked probes.

Other functions:
 Coloring
Opens the Coloring window for changing colors of different protein parts (see section 3.15).
 Screen Capture

10
 Allows the user to store current snapshot of the visualization window in the .png, .bmp or
.jpg format.
 Video Capture
Opens the panel for capturing the video of the main visualization window.
 Fog
Allows the user to turn fog on and off in the scene. Fog can enhance the perception of
depth in the scene.
 Selection Mode
Enables to define the selection mode is then used for direct selection of protein parts in the
main visualization window.

11
3 Windows and their Features
3.1 Structures Overview Window
This window allows the user to organize all structures currently loaded into the application (see figure
3.1.1). The basic principle of manipulation with structures is as follows. When the user wants to work with a
given structure, the structure must be ACTIVE. This means that it is selected in the Structures Overview
window. A selected structure is highlighted in orange. More structures can be activated at once. Using Ctrl
+ Left Mouse Button, the user can select or deselect any structure. The Shift + Left Mouse Button
combination allows the user to select or deselect more structures between two selected structures (similar
as in other, common applications).
All structures can be selected using the “Activate All” button at the top of the window. The “Invert Active”
button switches between selected and deselected items. The “Remove Selected” and “Remove All” buttons
can remove structures from the application.

Figure 3.1.1: Structures Overview Window example

IMPORTANT NOTE:
WHEN SOME FUNCTIONALITY OF CAVER ANALYST APPEARS TO BE DISABLED, PLEASE VERIFY THAT YOUR
DESIRED STRUCTURE IS ACTIVE!

Detailed Description of Structures Overview Window

In the Structures Overview Window, the loaded structures and tunnels are displayed as a list of individual
rows. Each structure row contains the 4-character PDB ID of the structure and a set of icons for direct
manipulation of the structure. The icons are (from left to right):
 Zoom
Positions the structure to the center of the visualization window.
 Changing Visualization Style
Shows the changing style window very similar to the window which can be activated in the main
menu – Visualization.

12
  Colors
Switches to the Colors window for changing coloring styles.
 Structure Symmetries
If the protein has some defined symmetries in the PDB file, these can be turned on and off using
this button.
 Lock Structure
Allows the user to lock the structure in a given position so that the user cannot rotate and move
with the structure within the scene.
 Show/Hide Structure
Shows/hides the structure in the visualization window.
 Close Structure
Removes the structure from the application.

When clicking on a structure with the right mouse button, the following menu appears:

Figure 3.1.2: Menu for each structure

 Show
Enables to define which parts of the loaded PDB structure should be visible in the visualization
window.
o Amino Acids
o Hydrogens
o Waters
o Ligands
 Show Labels
Shows labels on all atoms in the scene.
 Tunnel Computation
o Compute Tunnels…
Opens the Tunnel Computation window (see details in section 3.3).

o Import Tunnels…
Allows the user to import already computed tunnels from the hard drive in the .pdb
13
 format (stored previously using tunnel export – see below).
o Import Caver 3 Outputs…
Allows the user to load the whole CAVER 3 output directory containing computed
tunnels and all statistics.
 Hydrogen Computation
o Compute Hydrogens
Launches a computation of hydrogen atoms which are assigned to a given structure
and visualized.
o Configure Protonation of Titratable Residues…
Allows the user to change the protonation of titratable residues when computing
hydrogen atoms.
 Align With…
Opens the Alignment window (see details in section 3.16).
 Select
Creates a selection containing all instances of the type below present in the PDB file of given
structure.
o Amino Acids
o Backbone
o Waters
o Ligands
o Donors
o Acceptors
 Export
Allows the user to save the current structure in the given format to an external file.
o Export to PDB…
o Export Trajectory to PDB…
o Export to FASTA…
o Export to OBJ…
o Export to PLY…
o Export to STL…
 Properties…
Opens the Structure Properties window (see details in section 3.12).
 Rename..
Allows the user to rename the structure (the new name appears in the given structure
row).
 Remove
o Structure – removes the structure from the application.
o Waters – allows the user to remove water molecules from the structure. Please note that
this function cannot be undone!
o Hydrogens – removes hydrogens from the given structure. Recommended usage: when
some hydrogens are already present in the PDB file of the structure, first remove them and
then launch Compute Hydrogens.
o All Tunnels – removes all tunnels of a given structure from the application.
o Ligands… – if the input PDB file of a given structure also contains a ligand, it is recognized in
the loading phase and stored separately. This function opens the dialog window with all
ligands present in the structure and the user can select which ones should be removed.

14
All tunnels computed in one pass of the CAVER algorithm are grouped into one tunnel set, marked as
SET#X. Each tunnel (tunnel cluster, to be more specific) has also its own record (see figure 3.1.3). The
tunnel set is assigned to its structure and its name also contains the settings of two major computation
parameters – probe radius and clustering threshold. Moreover, individual tunnels are also named (starting
with tun_cl_# - abbreviation for tunnel cluster). They are sorted – a smaller tunnel number indicates its
greater relevance.

Figure 3.1.3: Example of tunnel set record

Each tunnel set and tunnel record contains the following functions controlled by icons (from left to right):
 Changing Visualization Style
Shows the changing style window where the user can change the visualization style of the tunnel
(same as in Toolbar or Visualization in main menu).
 Tunnel Color
Allows the user to change the color of given tunnel.
 Show Tunnel Graph
Opens the Tunnel Graph window (see section 3.14).
 Show/Hide Tunnel
Shows/hides the tunnel in the visualized scene.
 Close Tunnel
Removes the tunnel from the application.

When clicking with the right mouse button on a specific tunnel, the following menu appears:

3.1.4: Menu for each tunnel record

 Export “tun_cl_#”
Allows the user to store a given tunnel in .pdb format to the hard drive.
 Rename…
Shows the dialog window for renaming a given tunnel.
 Remove
Removes a given tunnel from the application.

15
3.2 Visualization Window
The visualization window displays all loaded structures (if they are not explicitly hidden) and allows the user
to control them via a standard interface. The user can rotate (left mouse button click + mouse drag), scale
(right mouse button click + mouse drag) or translate them (mouse wheel button click + drag), see figure 3.2.

Figure 3.2: Visualization window with gray background color and enabled fog

3.2.1 Manipulation with the Scene

The scene can be controlled using two different approaches:
 Global manipulation – users can manipulate (zoom, rotate, move) the whole scene (with all loaded
structures at once). This can be done by using three mouse buttons.
o Left mouse button – rotation with all structures (active and inactive) around the scene cen-
ter.
o Right mouse button – scales the whole scene along the Z-axis.
o Middle mouse button – translates the whole scene.
 Local manipulation – users can only manipulate active structures (highlighted in the Structures
Overview window). This manipulation is activated by pressing CTRL + mouse buttons.

o CTRL + left mouse button – rotation with active structures around their local center (de-
fined by the bounding box).
o CTRL + right mouse button – scales the ACTIVE structures in the Z-axis.
o CTRL + middle mouse button – translates the ACTIVE structures.

16
3.3 Tunnel Computation Window
The Tunnel Computation window controls the computation of tunnels (see figure 3.3.1) – setting the
binding site position, important parameters, etc. This panel is activated by selecting the desired structure in
the Structure Overview window. The panel has two modes – the basic and the advanced one. The advanced
mode can be activated using the checkbox on the top of the panel.
In the first section, the user specifies the starting point – the initial position of binding (or active) site. The
application first searches the active site in the Catalytic Site Atlas, according to the structure PDB ID code.
When the search fails or if the user decides to select other starting point, the active site can be set
manually (by determining the surrounding atoms and/or residues, or X, Y, Z coordinates). This starting point
can be subsequently visualized in the display window using the “Show Starting Point“ button. After that,
the user chooses the desired number of tunnels they want to compute and items included into
computation (waters are excluded by default) and clicks “Compute Tunnels”.

Figure 3.3.1: The advanced mode of the Tunnel Computation window

17
Proper descriptions of each function of the Tunnel Computation window are as follows:
 Starting point definition
This section allows the user to set the position of the binding site.
o Known binding sites
Shows the list of active sites for a given structure. This list is loaded from the Cata-
lytic Site Atlas (CSA) database. If no records for a given structure are present in CSA,
this section remains empty.
o Surrounding items (atoms or residues)
Displays the list of atoms or amino acids surrounding the active site. Users can ma-
nipulate this list by adding and removing atoms and/or residues. The “Restore
Items” button resets the list to initial values (loaded from CSA or empty). The “Add
Item” button opens the dialog window for setting parameters of a new atom
and/or residue included into the active site definition (see figure 3.3.2).

First, the user can select a structure for which they want to define the binding site.
If this structure contains more chains, they can also choose the one containing the
desired atom/residue. Then they can select atoms or residues for binding site defi-
nition. In both cases the user must know the ID of the atom or residue.
The user can add more atoms or residues at one time by dividing them by comma
(e.g., 123,247,256).
Then the user must check if a given ID is present in the given structure and chain by
pressing the “Validate” button. If it succeeds, the user obtains a list of selected at-
oms/residues along with their name abbreviation.

Figure 3.3.2 Definition of a binding site

o From Selection…
The binding site can be also defined by a selection (see figure 3.3.3). The user can
choose from a list of created selections (present in Structure Selections window)
and specify if the starting point should be defined by atoms or residues of the given
selection. Finally, the user assigns the starting point to one of the loaded struc-
tures.

18
 Figure 3.3.3 Creating the starting point from selection

o To Selection
This option adds the list of surrounding items into the selection. If a selection is ac-
tive, it adds the items to this selection. Otherwise it creates a new selection.
o Absolute position
Users can set the absolute X, Y, Z coordinates of the active site.
o Show Starting Point
Button activating visualization of a cross representing the active site.
o Copy to Another Structure…
A given binding site can be reused for another structure (see figure 3.3.3). This can
perform the structure alignment and stores the starting point for selected struc-
tures as absolute position.

Figure 3.3.3 Conversion of a starting point

 Starting Point Optimization (available only in the advanced mode)

o Maximum distance (Å)
Specifies the maximum distance of the calculation starting point from the initial starting
point (for more details see CAVER 3.0.2 user guide).
o Desired radius (Å)
The closest Voronoi vertex to the initial starting point, which is located within a specified
distance from the initial starting point and at least a desired_radius far from the balls rep-
resenting the input structure, which will be used as the starting point for the calculation of
tunnels (for more details see CAVER 3.0.2 user guide).
 Settings
o Dynamic tunnel computation (available only in the advanced mode)
Allows the user to set the range and sparsity of snapshots for which tunnels should
be computed. Active only for dynamic structures.
 From frame, To frame (available only in the advanced mode)
The number of frames defining the first and last frame for tunnel computa-
tion (for more details see CAVER 3.0.2 user guide).
 Sparsity (available only in the advanced mode)
Defines which snapshots tunnels will be computed for. For example, a
sparsity value of 5 means that tunnels will be computed in each fifth snap-
shot from the trajectory (for more details see CAVER 3.0.2 user guide).

19
 o Approximation
Specifies the number of balls which will be used to approximate individual atoms in
the input structure (for more details see CAVER 3.0.2 user guide). A higher number
means more precise results but also greater memory consumption.
o Min. probe radius
Defines the desired minimal radius of computed tunnels (for more details see CAV-
ER 3.0.2 user guide).
o Clustering threshold
Specifies the level of detail at which the tree hierarchy of tunnel clusters will be
cut, and thus influences the size of resulting clusters (for more details see CAVER
3.0.2 user guide).
o Shell depth
Specifies the maximum depth of a surface region, i.e., a part of the input structure
located below the bulk solvent region (for more details see CAVER 3.0.2 user
guide).
o Shell radius
Specifies the radius of the shell probe which is used to define which parts of the
Voronoi diagram represent the bulk solvent (for more details see CAVER 3.0.2 user
guide).
o Residues included
In this section the user can find all structures which are present in a given structure
(amino acids, ligands, waters, etc.). The user can select which of these structures
should be involved in a tunnel calculation.
This section includes two checkboxes:
 Detailed list
Displays the list of all amino acids present in the structure (along with their
ID).
 Exclude active selections
Excludes selections which are highlighted as active in the Structure Selec-
tions window.
o Output directory
Allows the user to select the destination folder for the results of tunnel computa-
tion.
 Compute Tunnels
Button for launching the tunnel computation.
 Import Tunnels
Enables to load the tunnels computed by the standalone CAVER tool.
 Set of icons for:
o Save computation settings to an external file
o Load computation settings from an external file
o Get computation settings from previously computed results

20
3.4 Tunnel Advanced Settings Window
This window allows the user to edit all parameters which influence the results of tunnel computation (see
figure 3.4). Changing or adding arbitrary parameters is further described in the CAVER 3.0.2 user guide.

Figure 3.4 Tunnel advanced settings panel

After changing a parameter, the “Save” button is enabled. To return to the default settings, the user can
use the “Restore” button. For a better orientation in the parameters, the user can jump to a section
containing the desired parameter using the combo box at the top right corner of the panel.

3.5 Tunnel Statistics Window

The Tunnel Statistics window allows the user to display statistical information about computed tunnels at
different levels of detail.
The first table displays a summary for computed tunnel clusters (see figure 3.5.1).

Figure 3.5.1 Table summarizing tunnel cluster statistics

21
The user can open more tunnel cluster statistics at once and they’ll be displayed as individual tabs. Each tab
is named according to the related tunnel set, tunnel cluster or individual tunnel. Each tab contains a set of
buttons and other controls for manipulation within the table.
 Save
Allows the user to save the table in the .csv format to a desired location on the hard drive.
 Summary (Clusters)
This button allows for navigation in the table. When the user switches to detailed tables
containing individual clusters (second level of detail) or tunnels (third level of detail) (see
below), this button returns to this table summarizing information about clusters.
 Cluster
This button has a similar function as the Summary (Clusters) button. It returns the user
from the third level of detail – information about individual tunnels – to the second level
containing cluster information.
 Detail (Tunnel)
Informative button, always disabled. It informs the user about the level of detail in the sta-
tistics panel.
 Hide inactive tunnels
When this checkbox is checked, the Visualization window shows only active tunnel clusters
(activated by clicking into the table on a corresponding row or activating them in the Struc-
tures Overview window – see section 3.1).
 Synchronize with visualization
Active only for molecular dynamics. When the dynamics is animated (animation can be
launched using the Structure Dynamics window – see section 3.11), it enables to animate
also the statistics table.
 Residues in current snapshot only
In molecular dynamics, it displays only residues relevant to tunnels only in given snapshot.
 Visible columns
Shows the list of all available columns and allows the user to show/hide them using their
checkbox.
 Show all
Displays all available columns, including the hidden ones.
 Invert Visibility
Inverts the visibility of columns (the visible ones will be invisible and vice versa).

Individual columns have the following meaning:

 ID – identification of a given tunnel cluster; ranks a given cluster based on their priority.
 No – total number of tunnels belonging to a given cluster.
 No_snaps – number of snapshots with at least one tunnel with a radius >= parameter
min_probe_radius
 Avg_BR [Å] – average bottleneck radius.
 SD – standard deviation (present more times in the table, always corresponds to preceding col-
umn).
 Max_BR[Å] – maximum bottleneck radius.
 Avg_L[Å] – average tunnel length.
 Avg_C – average tunnel curvature.
 Priority – tunnel priority calculated by averaging tunnel throughputs over all snapshots (zero value

22
 used for snapshots without tunnels).
 Avg_throughput – average tunnel throughput.

By double-clicking on a desired tunnel cluster (corresponding row in the table) the user can open the
details about the given cluster (see figure 3.5.2).

Figure 3.5.2 Table containing information about individual cluster

This table activates the “Cluster” button. To return to the summary about all clusters, the user can press
“Summary (Clusters)” button. This table contains the following columns:
 Snapshot – name of the input structure, in which the cluster was identified.
 Tunnel cluster – the ID of the tunnel cluster to which a given tunnel belongs (corresponds to the
Tunnel cluster ID in the summary.txt.
 Tunnel – ID of a given tunnel in a given snapshot.
 Throughput – the throughput of a given tunnel (throughput = e-cost).
 Cost – the cost of a given tunnel defined as the balance between the width and length of the tun-
nel.
 Bottleneck radius [Å] – the radius of the bottleneck, i.e. the narrowest part, of a given tunnel.
 Length [Å] – the length of a given tunnel.
 Curvature – the curvature of a given tunnel which is calculated as length/distance, where length is
the length of the tunnel (distance from the calculation starting point to the tunnel ending point
calculated along the tunnel axis) and distance is the shortest possible distance between the calcula-
tion starting point and the tunnel ending point.

By double-clicking on desired tunnel (corresponding row in the table) the user can open the details about
given tunnel (see figure 3.5.3).

Figure 3.5.3 Table containing information about individual tunnel

23
This table provides detailed information about an individual tunnel. It displays tunnel lining residues along
with their exact position in the tunnel. If a given residue influences the width of the tunnel in a position
given by the Distance parameter (counted from the binding site position to the outer surface), it is marked
by a green square.
Moreover, if the tunnel analysis was performed with the compute_bottleneck_residues parameter set to
yes, this table also shows the bottleneck residues which line the narrowest part of the tunnel. These
residues are marked with a blue square.
A yellow square represents the selected residue (it is also added to the active selection). It can be selected
using the left mouse button. Users can also utilize the combination with Shift and Ctrl having the standard
meaning. Selection can be deselected using the Ctrl+D.

If the bottleneck residues were not calculated, the user can start their calculation in the Structure window.
By clicking with the right mouse button on a desired tunnel set and choosing “Statistics → Bottlenecks“ the
user can also compute the bottleneck residues. Reopening the statistics table with details about clusters
(choosing “Statistics → Tunnels“ in the same window as with bottlenecks) displays blue squares represen-
ting bottleneck residues.

The rows of the table represent one sphere contained in the tunnel representation. It starts from the
binding site and ends at the protein surface. The green squares in each row signify their presence around
the sphere.

The columns of the table have the following meaning:

 Length [Å] – the Euclidean distance between the first sphere of the tunnel and the current sphere
(in given table row).
 Individual tunnel lining residues – a set of columns; each column represents one residue which
was present at least once at the tunnel neighborhood. Each column is named according to the one-
letter amino acid abbreviation and it’s PDB ID (e.g., A156).
 X, Y, Z – coordinates of the corresponding sphere center.
 Distance [Å] – represents the distance of the current sphere from the first sphere of the tunnel,
calculated along the tunnel centerline. It depends on the tunnel_sampling_step parameter.
 Radius [Å] – radius of the corresponding sphere.

Residues table

This table provides detailed information about the tunnel lining residues of a given tunnel set (see figure
3.5.4).

24
 Figure 3.5.4 Table containing information about residues lining given tunnel set

Each row of the table represents one tunnel cluster.

Each column of the table represents an amino acid which was marked as a tunnel lining residue at least
once throughout the whole tunnel set.
The green squares mark those residues which line the corresponding tunnel cluster. The yellow squares
determine the selected residues.

The table contains the following controls:

 Save
Allows the user to save the table in the .csv format to a desired location on the hard drive.
 Side chain only
Shows residues where at least one side chain atom of a given residue lines a given tunnel.
For this purpose, all atoms except those named H, N, C, O, CA, or HA are considered as side
chain atoms.
 Minimum residue occurence
This parameter is enabled only when analyzing trajectories of molecular dynamics. It signi-
fies the percentage of a given residue which lines a tunnel during the whole dynamics. For
example, when this number is set to 60%, the user obtains a set of residues which formed
the tunnel in at least 60% of the snapshots of the trajectory (so the less frequent ones are
filtered out).
 Visible columns
Shows the list of all available columns and allows the user to show/hide them using their
checkbox.
 Show all
Displays all available columns, including hidden ones.
 Invert Visibility
Inverts the visibility of columns (the visible ones will be invisible and vice versa).

Bottlenecks table
This table contains detailed information about the bottlenecks of a given tunnel set (see figure 3.5.5).

25
 Figure 3.5.5 Table containing information about bottlenecks of given tunnel set

The controls of the table are similar to the Residues table but some have slightly different meanings.
 Save
Allows the user to save the table in the .csv format to a desired location on the hard drive.
 Minimum bottleneck occurence
Takes into account all tunnels in a given cluster and their bottleneck residues. This number
signifies the frequency of occurrence of these residues in the cluster. For example, a value
of 30% means that a given row remains when all three bottleneck residues have the
occurrence in the corresponding tunnel cluster of at least 30%.
 Visible columns
Shows the list of all available columns and allows the user to show/hide them using their
associated checkboxes.
 Show all
Displays all available columns, including hidden ones.
 Invert Visibility
Inverts the visibility of columns (the visible ones will be invisible and vice versa).

Each row of the table contains information about one tunnel and its bottleneck. The columns have the
following meaning:
 Snapshot – name of the input structure.
 Cluster – ID of a tunnel cluster to which a given tunnel belongs (corresponds to the Tunnel cluster
ID in the summary.txt).
 Individual bottleneck lining residues – residues located within the specified distance from the bot-
tleneck of a given tunnel. They are ordered from the closest to the most distant ones.

26
3.6 Cavity Computation Window
The Cavity Computation window (see figure 3.6.1) controls the computation of pockets and cavities and the
visualization of their solvent-excluded surface (SES). When changing parameters, the surface is updated in
real time. Individual cavities and pockets are distinguished from each other by random colors attached to
them.

Figure 3.6.1 Cavity Computation window

Pockets can be defined with respect to the radii of a pair of spherical probes (a big probe and a small probe)
as the regions of empty space, into which the small probe can get from the outside but the large probe
cannot and where the small probe cannot collide with the large probe. Cavities are regions of empty space,
which can be occupied by the small probe but from which the probe cannot escape to the outside. A very
similar definition of pockets was introduced by Kwabata and Go in their article from 2007 [1].

The window contains the following parts and options:

 Compute/Reset
This action performs the initial computation or resets already computed cavities and pockets.
When the computation is finished, the resulting cavities appear in the visualization window and
their details in the Cavities Overview window. The computation may take a while for larger struc-
tures. Ongoing computations can be canceled in the progress bar.
 Pockets
This option enables or disables the computation of pockets and it can only be changed prior to the
computation or after resetting the result of the last computation. Its default value can be changed
in application settings.
Warning: The computation of pockets is more intensive and more sensitive to numerical errors
than the computation of cavities. The computation can be restarted several times with slightly per-
turbed coordinates if numerical errors are detected.
 Large probe
The radius of the large probe for the computation of pockets. Its default value can be changed in
application settings. Increasing this value will have a negative effect on performance.
 Probe
The radius of the probe for the computation of cavities and pockets. If the computation of pockets
was disabled, it is the radius the single probe for the detection of cavities. If the computation of
pockets was enabled, it is the radius of the second (small) probe. When cavities/pockets are al-
ready computed, this radius can be changed interactively.
 Outer surface
Enables or disables the visualization of the outer surface and allows to change its color. It is the sur-
face, which can be reached by the probe from the outside. This option is available only if the com-
27
 putation of pockets is disabled.
 Advanced coloring
By default, different cavities and pockets are distinguished by random colors. This option enables or
disables advanced coloring schemes according to the strategy chosen in the Coloring window. Col-
ors are interpolated over the whole surface from colors on the atoms.
 Residues included
This list contains residue types that are present in the current molecular structure (amino-acids,
solvent molecules, ligands and non-standard residues). Residue types selected in this list will be in-
cluded in the computation of cavities and pockets. The list of included residues can only be changed
prior to the computation or after resetting the result of the last computation. The height of this GUI
component can be changed manually.
 Frame
This button is for dynamic molecular structures only. It displays the frame number in which cavities
or pockets were computed and allows to move the dynamics to the frame.

Illustrative examples of cavities, pockets and the outer surface are depicted in Figures 3.6.2 and
3.6.3.

Figure 3.6.2 Left: All cavities detected for the probe radius 1.4 A. Right: All pockets detected for the
large probe radius 3 Å and the small probe 1.4 A. PDB: 1CQW.

Figure 3.6.3 Pockets detected for the large probe radius 5 Å and various radii of the small probe. Left: Small probe
radius 1 Å. Right: 0.6 Å. PDB: 1CQW.

[1] Kawabata, T. and Go, N. (2007), Detection of pockets on protein surfaces using small and large probe
spheres to find putative ligand binding sites. Proteins, 68: 516–529. doi:10.1002/prot.21283.
28
3.7 Cavities Overview Window
This window (see figure 3.7.1) displays detailed information about computed cavities and pockets, and
allows to select, hide, and perform also other actions. The table with cavities and pockets can be sorted by
clicking on a column header.

Figure 3.7.1 Cavities Overview window and the context pop-up menu

The window contains the following parts and options:

 Hide All
This action hides all cavities and pockets currently listed in the table. Hidden cavities will still
appear in the table and can be selectively made visible by switching the eye icon in the visibility
column. If the probe radius is manipulated in the Cavity Computation window, hidden cavities and
pockets will still be hidden, but new cavities or pockets can be detected and they will be visible by
default.

 Show All
This action makes visible all cavities and pockets that have ever been hidden, even those that have
been hidden in the past for a different choice of the probe radius. If the probe radius parameter is
manipulated in the Cavity Computation window and the cavity will be detected again, it will be
visible.

 Invert Selection
Inverts the selection of cavities and pockets currently listed in the table. Selected items will be
deselected and deselected items will be selected.
 ID
Zero-based indices of detected items (cavities and pockets).
 Color
The color assigned to each item. It cannot be changed to a custom color.

29
  Visible
The icon indicates whether the item is visible (eye icon) or invisible (crossed eye). A mouse click on
the icon will switch the visibility.
 Volume
A very rough estimation of the volume, in cubic Å. The volume of each item (cavity or pocket) is
computed by random sampling the bounding box of filling balls. Multiplying a rough estimate of the
hit probability by the bounding box volume gives an estimation of the real volume. The number of
samples has to be increased manually to get a better estimation.
 #Samples
This is the number of samples used for the volume estimation. The initial number of samples is
usually very low, hence the first estimation is imprecise. To get a better estimation, double click on
the number of samples and enter a custom value or select the item, open a pop-up menu and
choose “Increase volume precision”, which will add 5000 samples to selected cavities.
 Min. Probe
The limiting probe radius, for which the cavity will open up and disappear. Its surface will become
accessible from the outside and merge with the outer surface. The value is rounded to two decimal
places, so the precision is plus/minus 0.01 Å. This value is meaningful only for cavities. Pockets have
N/A in this field, which means that the value is “Not Available” (the small probe which has been
used to detect a pocket can always escape from the pocket).
 Max. Probe
The radius of the largest possible sphere, which fits inside the cavity / pocket. The value is rounded
to two decimal places, so the precision is plus/minus 0.01 Å.
 Pocket
This column indicates which items represent pockets and which represent cavities.

When clicking with the right mouse button on a row in the table, the following context menu appears:

Figure 3.7.2 Context menu

The context menu offers the following functionality:

 Create Starting Point
This action creates new starting points for the computation of tunnels. The starting point will be
generated in the center of the maximal sphere that fits in the cavity/pocket.
 Hide
This action hides all selected items. The items will still be listed in the table, only their visibility will
be turned off.

30
  Show
This action shows all selected items.
 Remove
This action removes all selected items. The removed items will no longer be available in the table
until the next re-computation of cavities/pockets.
 Select Atoms
This action selects the atoms that touch the surface of the cavity/pocket, but only those which have
been used for the computation. Selected items can be manipulated in the Structure Selections win-
dow.
 Select Residues
This action selects the residues of the atoms that touch the surface of the cavity/pocket, but only
those which have been used for the computation. Selected items can be manipulated in the Struc-
ture Selections window.
 Color by Max. Probe
This action changes the coloring of cavities according to their “Max. Probe” values. This is a simple
way of highlighting large cavities/pockets.
 Color by Volume
This action changes the coloring of cavities according to their volume estimation. This is another
way of highlighting large cavities/pockets.
 Increase Volume Precision
This action increases the precision of the estimated volume of selected cavities/pockets by using
next 5000 samples per each cavity/pocket. The number of samples is in the column #Samples and it
can also be increased manually by double-clicking the field and entering a custom value.

Advanced settings of cavities can be found in the Application Settings (File – Application Settings – Cavities).
There are the following options related to rendering the surface of cavities:
 Pixel Artifacts Correction
This option enables or disables a post-processing step in rendering that reduces minor pixel arti-
facts caused by the analytic computation of solvent-excluded surface on GPU. Disabling this option
may improve performance but some pixels can be rendered incorrectly.
 Face Culling
This option enables or disables back-face culling. Enabling this option may improve performance.
 Rendering Strategies
This section allows to choose a rendering strategy for elements of the solvent excluded surface.
o Point Sprites
This technique should be the fastest one but problematic on some graphic cards.
o Rectangles
This technique can be used instead of Point Sprites.
o Bounding Boxes
The slowest one but most robust technique.

The category Default Values contains several options, which will be used by default in the Cavities Compu-
tation window for newly opened molecular structures. These options decide whether to allow the compu-
tation of pockets, the default radii of the large and small probes, etc.

31
3.8 Locked Probes Window
When cavities are computed, it is also possible to show locked probes. This may help to indicate the
positions of active sites in enzymatic proteins. Locked probes are largest possible empty spheres touching
atoms of the molecular structure (see Figure 3.8.1). Such probes are held by the atomic balls they touch.
They cannot be moved and cannot be expanded without hitting some atomic ball.

Figure 3.8.1 Locked probes - spheres and centers

The parameters for the locked probes can be set in the Locked Probes window (see Figure 3.8.2) and their
visualization can be enabled by activating the corresponding button in the toolbar.

Figure 3.8.2 Locked Probes window

The window contains the following parts and options:

 Visualization
Switches the visualization of locked probes – either spheres or centers of the spheres.
 Probe Radius Filter
This filter limits the locked probes to spheres having the radii in the interval given by its endpoints –
the minimum and the maximum. The button with the Min label and a lock icon activates or deac-
tivates the synchronization of the minimum of the interval and the probe radius in the Cavities
Computation window.
 Probe Accessibility Filter
This filter limits the locked probes to accessible or inaccessible with respect to the probe radius in
the Cavity Computation window. The spheres of inaccessible locked probes are trapped in cavities.

32
3.9 Structure Sequence Window
This window shows the primary structure of all loaded molecules (see figure 3.9). Each row consists of the
following sections:
 Structure name
 List of one-letter abbreviations of residues
Clicking the right button mouse on a residue shows the list of all atoms of a given residue.

Figure 3.9: Structure Sequence Window

3.10 Structure Selections Window

The selections window allows the user to create a selection of atoms, residues and chains. This selection
can be performed within one molecule or can cover parts of more structures. The Selections window has
the following functions:
 New…
Allows the user to create a new selection with a user-specified name.
 None
Disables the selection.
 Atoms
Allows the user to select individual atoms.
 Residues
Allows the user to select whole residues.
 Chains
Allows the user to select whole polypeptidic chains.
 SS
Allows the user to select whole secondary structures.

Each selection has its own record. The record contains:

 Selection name
Displays the name of a given selection.
 Zoom
Zooms and centers the given selection in the visualization window.
 Changing visualization style
Allows the user to change the visualization style of a given selection. Supported styles are:
Dots, Sticks, Balls & Sticks, Van der Waals Radii, Surface.
 Color selection
Allows the user to change the color of a given selection in the scene.
 Show/Hide selection
Shows or hides a given selection in the scene.

33
  Lock selection
Allows the user to lock the selection in a given position so the user cannot rotate and move
the selection in the scene.
 Close
Removes a selection.

Figure 3.10.1: Structure Selections Window

When clicking with the right mouse button on a selection, the following menu appears:

Figure 3.10.2: Menu for each selection

It has the following functions:

 Show residue labels
Shows labeling of residues in a selection.
 Invert
Changes the selection to include all atoms in the scene which were not selected.
 Clear
Deselect all structures of a selection.
 Convert To Starting Point…
Opens the same dialog window as the “From Selection…” button in the Tunnel Computa-
tion window (see section 3.3).
 Rename…
Opens the dialog window for renaming a selection.
 Remove Atoms From Structure
Deletes the selected atoms from the structure.
 Remove Selection
Removes the given selection.
 What’s selected?
Shows the list of atoms/residues/chains present in the selection.
 Structure From Selection
Creates a standalone structure from the given selection. The new structure will appear in
the list of structures in the Structure Overview panel.
34
3.11 Structure Dynamics Window

This window controls the playback of molecular dynamics (see figure 3.11). It is active only when dynamics
are loaded and active.

Figure 3.11: Structure Dynamics Window

The Structure Dynamics window contains the following features:

 Current frame
Shows the number of currently displayed snapshots during the animation.
 Animation step (in frames)
Defines which snapshots will be included in animation (e.g., a value of 2 means that every
second frame will be included).
 Animation speed (in frames per second)
Controls the animation speed.
 Repeat animation loop
Defines whether the entire animation will be played back in a loop.
 Smooth animation
Enables the interpolation between frames.
 From frame, To frame
Defines the starting and ending snapshot of the animation.
 Animation control buttons
Standard buttons for controlling the playback of the animation (e.g., start, stop, pause, fast
forward, etc.)

3.12 Structure Properties Window

This window shows more information about active structure (see figure 3.12). It contains the PDB ID of the
structure, its classification and inner structure – the number of chains, all residues, amino acid residues,
water molecules, ligands, atoms, hetero atoms and atoms with alternate locations. The number of loaded
frames of the structure is also shown.

35
 Figure 3.12: Structure Properties Window

3.13 Structure Statistics Window

This window show detailed information about the constitution of given structure (see figure 3.13). It
contains the list of present atoms, their count in the structure and their radii. Moreover, the list of present
residues and their count is presented.

Figure 3.13: Structure Statistics Window

3.14 Tunnel Graph Window

This window provides settings, visualization, manipulation and animation of graph statistics of computed
tunnels. Basically, there are two types of graphs: those for static snapshots and those for molecular
dynamics (profiles and heat plots).

36
3.14.1 Static case

Figure 3.14.1.1: Tunnel Graph window for static case

In the static case (figure 3.14.1.1), users can:

 Save PNG
Saves the graph as a picture in the PNG format.
 Export CSV
Saves the graph as a table in the CSV format.

Users can also set the following parameters:

 Domain axis
Allows the user to select the feature which will be mapped to the X-axis.
 Range axis
Allows the user to select the feature which will be mapped to the Y-axis.
 Freeze
This function locks the currently displayed graph. It is useful when users want to combine
more graphs into one. E.g. in one graph the user can display the relationship between
tunnel length and width and also between length and distance.
 Clear
Clears the data from the Tunnel Graph window.

The Graph section contains the “Show marks” checkbox, which adds marks to the curve (see figure
3.14.1.2).

37
 Figure 3.14.1.2: Graph with highlighted marks on the curve

The Tunnels section then shows the legend of displayed curves. The color of each curve corresponds to the
color of the corresponding tunnel in the Structures window (where the color of tunnels and their graph
curves can also be changed).

3.14.2 Dynamic case

In the dynamic case (see figures 3.14.2.1 and 3.14.2.2), users can choose between a similar representation
as the static case and heat plots of computed tunnels. Switching between these two representations can be
obtained by choosing between Tunnel Profiles and Heat map (heat plot).

3.14.2.1 Tunnel Graph window for molecular dynamics

The visualization of dynamic tunnels using profiles has the same settings as profiles in the static case. It also
contains the following dynamics controls:
 Snapshot
Displays the currently processed and visualized snapshot. Checking the current checkbox
tightly connects the graph to Structure Dynamics window. When user starts playback in
that window, profiles of tunnels also animate.

38
 3.14.2.2: Example of heat plot (heat map)

With the Heat map (also known as heat plot), the user can set the following parameters:
 Domain axis
Allows the user to select the feature which will be mapped to the X-axis.
 Range axis
Allows the user to select the feature which will be mapped to the Y-axis.
 Scale axis
Allows the user to select the feature which will be mapped to colors specified by the range
displayed on the right side of the heat plot.
 Tunnel
Allows the user to switch between heat plots of individual detected tunnels.
 Clear
Clears the data from the Tunnel Graph window.

In the Radius section, the user can customize the color range of tunnel radii in the heat plot (see figure
3.14.2.3).
The user can manually set the range of radii and the color palette changes according to these new values.

3.14.2.3: Example of different settings of custom color values of tunnel radii

By clicking with the right mouse button on the graph, menu with advanced settings appears.

39
3.15 Coloring Window

The Coloring window (see figure 3.15) is used for coloring the scene background, structures and their parts,
selections, and tunnels. The Global Settings part enables to change the background color of the main
visualization window. There are three main tabs for changing the coloring schemes of structures,
selections, and tunnels.

Figure 3.15: Coloring Window for structures, selections, and tunnels

In all cases the user can choose between predefined Color schemes. Users can also change the colors of
individual atoms (residues, chains or secondary structures) by clicking on the desired chemical element (its
colored row).
The panel contains the following options:
 Structure (Selection, Tunnel)
Enables to select a structure (selection, or tunnel) which will be influenced by the color
change.
 What
Defines which representations will be influenced by the color change.

40
  Color All
This button transfers the color settings to all loaded structures.
 Color Active
This button transfers the color settings to all active structures.
 Color
Defines which parts of the structure will be used for coloring (atoms, residues, chains,
secondary structures, uniform coloring of the whole structure).
 By
For each type in the Color mode, it determines the coloring mode (i.e., hydrophobicity for
residues).
 Scheme
Enables to select a predefined coloring scheme.
 Reset
Returns colors to the currently selected type and color scheme.
 Save
Allows the users to save their own color scheme using the current settings.
 Delete
Deletes the given scheme.
 Export
Exports the given scheme in the .xml format.
 Import
Imports a scheme from the .xml format.

3.16 Alignment Window

This function allows the user to perform structure or sequence alignment.

In structure alignment, we define two structures to align. The result is shown in the visualization window
and the computed RMSD (Root Mean Square Deviation) value is displayed (see figure 3.16). Users can
select the first and second structure for alignment and press the “Align” button.
TIP: To align more structures, it is necessary to choose one “reference” structure as a First structure and
then select a Second structure for alignment.

41
 Figure 3.16: Structure Alignment Window

The sequence alignment enables the user to align two structures according to their sequential constitution.
The basic parameters are the allowed sizes of inserted gaps. The alignment can be applied using the “Align”
button or cancelled using the “Reset” button. The sequence alignment offers also the Advanced mode,
where we can refine the parameters for alignment. We can select the score used, define IDs of residues or
subsequence, which should correspond in the alignment.

3.17 Structure Clip Planes Window

This window allows the user to activate work with several clip planes and slices of clip planes for each
structure, which can stress important features of proteins as well as tunnels (see figure 3.17).

Figure 3.17: Clip Planes Window

The Clip Planes window allows the user to activate several clip planes and slices for each structure and
configure the following settings and functions:
For clip planes:
 Activate or deactivate the clip plane using the On/Off button.
 Set to color of the clip plane.
 Lock the clip plane, i.e., all transformations applied to the structure are applied to the clip plane as
well.
42
  Show or hide the clip plane representation (colored rectangle).
 Distance
Defines the distance of the clip plane from the camera.
 Invert
Inverts the visibility of the clipped structure.
 Realign
In case when the clip plane is locked, this button resets the clip plane position to be per-
pendicular to the camera.
 Define if the clip plane should be applied also to Selections, Tunnels, and Cavities.

For slices we have one extra option:

 Thickness
Defines the thickness of the slice.

3.18 Surface Configuration Window

The Surface Configuration window (see figure 3.18) directly influences the computed and visualized surface
of a selected structure, tunnel, or selection. It enables to define the structure for which we are changing
the surface parameters and determine the surface type. Then we can change the transparency of a surface
for the structure, its tunnels and selections, and change the probe size for the structure's surface and also
for the Detailed Surface method for tunnel visualization.

Figure 3.18: Surface Configuration Window

3.19 Search in Structure

The Search in Structure window (see figure 3.19) enables the user to select a specific part of the structure
by defining the identifiers of residues and/or atoms. First, the user defines the structure in which the
selection should be performed (can be applicable also to all structures in the scene), then defines the chain
identifier and the identifiers of residues and atoms (individual numbers divided by comma, ranges). Then
the selected residues and/or atoms can be added to an existing selection or can form a new selection. The
“Select“ button finishes the process. The “Center & Zoom“ button centers and zooms the selection in the
visualization window. The “Close“ button closes the window without any action.

43
 Figure 3.19: Search in Structure Window

3.20 Delete from Structure

The Delete from Structure window (see figure 3.20) serves for deleting the user-specified part of structure.
The user can delete a selection or specify the structure, its chain, and identifiers of residues and/or atoms,
which should be deleted. The “Remove“ button finishes the process. The “Center & Zoom“ button centers
and zooms the selection in the visualization window. The “Close“ button closes the window without any
action.
Warning: This process is irreversible, after deleting a given part of structure the structure will contain a gap.

Figure 3.20: Delete from Structure Window

3.21 Measurement Window

The Measurement window (see figure 3.21) enables the user to measure distances, angles, and dihedral
angles between atoms. First, the user has to select the type of measurement in the top part of the panel.
Then, by clicking to the scene (Ctrl + left mouse click on a selected atom) or specifying the atom identifier
directly in the text field, the user defines a given number of atoms – two for distance, three for angle, four
for dihedral angle. After adding the last atom, the measurement is automatically performed and it is listed
in the panel. The measurement is also visualized in the main window as a line connecting the corresponding
atoms and with the resulting measured number next to the line. The user can change the color of the line,
show measurement graph, show/hide the measurement, and delete the measurement.
When performing the atom selection, the given icon next to the atom field has to be selected (it is stated
by blue background of the icon).

44
 Figure 3.21: Measurement Window

3.22 Mutagenesis Window

The Mutagenesis window (see figure 3.22) enables the user to perform mutations of individual residues on
static structures. The panel contains the list of performed mutations, which can be cancelled by using the
“Restore checked“ or “Restore all“ button.
The Add mutation section serves for specification of the mutation. First the user defines the residue
identifier (numerical value). Then the rotamer library has to be selected which will be used for searching
the rotamers. Then the user specifies the type of new amino acid and selects the desired rotamer from the
given list. Here the decision can be based on the score of individual rotamers, based on their Probability of
occurrence, Standard deviation, and Collision penalty score. The “Apply mutation“ button performs the
mutation and creates an item in the list of mutations. The “Cancel mutation“ button cancels the operation.
The ”Show collisions” button displays the collision of the selected rotamer with the rest of the structure in
the visualization window.

Figure 3.22: Mutagenesis Window

45
3.23 Asymmetric Tunnels Window

The Asymmetric Tunnels window (see figure 3.23.1) helps to specify parameters for the asymmetric
representation of tunnels (see figure 3.23.2). This representation can be calculated either automatically
based on the surrounding atoms or can be controlled by the user-defined parameter, specifying the
enlargement of the probe size used for the tunnel computation. This method searches for tunnel surface in
highly asymmetric cavities and calculates the tunnel volume. This is the difference between the asymmetric
tunnel and the detailed surface representation, available in the list of possible tunnel representations. The
tunnel detailed surface is a simple extension of the probe size used for tunnel computation. It is set
automatically and the user does not have any control of the appearance. When the tunnel is highly asym-
metric, this representation does not cover completely the bulky parts of the tunnel. The detailed surface is
used for the computation of contours in cross cuts, used in the Contours window (see section 3.25).

The Asymmetric Tunnel Computation window (see figure 3.23.1) contains the following items:
 Extrusion – defines the tunnel surface enlargement. It can be automatic (then it searches for the
tunnel-lining residues and their atoms) or manual (enlargement of radii of spheres forming the
tunnel computed by the CAVER tool).
 Voxel size – defines the density of the grid which is used for searching for the tunnel-lining resi-
dues.
 Probe radius – size of the probe. Can be defined manually by the user or the tool can automatically
use the size of the bottleneck as a probe size.
 Change settings – recalculates the asymmetric tunnel after changing the parameters.
 Select lining residues – selects the closest residues lining the tunnel (a selection will be created
which will be displayed in the visualization window).
Then follows the list of calculated asymmetric tunnels, along with their volume.

Figure 3.23.1: Asymmetric Tunnel Computation window

46
 Figure 3.23.2: Asymmetric representation of a tunnel

3.24 Residue Graph Window

The Residue Graph window (see figure 3.24.1) enables the user to visually explore the evolution of a
selected tunnel over time in detail. The window consists of the following parts:
 Manipulation panel
In this top part of the window, the user selects the structure and one of its tunnels, which should
be explored in the bottom parts. The computation of the tunnel profile and its surrounding resi-
dues is performed using the “Compute“ button.
o Analyse Selected tunnels – enables to explore more tunnels at once.
o Import Contour Data – enables to import the stored data about the contours, computed us-
ing the Contours window (see section 3.25).
o Graph – enables to color the tunnel profile below based on different properties – time,
tunnel length, tunnel maximal radius, and tunnel ID.
o Residues – enables to color the surrounding residues based on their different properties –
hydrophobicity, partial charge, donors and acceptors, tunnel ID, and influence of the tunnel
by these residues.
o Detailed – changes the representation of the surrounding residues below from an interpo-
lated representation (default) to line-by-line representation (each line corresponds to one
timestep).

 Tunnel profile
This graph representation plots the tunnel profile (width) along its centerline. On the left side is the
starting point of the tunnel in the active site, the right side represents the molecular surface (the
tunnel gorge). Each line represents the tunnel in one timestep. From the graph, the user can clearly
detect the narrowest parts of the tunnel (bottlenecks) and the most stable and unstable parts of
the tunnel over time.
 Surrounding residues
The bottom part of the window shows the list of all residues contributing to the tunnel. Each of
them is represented by a thick colored line, and the residue identifier is positioned on the left side
of the line. The color corresponds to a selected property (i.e., hydrophobicity). The horizontal posi-
tion of the color in the line shows which part of the tunnel this residue influences (related to the

47
 tunnel profile above). The longer the colored line, the more it influences the tunnel.

Figure 3.24.1: Residue Graph window

The vertical red slider shows the constitution of the tunnel (its surrounding residues) in a given position
along the tunnel centerline. By clicking on a specific site, the user launches the calculation of the detailed
cross cut of the tunnel in this site, which can be explored in detail using the Contours window (see section
3.25).

Right mouse button click in the window opens the menu containing the following items (see figure 3.25.2):
o Properties… – opens a window where the user can set various properties influencing the
appearance of the tunnel plot.
o Copy – enables to save the content of the window (without the top panel) to clipboard.
o Save as – enables to save the content of the window in the PNG format.
o Clear Selection & Filters – removes all applied filters (shows data for all timesteps).
o Show Filters – enables to use filtering based on selected values of the residue properties.
o Sort Residues – enables to sort residues according to a selected property (6 options available).

Figure 3.24.2: Menu with advanced options

Details about the design rationale of the Residue Graph and examples of usage can be found in [2].

[2] Byska, J., Le Muzic, M., Gröller, E. M., Viola, I., Kozlikova, B. AnimoAminoMiner: Exploration of Protein
Tunnels and their Properties in Molecular Dynamics. IEEE Transactions on Visualization and Computer
Graphics, IEEE, 2016, 22(1), 747-756. ISSN 1077-2626. doi:10.1109/TVCG.2015.2467434.

48
3.25 Contours Window
The Contours window (see figure 3.25.1) enables the user to visually explore the evolution of a selected
tunnel cross cut (i.e., bottleneck) over time in detail.

Figure 3.25.1: Contours window

The window consists of the following parts:

o Manipulation panel
Colored by – changes the coloring of the contours according to time or contour area.
Show Circles – shows the “coordinate system” with a measure.
Show 3D – projects the contour to the 3D main visualization window so the user can observe simul-
taneously the 3D representation of the tunnel, its cross cut, and the contour representation.
o Contour visualization
The main part of the window contains the 2D contour representation. Each contour represents one
timestep. It can be colored according to time or area occupied. The surroundings of the contours is
filled with bars representing individual residues forming the tunnel in this cross cut part. The posi-
tion of the bars corresponds to their actual positions around the cross cut. Each residue can be rep-
resented by more bars, arranged in a co-circular manner – each bar represents one selected prop-
erty (in figure 3.25.1 hydrophobicity and partial charge). When a bar is filled only partially with a
color, it means that this residue did not contribute to this tunnel cross cut in all timesteps.

Right mouse button click in the window opens the menu containing the following items (see figure 3.25.2):
o Properties… – opens a window where the user can set various properties influencing the
appearance of the tunnel plot.
o Copy – enables to save the content of the window (without the top panel) to clipboard.
o Save as – enables to save the content of the window in the PNG format.
o Clear Selection & Filters – removes all applied filters (shows data for all timesteps).
o Show Filters – enables to use filtering based on selected values of the residue properties.

49
 Figure 3.25.2: Menu with advanced options

The Contour window is interactively connected with the Residue Graph window (see section 3.24).

Details about the design rationale of the Contours and examples of usage can be found in [3].

[3] Byska, J., Jurcik, A., GRÖLLER, E. M., Viola, I., Kozlikova, B. MoleCollar and Tunnel Heat Map
Visualizations for Conveying Spatio-Temporo-Chemical Properties Across and Along Protein
Voids. Computer Graphics Forum, Wiley-Blackwell, 2015, 34(3), 1-10. ISSN 0167-7055.
doi:10.1111/cgf.12612.

3.26 Console Window

The Console window (see figure 3.26) enables the user to control the most important functions of CAVER
Analyst using the command line.
When typing help to the command line, the list of all commands appears.
Using help xxx, where xxx stands for one of the commands, displays the description of usage of the given
command.

Figure 3.26: Console Window with the list of all supported commands

50