FEP+ Panel — Map Tab

This tab displays the perturbation graph for relative ligand binding FEP or protein mutation FEP, with tools for editing it, and a summary of the ligand or protein data and hysteresis data (after a calculation has been run).

This tab is present for relative binding FEP and for protein mutation FEP. For protein mutation, some of the ligand-related controls are not present.

  • Using
  • Features
  • Additional Resources

Using the Map Tab

Some of the main tasks are described below. For protein mutation FEP, you cannot add or remove proteins or add connections.

To rearrange the layout:

  • Drag a ligand box to reposition it.
  • Drag over several ligand boxes to select them, then drag them to reposition the group.
  • Drag with the right mouse button to move the entire diagram.
  • Use the mouse wheel or buttons to zoom in or out.
  • Use the Auto-layout button to optimize ("clean up") the layout.

To highlight particular perturbations:

  • Pause the pointer over a ligand. The ligands to which it is not directly connected, with their connections, are dimmed.

To delete ligands or connections from the diagram:

  1. Select the ligands and connections.

  2. Click the Delete button.

To add connections to the diagram:

  1. Click the Add new connection button.

  2. Select the ligand pairs.

Note: This feature is not available for Academic FEP+ users

To add ligands from a file to the diagram:

  1. Click the Import button and choose Import Ligands → From File from the menu.

  2. Select the file in the dialog box that opens.

To sketch new ligands to add to the diagram:

  1. Click the Sketch button.

  2. Sketch a ligand in the 2D Sketcher.

  3. Click Save as New.

  4. Repeat steps 2 and 3 for each new ligand.

  5. Click Close.

To add ligands from the Project Table to the diagram:

  1. Select the ligand entries in the Project Table or Entry List.

  2. Click the Import button and choose Import Ligands → From Project Table from the menu.

You can also add ligands in the Project Table by adding them directly to the FEP+ entry group that is created for the FEP+ map. With this method, it is recommended that you duplicate the ligand into the entry group, so that the original ligands are preserved.

To export the FEP+ map:

  1. Click the Export button and choose FEP → Map to export the map as a PNG or Perturbation Map to export the map as a .fmp file. See the Toolbar for more Export options.

Map Tab Features

Summary tab

This tab contains a table of the ligands that have been imported into the workflow. The table columns are described below.

# Ligand number: index of the ligand in the order in which it was imported.
Ligand or Protein Title of the ligand or the mutated protein. This is linked to the title in the Project Table, so updating it there updates it in this table.
Bias Column of check boxes that can be use to bias the graph construction to draw more connections to the ligand. The biasing is done by boosting the calculated similarity to other ligands. If you change the bias settings, you must click Regenerate Map to generate a graph with the new settings.
Exp. Affinity

Experimental property free energy value, which can be affinity or solubility depending on the simulation. Any ligand whose binding affinity or solubility free energy you enter is favored in the construction of the graph, by adding a further bias to the similarity. The property is also used in the analysis of the FEP results. To add experimental data, click the Affinity button. Once you have added experimental data, you can edit individual cells to set or change the values. For binding affinity, the value is preceded by a > or a < sign if the ligand has been designated as Top of assay or Bottom of assay (values outside the assay range). To clear all experimental data, right-click on the column heading and choose Clear All.

Depending on the quantity represented, this column shows the predicted error in pKi or ΔG, or the predicted error factor for the Ki value. Multiplying and dividing the Ki value by this factor gives the error bounds. You can double-click in the cell to edit the value (except if the units are Ki) and error on both sides of the "±" indicator (as long as the values are valid numbers). To clear all experimental error data, right-click on the column heading and choose Clear All. Additionally, right-clicking on the column displays the option to sort by either value or error.

The heading changes when you click the Units button and choose a different quantity to display.
Pred. Affinity Final predicted binding affinity (ΔG) in kcal/mol for a set of ligands. This prediction requires the experimental binding free energy for at least one of the ligands, which is used as the reference. If the experimental binding free energies for multiple ligands are available, then all these ligands are used as the reference to convert the relative affinity to the final predicted affinity, forcing the average of the predicted affinities and average of the experimental affinities to be equal among the ligands with experimental binding free energies. See Conversion of ΔΔG Values to ΔG Values for more information. Also shown is the minimum error in kcal/mol along all paths from the reference ligand to the ligand.

The table also has a shortcut menu, with which you can perform the following actions:

  • Export Structures—Export the structures in the selected rows in Maestro or (3D) SD format. To export in 2D format, use the 2D Viewer Panel.
  • Select All—Select all the table rows.
  • Select Inverse—Invert the selection of table rows. The selected rows are deselected, and the unselected rows are selected.
  • Set as Reference—Use the selected rows as the set of reference ligands.
  • Set as Top of Assay—Mark the selected ligands as ones whose experimental affinity is above the upper limit of the assay, and therefore a reliable error cannot be assigned. These ligands are excluded from cycle closure calculations and from the statistical analysis in the correlation plots. A > sign is inserted before the affinity value to show its relation to the assay limit.
  • Set as Bottom of Assay—Mark the selected ligands as ones whose experimental affinity is below the lower limit of the assay and therefore a reliable error cannot be assigned. These ligands are excluded from cycle closure calculations and from the statistical analysis in the correlation plots. A < sign is inserted before the affinity value to show its relation to the assay limit.
  • Clear Top/Bottom of Assay—Clear the designation of the ligand as outside the assay range.
  • Show Representative Structures—Add an entry group in the Project Table containing the representative structures of the selected ligands for each connection in the map to these ligands, and show them in the Workspace in a tiled view, replacing the Workspace. The group is named Node representatives, and contains a subgroup for the representatives of each selected ligand, if there is more than one ligand selected. The entries in the group are replaced if you show a different set of representative structures.
  • Clear Affinity Data—Clear the experimental affinity and error data from the selected rows.
  • Delete—Delete the ligand from the map. You can also use the Delete key.
Hysteresis tab

This tab reports hysteresis values around various cycles (closed paths in the graph). When no edges are selected, all cycles are shown in the table. When edges are selected in the map, only cycles containing one or more of the edges are shown in the table. If only a single edge is selected, a Show Atom Properties panel is shown below the table. The tab is not displayed if the map has no edges.

# of Nodes

Number of nodes in the cycle for which the hysteresis was evaluated. The cell is colored by the quality of the hysteresis; a legend is shown below the table.
Hysteresis

Value of the hysteresis around a cycle. The hysteresis is scored as value/√N, where N is the length of the cycle. The value is colored according how well it has converged, based on this score:

  • Good—green, score less than 0.5
  • Yellow—acceptable, score between 0.5 and 0.8
  • Red—poor, score greater than 0.8.
Cycle

Path taken in the cycle for which the hysteresis was evaluated, given as a list of nodes connected by arrows indicating the direction of the path. If the entire path is not visible in the table, it is displayed in a tooltip when you pause the pointer over the table cell.
Groups tab

This tab shows groups of protomers (protonation states), tautomers, or conformers of a given compound, for which binding free energies can be calculated taking account of the populations of these species in solution. The structures belonging to each compound are automatically detected.

Protonation and deprotonation events are fast and ubiquitous. What may be thought of as a single ligand may in fact be an ensemble of rapidly interconverting tautomers and protomers. The tautomers and protomers of ligands in FEP+ can be accounted for by adding every state as a separate node in the perturbation map. If the populations or pKa values of the protomers or tautomers in water are known, the complex populations or pKa values can be calculated by FEP+ using the method described by de Oliveira et al. [9].

Likewise, if there is uncertainty about which conformation of a ligand binds to a protein, or if a ligand can bind in multiple different poses, each conformation can be added as a separate node in the perturbation map. After the FEP+ calculation has finished, the effect of the conformations can be automatically accounted for on the assumption that the ligand cannot switch between the different conformations in the complex, but can and readily does so in the solvent.

Only present if the fmp file is from a small molecule relative binding calculation and Binding is selected from the Affinity button menu.

Use pH text box

Use the specified pH to calculate the populations of protomers in solution.

Auto-populate

Use Epik to calculate the pKa values for a given pH and populate the table.

Groups table

This table shows the groups that were identified and allows you to edit the data to specify the populations or pKa values for the compound represented by the group. Hover over the row and click the edit (pencil) icon, or double-click the row to open the Grouping Dialog Box for editing the data for a compound (group).

Titles This column lists the structures in the group (or compound) by title as a comma-separated list.
# This column shows the number of structures in the group.
Solvent Population This column shows the populations of each structure in solution, separated by slashes. The populations are specified either via the pKa (for protomer pairs) or directly (for tautomers). Conformer populations in solution are assumed to be equal.
Solvent pKa This column shows the pKa for the compound (as applicable).
Complex Population This column shows the populations of each structure in the complex, separated by slashes. The populations are derived from the values input for the compound in the solvent.
Complex pKa This column shows the pKa for the structure in the complex (as applicable).
Clear All link

Clear the population and pKa data from the table.

Show group information in options

  • Table only—Only show group corrections in the Groups table. This allows you to change the map after applying group corrections to the free energy.

  • Map and other data displays—Apply corrections based on the populations of the groups to the free energies. The group free energy value is applied to all structures in the group. You cannot change the map while the group free energies are in effect.

Show Atom Properties option menu and display

This option menu and display is shown when you click on a connection (edge) in the perturbation map. It contains the 2D structures for the connection, with options for annotating the structures on the option menu, and arrow buttons below the structures for stepping through the connections (edges).

  • Mutation—highlight the mutation. The mutated atoms are marked with red circles and the changed bonds are colored red.

  • Atom mapping—Show the atom numbering on the structures.

  • Common core—Mark the common core with green bonds.

  • Hot atoms (solvent leg)—Mark the ligand atoms that are sampled more extensively in the solvent, effectively simulated at an elevated temperature (in the REST region). The hot atom choice can be changed in the FEP+ Hot Atoms Dialog Box.

  • Hot atoms (complex leg)—Mark the ligand atoms that are sampled more extensively in the complex, effectively simulated at an elevated temperature (in the REST region). The hot atom choice can be changed in the FEP+ Hot Atoms Dialog Box.

  • None—Show the structures without any annotation.

Show in Workspace buttons

Show structures or trajectories in the Workspace for the selected connection. These buttons are only displayed when you have a single connection (edge) selected in the map.

  • Representative structures button—show the representative structures for the connection in the Workspace, in a tiled view, excluding all other entries. The structures are imported into the project as entries in a special group, named Edge representatives, and are replaced if you show representative structures for a different connection (edge) or for a ligand (node).

  • Trajectories—Import the trajectories for the two ends of the connection, along with the structures, into a special entry group. The button is a menu from which you can import the trajectories for the complex leg, the solvent leg, or the vacuum leg. You can then play the trajectory for one of the entries by clicking the T button for the entry.

Perturbation map

The perturbation diagram consists of boxes connected by lines. The boxes (nodes in the graph) represent the ligands; the lines (edges or connections) represent allowed perturbations. You can drag the boxes to reposition them. The box is green for unconnected (new) ligands, and black for connected ligands.

The buttons at the top left allow you to zoom in or out on the diagram, and fit the diagram to the display area.

Zooming in or out can also be done with the scroll wheel. When zoomed out, the ligands are represented by a number in a box. As you zoom in, the number is replaced by the 2D structure when the resolution is high enough to display it. Double-clicking on a connection zooms in to that connection.

When you pause the pointer over a ligand, the parts of the graph to which it is not connected are dimmed. When zoomed out, the 2D structures are also displayed in a tooltip.

Right-clicking on a ligand (node) displays the same shortcut menu as right-clicking on a row in the Summary tab (see above).

You can select ligands and connections by clicking on them, or dragging over them. You can add ligands or connections to the selection, or remove them from the selection, with control-click. The selected objects are highlighted in yellow. Selected ligands are included in the Workspace.

Right-clicking on a connection (edge) displays a menu, with the following items:

  • Analyze—opens a panel with a more detailed analysis of the perturbation, including plots of data as a function of simulation time and analyses of protein-ligand interactions. See FEP+ — Analysis Panel for more information. The same analysis is done when you click on a ligand pair row in the Analysis tab. This item is present if the FEP results are available.

  • Import Trajectories (Complex Leg)—import the trajectories for the complexes at either end of the connection. The structures are added as entries to a Trajectories subgroup in the entry group for the structures in the FEP+ panel, and have the usual T button for viewing the trajectory.

  • Import Trajectories (Solvent Leg)—import the trajectories for the ligand structures in solvent at either end of the connection. The structures are added as entries to a Trajectories subgroup in the entry group for the structures in the FEP+ panel, and have the usual T button for viewing the trajectory.

  • Import Trajectories (Vacuum Leg)—import the trajectories for the ligand structures in vacuum at either end of the connection. The structures are added as entries to a Trajectories subgroup in the entry group for the structures in the FEP+ panel, and have the usual T button for viewing the trajectory.

  • Show Representative Structures—Add an entry group in the Project Table containing the structures at either end of the connection (the "representative structures"), and show them in the Workspace in a tiled view. The tiling is canceled if you add a structure to the Workspace, but the entry group remains unless it is replaced by another set of representative structures. The group is named Edge representatives If you have more than one connection selected, the representative structures for each connection are added as subgroups.

  • Delete—delete the connection.

  • Clear Predictions—Deletes predicted ΔΔG values and data found in the Analysis tab for relative binding FEP+ calculations..

Note that, if ligands are selected as well as connections, the ligands are deselected when you right-click. Some of the items are not available for protein mutation FEP.

Clicking on a ligand connection displays the Show Atom Properties section, which contains the 2D structures for the connection and options for annotating the structures. In addition, the two ligands at the ends of the connection are added to the entry group for the ligands, in a new subgroup labeled Representatives. The ligands are displayed in the Workspace side by side in a split screen view (Tile mode), zoomed in to the ligands, with protein residues, ribbons, waters, and any interactions displayed. When you click elsewhere in the map so that the connection is no longer selected, this subgroup is removed from the project and the ligands are no longer displayed in the Workspace. When a connection is selected, you can use the arrow buttons below the 2D structures to step through the connections.

Generate Map button

Generate the perturbation scheme ("map") from the ligands in the table. The Map Options Dialog Box opens so you can set options for map generation; clicking Generate Map or Update Map in this dialog box starts the map generation. The ligand structures are analyzed to find possible perturbations between them. This may take a few minutes, depending on the number of ligands. A progress dialog box is displayed while the analysis is running. When the analysis finishes, the diagram showing the possible perturbations is displayed. The previous results and analyses are cleared when you regenerate the map. However, you can use the Undo button to undo a map generation.

You can generate a map for all ligands or for selected ligands only. If ligands are selected in the map, the map is generated or updated for these ligands only; the rest of the map is not changed. If no ligands, one ligand, or all ligands are selected, the map is generated for all ligands. Note that, while you can change the ligand selection while the Map Options Dialog Box is open, the map is generated for the ligands that are selected when you click Generate Map.

If ligands have identical titles, a dialog box warns you of the duplication. You can click OK to automatically rename the ligands with duplicate titles, by appending a numerical suffix. The map generation then proceeds. If you do not want to automatically rename the ligands, click Cancel; you will then have to remove the duplicates in order to generate a map. To find duplicate ligands, you can click the Ligand column heading in the summary table to sort by title.

The graph is constructed by determining the maximum common substructure between the ligands (using Canvas MCS with special atom types for FEP), and calculating the similarity. If you want to redefine the core for comparison of ligands, you can do so in the Map Options Dialog Box. Connections are made between structures with high similarity. The similarity between one ligand and the rest can be boosted by setting the affinity or selecting the Bias option for that ligand in the Summary tab. This is the default. Alternatively, the graph can be constructed by making connections from all the ligands selected for bias to all the other ligands, with no connections between the other ligands. To choose this option, select Star topology under Map topology type in the Map Options Dialog Box. The default option selected is Optimal topology. The choice of topology is saved and is used in all sessions until you change it again.

The similarity score is based on several ingredients:

  • Difference in heavy-atom counts in the mutated part of the structures: a larger difference gives a smaller score
  • Whether a covalent bond needs to be formed or broken for the perturbation
  • Whether the net charge changes, or whether a charged group is mutated to a different charged group
  • RMSD of the core: a larger RMSD gives a smaller score
  • Bias towards selected molecules
  • Flipped rings cause a low similarity

Perturbations that change both the net change and the ring topology (core-hopping) are currently not supported in FEP+. These perturbations have a similarity score of 0.

Hot Atoms button

Make settings for hot atoms. Opens the FEP+ Hot Atoms Dialog Box.

Display perturbation properties button menu

Display perturbation properties on the map. The properties are color-coded and placed at the midpoints of the lines. You can select multiple properties to display. The selected properties are highlighted on the menu, and shown in a legend at the lower left of the map.

  • Predicted ddG (Raw)—raw ΔΔG from FEP calculation
  • Predicted ddG (Corrected)—final ΔΔG after corrections
  • Experimental ddG—value derived from input experimental affinities
  • Solvation ddG—ΔΔG of solvation
  • Bad perturbations—mark perturbations with a high error. The lines are drawn in orange with the text "Bad perturbation" below the line.
  • Similarity scores
  • Protocol—protocol used for the edge. Can be one of default, core-hopping, fragment-linking, charge-hopping.
  • SMARTS matches—mark ligands that matched the core SMARTS pattern with an asterisk (*).
Toolbar

The toolbar has the following buttons:

Auto-layout

Automatically optimize the layout of the graph in the panel.

Add new connection
Add a connection between two ligands in the perturbation diagram. Click this button, then click the ligands to add the connection. Not present for protein mutation FEP

This feature is not available for Academic FEP+ users
Undo

Undo changes to the perturbation scheme, including map generation and addition of ligands. You can undo multiple changes. Undo also applies to setting affinity values.
Redo

Redo changes to the perturbation scheme. You can redo multiple undone changes.
Delete

Delete the selected ligands and connections from the perturbation scheme. The ligand entries are also deleted from the FEP+ entry group in the Project Table.

Related Topics