bioluminate quick start guide

BioLuminate Quick Start Guide

BioLuminate 1.0

Quick Start Guide

Schrödinger Press

BioLuminate Quick Start Guide Copyright © 2012 Schrödinger, LLC. All rights

reserved.

While care has been taken in the preparation of this publication, Schrödinger

assumes no responsibility for errors or omissions, or for damages resulting from

the use of the information contained herein.

BioLuminate, Canvas, CombiGlide, ConfGen, Epik, Glide, Impact, Jaguar, Liaison,

LigPrep, Maestro, Phase, Prime, PrimeX, QikProp, QikFit, QikSim, QSite, SiteMap,

Strike, and WaterMap are trademarks of Schrödinger, LLC. Schrödinger and

MacroModel are registered trademarks of Schrödinger, LLC. MCPRO is a trademark

of William L. Jorgensen. DESMOND is a trademark of D. E. Shaw Research, LLC.

Desmond is used with the permission of D. E. Shaw Research. All rights reserved.

This publication may contain the trademarks of other companies.

Schrödinger software includes software and libraries provided by third parties. For

details of the copyrights, and terms and conditions associated with such included

third party software, see the Legal Notices, or use your browser to open

$SCHRODINGER/docs/html/third_party_legal.html (Linux or Mac OS) or

%SCHRODINGER%\docs\html\third_party_legal.html (Windows OS).

This publication may refer to other third party software not included in or with

Schrödinger software ("such other third party software"), and provide links to third

party Web sites ("linked sites"). References to such other third party software or

linked sites do not constitute an endorsement by Schrödinger, LLC or its affiliates.

Use of such other third party software and linked sites may be subject to third

party license agreements and fees. Schrödinger, LLC and its affiliates have no

responsibility or liability, directly or indirectly, for such other third party software

and linked sites, or for damage resulting from the use thereof. Any warranties that

we make regarding Schrödinger products and services do not apply to such other

third party software or linked sites, or to the interaction between, or

interoperability of, Schrödinger products and services and such other third party

software.

Revision B, September 2012

Contents

Document Conventions ...................................................................................................... v

Chapter 1: Introduction ....................................................................................................... 1

Chapter 2: Using the Toggle Table ............................................................................. 3

Chapter 3: Examining Sequences ............................................................................ 11

Chapter 4: Identifying Reactive Residues ........................................................... 17

Chapter 5: Identifying Consensus Waters........................................................... 21

Chapter 6: Homology Modeling of Proteins ....................................................... 27

Chapter 7: Scanning for Residue Mutations ...................................................... 35

Chapter 8: Locating Possible Mutations for Disulfide Bridges .............. 41

Chapter 9: Modeling an Antibody.............................................................................. 47

Chapter 10: Protein-Protein Docking ...................................................................... 59

10.1 General Protein-Protein Docking........................................................................ 59

10.1.1 Docking the Ligand Protein As Is.................................................................... 60

10.1.2 Docking the Prepared Ligand Protein ............................................................. 61

10.1.3 Docking the Native Ligand to the Native Receptor ......................................... 63

10.1.4 Comparing the Top Poses............................................................................... 66

10.1.5 Calculating the RMSD Between the Top Poses .............................................. 67

10.1.6 Examining the Poses and the Binding Sites ................................................... 69

10.2 Antibody-Antigen Docking .................................................................................. 71

10.2.1 Preparing the Antibody ................................................................................... 71

10.2.2 Setting Up and Running the Job ..................................................................... 72

10.2.3 Comparing the Best Pose with the Reference ................................................ 73

Getting Help ............................................................................................................................. 77

BioLuminate 1.0 Quick Start Guide iii

iv
Schrödinger Suite 2012 Update 2

Document Conventions

In addition to the use of italics for names of documents, the font conventions that are used inthis document are summarized in the table below.

Links to other locations in the current document or to other PDF documents are colored likethis: Document Conventions.

In descriptions of command syntax, the following UNIX conventions are used: braces { }

enclose a choice of required items, square brackets [ ] enclose optional items, and the barsymbol | separates items in a list from which one item must be chosen. Lines of commandsyntax that wrap should be interpreted as a single command.

File name, path, and environment variable syntax is generally given with the UNIX conven-tions. To obtain the Windows conventions, replace the forward slash / with the backslash \ inpath or directory names, and replace the $ at the beginning of an environment variable with a %at each end. For example, $SCHRODINGER/maestro becomes %SCHRODINGER%\maestro.

Keyboard references are given in the Windows convention by default, with Mac equivalents inparentheses, for example CTRL+H (H). Where Mac equivalents are not given, COMMANDshould be read in place of CTRL. The convention CTRL-H is not used.

In this document, to type text means to type the required text in the specified location, and toenter text means to type the required text, then press the ENTER key.

References to literature sources are given in square brackets, like this: [10].

Font Example Use

Sans serif Project Table Names of GUI features, such as panels, menus, menu items, buttons, and labels

Monospace $SCHRODINGER/maestro File names, directory names, commands, envi-ronment variables, command input and output

Italic filename Text that the user must replace with a value

Sans serif uppercase

CTRL+H Keyboard keys

BioLuminate 1.0 Quick Start Guide v

vi


Chapter 1

Chapter 1: Introduction

The exercises presented in this manual illustrate the use of BioLuminate for a variety of appli-cations. As well as the steps involved to run the main application, each exercise includes anynecessary preparatory steps and steps for analysis of results. The exercises all use structures orsequences from the PDB. Some of them run BLAST searches. You should ensure that you haveaccess to the PDB and BLAST databases, either locally or on the web.

• Chapter 2 covers basic operations in the BioLuminate interface, focusing on the ToggleTable.

• Chapter 3 introduces the Multiple Sequence Viewer and some of its functions.

• Chapter 4 contains an exercise on identifying and assessing reactive residues in a protein.

• Chapter 5 contains an exercise on visualizing the consensus between homologous pro-teins for species other than the protein itself—in this case, water.

• Chapter 6 contains an exercise on homology modeling of a small protein, with analysis ofthe quality of the model using a Ramachandran plot, deviations from ideal values ofstructural parameters, and visual comparison to the known X-ray structure after align-ment of the model and the known structure.

• Chapter 7 contains an exercise on scanning a protein for residue mutations and assess-ment of the binding affinity of the two protein chains for each other as a function of themutation. It includes a full protein preparation.

• Chapter 8 contains an exercise on scanning a protein for pairs of residues that could bemutated to cysteine to form a disulfide bridge, and ranking of the candidates by theireffect on protein stability.

• Chapter 9 contains an exercise on homology modeling of an antibody, with visual com-parison to the known X-ray structure after structural alignment. The exercise also demon-strates use of the Workspace sequence viewer to select chains and residues.

Information on getting help is provided at the end of this document.

BioLuminate 1.0 Quick Start Guide 1

2


Chapter 2

Chapter 2: Using the Toggle Table

The set of exercises in this chapter provides a basic introduction to the use of the BioLuminateinterface for displaying structures. The focus is on the Toggle Table panel, which is the primarytool in BioLuminate for interacting with objects in the Workspace.

First, a structure must be imported.

1. Choose File → Get PDB.

The Get PDB File dialog box opens.

2. Enter 2ot3 in the PDB ID text box, and click Download.

An information box is displayed, indicating that there are problems in the structure. Thisoften means that there are atoms missing, or nonstandard residues. It is an indication thatsome action to fix the structural issues must be taken before using the structure for mod-eling. This exercise does not involve any modeling, so the warning is of no concern here.If you don’t want to see this message again check the Do not show this dialog again box.

3. Click OK in the Information box.

The 2ot3 structure is imported into the project and displayed in the Workspace. Thebonds are represented as lines. The color scheme indicates what kinds of problems werefound when importing the structure—see Section 3.1.6 of the Maestro User Manual formore information. These structural defects are of no concern for this exercise, but if youwant to use the structure for modeling, these defects would need to be fixed. Later chap-ters include exercises in which the defects are fixed in the Protein Preparation Wizardpanel.

Now that the structure is imported, you can start with some basic operations on the view.

4. Rotate the structure in the Workspace:

• If you have a mouse, hold down the left mouse button and drag horizontally. • If you have a trackpad, hold down CTRL () and click and drag on the trackpad.

If you have not configured your trackpad, go to Edit → Settings → Mouse Actionsand choose Customize mouse actions for → Trackpad.

As the structure rotates, parts of it become lighter in shade and parts become darker. Thisis a 3D effect called fogging that helps distinguish closer atoms from more distant atoms.



4

5. Click Reset in the Toggle Table panel.

The view of the structure is reset to the original view.

6. Zoom in on the structure:

• If you have a mouse, hold down the right button and drag.• If you have a trackpad, use the pinch gesture or a two-finger swipe.

7. Click Zoom in the Toggle Table panel.

The view of the structure zooms out so you can see all the visible atoms.

8. Click Orient.

The structure is reoriented so that it is centered in the Workspace, and the view is rotatedso that the structure is most spread out in the plane of the screen (the xy plane).

Note: This action changes the coordinates of the structure, so you should only use it whenchanging the coordinates does not matter.

Figure 2.1. The BioLuminate main window with the oriented 2OT3 structure.



In the Toggle Table, there are two rows under Entries in the Workspace, labeled All, and 2OT3.Each row corresponds to a project entry (a structure). The All row is a special row that appliesto all structures in the Workspace.

Each row has a set of buttons for doing operations on the structure. These buttons are actuallymenus:

• A (Action)—perform various actions on the structure• S (Show)—show the structure or parts of the structure in different representations (line,

stick, ball and stick, sphere, ribbon, cartoon)• H (Hide)—Hide various parts of the structure or its representations.• L (Label)—Label the structure or parts of it with various properties.• C (Color)—Color the structure or parts of it with various color schemes.

The next few steps illustrate the use of these menus.

9. In the 2OT3 row, choose S → Sticks.

Bonds are shown in stick (tube) representation.

10. In the 2OT3 row, choose S → As → Lines.

The sticks are no longer shown, and the bonds are represented by lines again.

11. In the 2OT3 row, choose H → Waters.

The water molecules are no longer shown in the Workspace. However, they are stillcounted as being in the Workspace, even though they are not visible.

12. In the 2OT3 row, choose S → Lines.

The water molecules are redisplayed. The representation applies to all atoms in the struc-ture, and any hidden atoms are made visible again.

13. In the 2OT3 row, choose S → As → Ball and Stick.

Atoms are now represented by balls, and bonds by sticks.

14. In the 2OT3 row, choose L → Other Properties → Stereochemistry.

The chiral atoms are labeled R or S. Zoom in for a clearer view, then click Reset to zoomout again.

15. In the 2OT3 row, choose L → Clear.

The labels are removed.

16. In the 2OT3 row, choose C → Color by Element → Custom Color {C}HNOS.

This color scheme colors all atoms by element, and applies a user-selected color to car-bons. A color selector opens so you can choose a color for the carbons.



6

17. Click the gray cell in the lower right.

The name of the color is shown in a tool tip. The structure is colored by element with graycarbons, blue nitrogens, red oxygens, orange sulfurs. For coloring, bonds are split in two,and colored by the color of the nearest atom.

18. In the 2OT3 row, choose C → Custom Color All Atoms.

19. Click the gray cell in the lower right of the color selector.

All atoms are colored gray.

20. In the 2OT3 row, choose C → Color by Spectrum → B Factors (Calpha).

The carbon atoms are colored according to the PDB B factors (temperature factors).

You can experiment further with the representation and color settings.

There are also customizations of color and representation that can be applied to particularkinds of structures, such as protein interfaces. This structure has two chains, and hence aninterface that can be displayed. (You can find the number of chains by looking in the status bar,below the Workspace, for the text Chn:2. The status bar gives information about the Workspacestructure by default.)

21. In the 2OT3 row, choose A → Preset →Protein Interface.

A progress bar is displayed briefly, and then the protein is redisplayed. The atoms in theinterface region are displayed in ball and stick, while the rest of the protein is displayed incartoon. Each chain is displayed in a separate color.

So far the operations have been mostly on the entire structure. To perform operations on part ofa structure, you have to select the atoms you want to work on. You can do this by “picking”atoms in the Workspace, either individually or as part of a group such as a residue, a molecule,or a chain.

22. Choose Edit → Pick Mode → Chains on the main menu bar.

Now, when you pick (click on) atoms in the Workspace, the entire chain that the atom ispart of is selected. The cursor in the Workspace changes to a box with a C next to it, toindicate that you are picking chains.

23. Click on one of the chains in the Workspace.

You can click on the cartoon or on one of the atoms. The atoms are marked with yellowboxes, and a new row is added to the Toggle Table, labeled (Selection). This is the row thatyou can use to perform actions on the selected atoms.



24. In the (Selection) row, choose S → As → Lines.

The chain that you selected is now shown with bonds as lines and atoms as points.

25. In the (Selection) row, choose A → Copy to New Project Entry.

The chain is copied to a new entry in the project, and a new row is added to the ToggleTable, labeled Selection001. This name is chosen as a unique name, but it doesn’t provideany information about the structure. You can rename it.

26. In the Selection001 row, choose A → Rename.

The row name is replaced by a text box.

27. Click in the text box and type 2OT3 chain B.

If the chain you selected is chain A rather than chain B, name it 2OT3 chain A. You cancheck which chain you selected by pausing the cursor over an atom in the Workspace.The identity of the atom is shown in the status bar below the Workspace, and includes thechain name, residue name and number, and atom name.

Figure 2.2. The BioLuminate main window with the selected chain.



8

28. In the (Selection) row, choose A → Modify → Invert → Within Any.

The selection is inverted: all the atoms that were selected are now not selected, and theatoms that were not selected are selected. In other words, the other chain is now selected.

29. In the (Selection) row, choose A → Copy to New Project Entry.

The chain is copied to a new entry in the project, and a new row is added to the ToggleTable, labeled Selection002. If you wanted to remove the chain from the original proteinrather than copying it, you could choose Extract to New Project Entry instead.

30. In the Selection002 row, choose A → Rename.

The row name is replaced by a text box.

31. Click in the text box and type 2OT3 chain A.

Use B instead of A for the chain name if appropriate.

You have now extracted the two chains from this protein into separate project entries.Each of these is included in the Workspace, as well as the original protein.

32. In the 2OT3 row, choose A → Remove from Workspace.

The original protein is no longer in the Workspace, and its row is no longer in the ToggleTable. The (Selection) row has also gone, because the selected atoms were all in the orig-inal protein.

33. Click the title of the 2OT3 chain A row in the Toggle Table.

The atoms of this entry are hidden. As previously, the atoms are still considered to be inthe Workspace, even if they are not visible.

34. Click the title of the 2OT3 chain A row in the Toggle Table again.

The atoms of this entry are displayed again.

35. Click Include in Workspace.

The Include in Workspace dialog box opens, listing all the entries in the project by titleand entry ID. Entries that are already included in the Workspace have (Included) at thebeginning of the row.

36. Select the 2OT3 row and click OK.

The entry is included in the Workspace and the Toggle Table again. The selection youmade earlier is not restored, because selection applies only to the Workspace, and onlywhile the selected atoms are in the Workspace, and the selection was removed when thisentry was removed from the Workspace.



37. In the All row, choose A → Remove Everything from Workspace.

All of the atoms are now removed, and only the All row remains in the Toggle Table.

The exercises are complete, and the final step is to clean up by removing all the structuresand closing the project. The project you have been working in is a temporary project,called a scratch project.

38. Choose File → Close Project, and click Discard in the Save Scratch Project dialog box.

The project is closed and removed, and a new scratch project is opened. If you want tosave and close the project instead of discarding it, you can click Save in the Save Scratch

Project dialog box, then name the project in the file selector that opens. The newly savedproject is closed. You can open it again from the File menu when you want to use it.

You can convert a scratch project to a named, saved project at any point by choosing File

→ Save As. Project data is automatically saved when it is updated, so you don’t need tokeep saving a project to preserve your data.

For further information on the functions of the BioLuminate interface, you can use the helpfacility, either from the Help menu or by clicking the Help buttons.

Figure 2.3. The Include in Workspace dialog box


10


Chapter 3

Chapter 3: Examining Sequences

This exercise demonstrates the use of the Multiple Sequence Viewer to align, compare, andexamine the sequences for a small set of homologous proteins.

The first stage of the exercise is to import a single chain from each of three proteins.


The Get PDB File dialog box opens.

2. Enter 3max in the PDB ID text box.

3. Enter B in the Chain name text box.

When you specify a chain, only that chain is imported. If you leave the Chain name textbox blank, the entire protein is imported. To import a chain with a blank name, use ' '.

4. Click Download.

An information box is displayed, indicating that there are problems in the structure. Sincethis exercise is concerned largely with sequences, the problems can be ignored.


Chain B of the 3MAX structure is imported into the project and displayed in theWorkspace.


7. Enter 1t64 in the PDB ID text box and click Download.

The Chain name text box still contains B. As chain B of 1T64 is the chain that we want,the text in this text box does not need to be cleared or modified.

Figure 3.1. The Include in Workspace dialog box



12

As for the previous protein, an information box opens, and the problems it describes canbe ignored.


Chain B of the 1T64 structure is imported into the project and added to the Workspace.


10. Enter 1c3r in the PDB ID text box and click Download.


Chain B of the 1C3R structure is imported into the project and added to the Workspace.

All three proteins are now in the Workspace.

12. Choose Tools → Multiple Sequence Viewer.

The Multiple Sequence Viewer panel opens, showing the sequences for the three proteins.The sequences are colored by residue type, and each sequence is annotated with its sec-ondary structure assignment (labeled SSA on the left). Note that all the sequences start atthe left side and have no gaps.

Figure 3.2. The Multiple Sequence Viewer panel, initial view.



The next step is to align the three sequences, with 3MAX as the query.

13. Right-click on 3MAX_B in the left pane and choose Set as Query Sequence.

This sequence is placed at the top of the sequence list, and it is marked in blue to indicatethat it is the selected sequence. It is also the query sequence, and its identity is reported inthe status bar at the bottom of the panel, along with information on the number ofsequences.

The shortcut menu has a number of actions that you can perform on the sequence.

14. Shift-click on the last sequence in the left pane.

The three sequences are now selected. The alignment is performed on the selectedsequences.

15. Click the Multiple Alignment toolbar button.

A dialog box is displayed briefly as the alignment is done.

When the alignment finishes, you can see that there are now gaps in the displayedsequences. It is a little easier to compare the sequences if the annotations are removed.

16. Choose Annotations → Clear Annotations.

The secondary structure assignment is removed. You can also hide annotations using theturner for the tree view in the left pane.

17. Move the pointer slowly over one of the gaps, so that you can see the tooltip informationfor each residue.

The position in the sequence viewer is given first, followed by the 3-letter residue nameand number. The residue numbers have not changed, only the position in the viewer.

18. Click the Color Matching Residues Only toolbar button.

The coloring of the second and third sequences changes so that only the residues thatmatch the query sequence are colored; the rest are given a white background. The querysequence remains colored. This tool makes it easy to identify the residues that match thequery.



14

19. Click the Weight Colors by Alignment Quality toolbar button.

The intensity of the coloring changes to reflect the fraction of matching residues at eachresidue position. Where all three sequences have the same residue, the color intensity is atits maximum; where only two residues match at a position, the color is medium, andwhere only one matches (in the query sequence) the color is light. This tool makes it easyto identify the positions that have the most matches.

The Multiple Sequence Viewer has a number of useful annotations that can be displayed belowthe sequence.

20. Choose Annotations → Ligand Contacts.

The structure is analyzed to find the contacts of the ligand with the protein. After a while,the annotation is displayed. The ligand is represented by a row, with positions marked inred where there are close (< 4 Å) contacts between heavy atoms, and in orange for othercontacts (< 6 Å). The rest of the row is gray.

Figure 3.3. Coloring of matching residues.



21. Choose Annotations → B Factor.

The temperature factor is shown as a bar chart below the sequence.

The Multiple Sequence Viewer contains many more features, for manual alignment and editingof sequences, selection of residues, homology modeling, and prediction of structural features.For more information, see the online help or the Multiple Sequence Viewer document.

A restricted version of the Multiple Sequence Viewer is used in many of the BioLuminatepanels.

22. Close the Multiple Sequence Viewer panel.


The project is closed and discarded.

Figure 3.4. Ligand contact and B-factor annotations.


16


Chapter 4

Chapter 4: Identifying Reactive Residues

This exercise demonstrates how to identify reactive residues in a protein, which is importedfrom the PDB. The identification is done by matching residue patterns in the sequence.

1. Choose Tools → Protein Preparation.

The Protein Preparation Wizard panel opens. This panel is available from both the Toolsmenu and the Tasks menu.

2. Enter 2pcy in the PDB text box, and click Import.

The 2PCY structure is imported into the project and displayed in the Workspace. Thispanel provides a wide range of options for preparing proteins for modeling: fixing struc-tural defects and making assignments required by modeling applications. For thisexercise, the default settings will be used.

Figure 4.1. The Protein Preparation Wizard panel.



18

3. Click Preprocess.

The structure is processed, and a new structure is added to the project and displayed in theWorkspace, replacing the imported structure.

You can close the Protein Preparation Wizard panel.

4. Choose Tools → Reactive Residue Identification.

The Reactive Protein Residues panel opens. The predefined reaction types are selectedby default. You can define your own (click Edit), but for this exercise the defaults will beused.

5. Click Analyze Workspace.

The structure in the Workspace is analyzed to identify residues that match the patternsdefined for deamidation, oxidation, glycosylation, and proteolysis. The results are listedin the table, and the sites are marked with spheres in the Workspace. The spheres are col-ored according to the reaction type, and the color legend for the spheres is given belowthe table. Also below the table are some tools for filtering the residues shown in the tableand in the Workspace.

Figure 4.2. The Reactive Protein Residues panel.



6. Enter 80 in the Show only residues with solvent exposure >= text box.

The list is reduced to five residues, two proteolysis sites and three glycosylation sites,which have a high solvent exposure.

7. Click the row with 100% exposure (the row numbered 12).

The view zooms in to this residue, whose side chain has no contact with the rest of theprotein.

8. Set the solvent exposure threshold back to 15.

9. Choose Oxidation from the Show option menu.

Only one residue is listed in the table.

10. Choose All from the Show option menu.

11. Click on the heading of the Exposure column.

The residues are sorted by exposure. The oxidation site is the one that has the lowest sol-vent exposure.

12. Close the Reactive Protein Residues panel.

Figure 4.3. Reactive residue sites in 2PCY.



20





Chapter 5

Chapter 5: Identifying Consensus Waters

This exercise demonstrates the use of the Consensus Viewer to identify conserved non-proteinstructural elements in a set of homologous proteins—in this case, waters.


The Protein Preparation Wizard panel opens.

2. Enter 2pcy in the PDB text box, and click Import.

The 2PCY structure is imported into the project and displayed in the Workspace.

3. Deselect Delete waters beyond N Å from het groups.

This exercise uses the waters, so they should not be deleted.

Figure 5.1. The Protein Preparation Wizard panel.



22



You can close the Protein Preparation Wizard panel.

5. Choose Tools → Protein Consensus Viewer.

The Consensus Visualization panel opens.

6. Click Import and choose From Workspace.

The structure in the Workspace is imported into the sequence viewer in the panel.

7. Click Find and Align Homologs.

The Blast Search Settings dialog box opens.

8. From the Database option menu, choose NCBI PDB (non-redundant).

9. Choose a BLAST Server option.

Figure 5.2. The Consensus Visualization panel after import.



If you have a local installation of the BLAST database, you can choose Local. Otherwise,choose Remote (NCBI).

10. Click Start Job.

A Job Progress dialog box replaces the Blast Search Settings dialog box, and displaysthe log file from the BLAST search.

After a few minutes, the job finishes and the BLAST Search Results dialog box opens,with the results of the search. The top ten results are selected in the table.

11. Click Incorporate Selected Rows.

If you do not have a local installation of the BLAST or PDB databases, the search is doneon the web, and a warning is displayed: “Multiple Sequence Viewer is attempting toaccess a remote server. Would you like to continue?” You can select Do not ask this ques-tion again, to prevent it from opening each time a structure is downloaded, and click OK.

If an information box opens stating that problems were found when importing a structure,you can select Do not show this dialog again to prevent it from opening for each structurethat has problems, and click OK.

Figure 5.3. The BLAST Search Results dialog box.



24

The ten homologs are added to the sequence viewer and displayed in the Workspace. Allatoms are selected in all of the homologs.

12. In the Water section, choose Consensus from the Display option menu.

The consensus water structures are displayed with the proteins in the Workspace, repre-sented as spheres and outlined. In this case there is only one consensus water structure,because consensus is defined as having a water at this position in at least 60% of struc-tures. This percentage is specified under Define consensus.

13. Change the percentage under Define consensus to 50.

More consensus water structures are shown, at several locations. At some locations, thespread in the positions of the waters is greater than others. Some spheres have hydrogensin white, others do not. This is because hydrogen atoms are present on the waters in the2PCY structure, but are not present in the other structures.

14. In the All row of the Toggle Table, choose A → Remove Everything from Workspace.

Figure 5.4. The Consensus Visualization panel with results.



15. In the Consensus Visualization panel, click the check box for 2PCY_A in the sequenceviewer.

The 2PCY protein is displayed, with its consensus waters. The color scheme (red oxygen,white hydrogen) was set by the protein preparation process, and is not altered by thispanel.

16. Click the check box for 1JXG_A in the sequence viewer.

The 1JXG protein is displayed with its consensus waters, along with 2PCY. The watersfor 1JXG are represented by gray spheres for the oxygen atoms. There are no hydrogenson the 1JXG waters, and the structure is colored gray because it has only been imported:no protein preparation has been done on this protein. The spheres overlap with the watersfor 2PCY at four of the five water locations.

17. Click the check box for 2PCY_A again in the sequence viewer.

The 2PCY protein is removed from the Workspace, and only the 1JXG protein remains,with its consensus waters shown as gray spheres.

The identity of any consensus water can be ascertained by moving the cursor over theconsensus water sphere and viewing the text in the status bar, below the Workspace.

18. Close the Consensus Visualization panel.




26


Chapter 6

Chapter 6: Homology Modeling of Proteins

This exercise demonstrates how to run a simple homology model of a protein. A structure fromthe PDB is used, and a template is chosen that represents a realistic model for an unknownprotein. If you are interested in homology modeling of antibodies, see Chapter 9.

1. If the Workspace is not empty, in the Toggle Table panel click the A button in the All rowand choose Remove Everything from Workspace.


3. Enter 2pcy in the PDB ID text box, and click Download.

The 2PCY structure is displayed in the Workspace.

4. Choose Tasks → Homology Modeling → Simple Homology Modeling in the main window.

The Homology Model panel opens.

5. Click the Query button and choose From Workspace.

The box to the right of the button displays the text ( Query structure ) and the title of theWorkspace structure.

Figure 6.1. The Homology Model panel, initial view.



28

6. Click the Template button and choose BLAST Homology Search.

The Blast Search Settings dialog box opens. This dialog box allows you to changeparameters of the BLAST search. For this exercise (and for most purposes), the defaultscan be used.

7. Click Start Job.

A Job Progress dialog box replaces the Blast Search Settings dialog box, and displaysthe log file from the BLAST search.

After a few minutes, the job finishes and the BLAST Search Results dialog box opens,with the results of the search. The best homolog is selected. For this exercise, we willchoose one of the lower-ranking homologs to illustrate what would happen in a real situ-ation with an unknown structure.

8. Scroll down to 1M9W_A (row 21) and select it.

This homolog has an identity and a similarity that are more typical for an unknownstructure.

Figure 6.2. The BLAST Search Results dialog box.



9. Click Select Template.

If you do not have a local installation of the BLAST or PDB databases, a warning is dis-played by default: “Multiple Sequence Viewer is attempting to access a remote server.Would you like to continue?” If you have not already turned this warning off, Select Do

not ask this question again, to prevent it from being displayed each time a structure isdownloaded, and click OK.

The template is selected and the BLAST Search Results dialog box closes. The table rowse in the Homology Model panel are filled in with information on the template. Note therankings: the similarity and homology are considered good, but the identity is only fair.

10. Click Generate Models.

The Homology Model - Start dialog box opens. You can name the job and select a host torun it on. The job includes alignment of the template and the query using ClustalW, andthe structure is built on the basis of the template and an analysis of structural elements inthe PDB for non-templated regions (a “knowledge-based” selection of the coordinates).

After a few minutes, the job finishes. The model is added to the Workspace. The cartoonis colored by how the template was used: dark blue for residues for which all coordinateswere taken from the template, cyan for residues for which the backbone was taken fromthe template, and red for residues that were entirely modeled, not using the template.

The buttons in the Homology results section of the Homology Model panel are nowavailable.

Figure 6.3. The Homology Model panel after selecting the template.



30

11. Click Examine Model Quality.

The Protein Structure Quality Viewer panel opens, with the Ramachandran Plot tab dis-played. The model is the only structure in the Workspace, and the residues are colored bythe plot region that they are in.

12. Pause the pointer over a point in the disallowed region.

The identity of the residue and its angles are displayed at the top right of the plot, and theresidue is selected and highlighted in the Workspace. If you move the pointer off thepoint, the residue is no longer highlighted, and the Selection row in the Toggle Table isdimmed, indicating that it is no longer selected.

Figure 6.4. Protein Structure Quality viewer with point selected in Ramachandran Plot, and Workspace showing selected residue.

Selected point



13. Now click on the same point.

The view zooms in to show the residue centered in the Workspace with its nearest neigh-bors. This happens when you click on a point but not when you pause over a point. Theselection in the Selection row is now persistent: it remains until you make a new selec-tion. If you pause over a different point, its details are shown to the top right of the plotand it is highlighted and selected in the Workspace (although it may be off screen), butwhen you move the pointer off the new point, the details of the point you clicked on aredisplayed again and it is highlighted and selected again.

14. In the Protein Report tab, choose Backbone Dihedrals from the Display option menu.

The table is populated with values of the backbone dihedrals and the derived G-factor,which measures the likelihood of the combination of dihedrals, and the G-factor is plottedbelow the table for all the residues in the protein.

Figure 6.5. Protein Structure Quality viewer with protein report for backbone dihedrals, and Workspace showing residues selected in dihedral plot.



32

15. Move the pointer across the plot, and pause over one of the large negative peaks.

The identity of the residue for each plot point is displayed to the top right of the plot withits G-factor. For the large negative peaks, the text “disallowed” is displayed instead of theG-factor.

16. Drag the dotted blue vertical lines of the plot inward so that they are on either side of thethree large peaks above row number 50.

17. Drag the top dotted red line down so that it is just below the –9 mark on the vertical axis.

As you drag, the residues in the area between the blue and red lines are selected, and theWorkspace view zooms to these residues. On the plot, the background of the area outsidethe rectangle bounded by these lines is colored gray. This is a way of selecting residuesthat have particular values of a property.

18. Close the Protein Structure Quality Viewer panel.

19. In the Selection row of the Toggle Table choose A → Delete Selection.

The selection row disappears, indicating that there is now no selection of atoms in theWorkspace.

To further check the accuracy of the model, it can be visually compared with the X-ray struc-ture in the Workspace. The homology model should be the only structure in the Workspace. Itis titled Model of 2PCY built on 1M9W.


The Include in Workspace panel opens, listing the titles and IDs of the project entries thatyou can include.

21. Choose the 2PCY entry and click OK.

The X-ray structure is now in the Workspace.

22. In the All row of the Toggle Table choose C → Auto → All Atoms by Object.

23. In the All row of the Toggle Table choose S → As → Cartoon.

24. On the main menu bar, choose Tools → Protein Structure Alignment.

The Protein Structure Alignment panel opens below the Toggle Table. The default settingsshould be appropriate for this task:

• Use proteins from is set to Workspace (included entries).• The text box in the Reference residues section contains all.• Use same ASL as reference residues is selected under Residues to align.



25. Click Align.

After a short while, the proteins are aligned. The alignment score and the RMSD arereported at the bottom of the Results text area. The Workspace should look somethinglike Figure 6.6. If the structures are not visible, click the Zoom button in the Toggle Table.

26. Close the Protein Structure Alignment panel.

27. Rotate the structure around to compare the X-ray structure (green in Figure 6.6) with themodel (cyan).

Overall agreement between the predicted model and the X-ray structure is generally verygood. Differences in the secondary structural elements between the two are mostly small,with the exception of the loop formed by residues 42-52. This is not unexpected, giventhe lack of template residues (a sequence alignment gap) between residues 47 and 50. Animproved model for this loop could be set up and run in the Refinement panel by clickingRefine Loops in the Homology Model panel. Loop prediction can take a long time, so it isnot included as part of this exercise.

28. Close the Homology Model panel.



Figure 6.6. The model and the X-ray structure of 2PCY.


34


Chapter 7

Chapter 7: Scanning for Residue Mutations

This exercise demonstrates how to scan a protein for potential residue mutations and comparethe relative binding affinity of two protein chains for each of the mutants. The protein mustfirst be prepared, as only two of the chains will be kept for this exercise.




3. Enter 1brs in the PDB text box, and click Import.

The 1BRS structure is imported into the project and displayed in the Workspace. For thisexercise, the unwanted chains will be removed first, then the structure will be processed.

Figure 7.1. Deleting chains.

1. Click this button.


2. Select chains.



36

4. In the Review and Modify tab, click Analyze Workspace.

The tables of chains, waters, and het groups are filled in. This is done automatically whenyou process the structure in the Import and Process tab.

5. Select chains B, C, E, and F in the table of chains.

6. Click Delete.

The selected chains are deleted, leaving just chains A and D. Now the trimmed structurecan be processed.

7. In the Import and Process tab, select Fill in missing side chains using Prime.

This protein has some side-chain coordinates missing in the PDB structure. Adding themat this stage performs a quick addition of the side chains. Refinement of their locationswill be done later.

8. Click Preprocess, and confirm the addition of the side chains if prompted.


Figure 7.2. Preprocessing the protein with side-chain addition.


1. Select this option.



The Protein Preparation - Problems dialog box opens, reporting an atom clash for a lysinechain on the surface of the protein. This problem will be fixed later in the process.

9. Click OK to dismiss the Protein Preparation - Problems dialog box.

10. In the Refine tab, click Optimize in the H-bond assignment section.

Optimizing the hydrogen bonding is important because X-ray structures do not usuallyhave enough resolution to fix the orientation of terminal amides or histidines, or the ori-entation of hydroxyls and thiols.

When the optimization finishes, a new structure is added to the project. The structure islabeled with the changes that were made to terminal amides and histidines, to make iteasy to examine these changes if you wish.

11. In the 1BRS row of the Toggle Table, choose L → Clear.


Figure 7.3. Refining the protein structure.





38

12. In the Protein Preparation Wizard panel, click Minimize.

A restrained minimization is done, which removes the atom clash and relaxes the sidechains and other modifications made to the protein. The results are added as a new entryto the project. You can click View Problems to verify that the clash has gone. The proteinis now ready for use.

13. Choose Tasks → Residue Scanning → Perform Calculations in the main window.

The Residue Scanning panel opens.


A dialog box opens, prompting you to choose a chain. The default is chain A.

Figure 7.4. Setting up the residue scanning job.



2. Select residues.


4. Choose method.



15. Click OK to accept chain A for analysis.

The structure in the Workspace is analyzed and the residues table is filled in with all theresidues in chain A.

16. Select the rows for the following residues:

A:27 (LYS), A:54 (ASP), A:58 (ASN), A:59 (ARG), A:60 (GLU), A:73 (GLU), A:87 (ARG),A:102 (HIS).

17. Check that the residue in the For all selected rows, set mutations to is ALA, and clickApply.

The selected rows are marked for mutation and the mutation is set to alanine. The textmessage at the bottom of the tab should read Will mutate 8 of 108 residues; Number of

mutations: 8.

18. From the Refinement option menu, choose Side-chain prediction (bb).

This option allows for some movement of the backbone in response to the change in sidechain. The rest of the settings can be left as they are.

19. Click Start to open the Start dialog box.

20. Choose a host from the Host option menu.

If you chose a multiprocessor host, you can distribute the job over multiple processors.For this job, you can specify a maximum of 8 subjobs, because there are 8 mutations, andyou can also specify up to 8 processors.

21. Click Start to start the job.

A status bar showing the progress of the job is displayed above the Start button. The jobtakes about an hour to run on a single processor.

When it finishes, the Residue Scanning Viewer panel opens, displaying changes in prop-erties for each mutation, and a graph of one of the properties.

22. Choose Δ Affinity from the Graph property option menu.

The graph shows the change in affinity of the mutated chain for the other chain on muta-tion, as a function of the row number in the table. Note that some mutations improve theaffinity between the chains, while others destabilize the complex.

23. Close the Residue Scanning panel and the Residue Scanning Viewer panel.





40

Figure 7.5. The Residue Scanning Viewer panel.



Chapter 8

Chapter 8: Locating Possible Mutations for Disulfide Bridges

This exercise demonstrates how to run a cysteine mutation calculation to locate and rankpossible disulfide bridges. A structure from the PDB is used that already has a disulfide bridge.After preparation of the protein, the cysteine residues are mutated by hand to alanine, and theCysteine Mutation panel is then used to locate and restore the bridge.



3. Enter 1vhu in the PDB text box, and click Import.

The 1VHU structure is imported into the project and displayed in the Workspace. Thisstructure is missing a side chain, which can be filled in.

4. Select Fill in missing side chains using Prime.

5. Click Preprocess, and confirm the addition of the side chains when prompted.


6. In the Refine tab, click Optimize in the H-bond assignment section.

Optimizing the hydrogen bonding is important because X-ray structures do not usuallyhave enough resolution to fix the orientation of terminal amides or histidines, or the ori-entation of hydroxyls and thiols.

When the optimization finishes, a new structure is added to the project. The structure islabeled with the changes that were made to terminal amides and histidines, to make iteasy to examine these changes if you wish.

7. In the 1VHU row of the Toggle Table, click L and choose Clear.


8. In the Protein Preparation Wizard panel, click Minimize.

A restrained refinement is done, which relaxes the side chains and other modificationsmade to the protein. The results are added as a new entry to the project.



42

9. If the sequence viewer is not already displayed at the bottom of the Workspace, chooseEdit → Settings → Show Sequence Viewer on the main menu bar.

You might need to make the sequence viewer bigger so you can see the entire sequence.To do so, drag the “handle” (dotted line) just above the sequence in an upward direction.

10. If the Find toolbar is not already displayed, press CTRL+F (F).

11. Choose Residue type from the first option menu, then choose CYS from the next optionmenu.

12. Click the N button or press the N key.

The Workspace view zooms in to the first cysteine (Cys 111), which is one of the pair thatforms a disulfide bridge. The cysteine is selected, and highlighted in the Workspace andthe sequence viewer.

13. Right-click on the cysteine in the sequence viewer and choose Mutate Residue → ALA.

14. Click on the mutated residue.

Figure 8.1. Refining the protein structure.





15. Click the N button or press the N key again.

The Workspace view zooms in to the second cysteine (Cys 154), which is selected andhighlighted in the Workspace and the sequence viewer.

16. Right-click on the cysteine in the sequence viewer and choose Mutate Residue → ALA.

A further refinement of the protein should now be done to relax the mutated cysteines.

17. In the Refine tab of the Protein Preparation Wizard panel, click Minimize.

When the job finishes, you can close the Protein Preparation Wizard panel.

18. Choose Tasks → Cysteine Mutation.

The Cysteine Mutation panel opens.

Figure 8.2. The Run tab of the Cysteine Mutation panel.



44


The structure in the Workspace is analyzed to identify residues that could be mutated toform or to break disulfide bonds. There are only two pairs that involve a single cysteine,and the rest involve two non-cysteine residues, including the pair you mutated.

Because the protein has no disulfide bridges, there are no candidates for removing thebridge by mutation, but the analysis identifies such Cys-Cys pairs if they are present.

20. Place the pointer in the table and press CTRL+A (A).

All of the table rows are selected, and the selection is noted below the table.

21. Click Start to open the Start dialog box, and click Start in the dialog box to start the job.

There is no need to change the default values in the dialog box. The job takes about halfan hour. Its progress is displayed in a progress bar above the Start button. When the jobfinishes, the results are displayed in the Results tab.

Figure 8.3. The Results tab of the Cysteine Mutation panel.



22. Click on the heading of the Δ Ei column.

The residue pairs are sorted by the change in interaction energy on mutation. The originalcysteine pair, A:111-A:154 is ranked near the top, with a substantial negative energy,which means that the mutation is favorable for stabilizing the protein.

23. Click the A:111-A:154 row.

The residue pair is displayed in the Workspace, colored by element and represented inball-and-stick. The original protein is also present, colored gray.

All of the mutated structures are included in the project. The mutation is identified by the res

pair property. You can view the properties in the Project Table, which you open by pressingCTRL+T (T) in the main window.

24. Close the Cysteine Mutation panel.




46


Chapter 9

Chapter 9: Modeling an Antibody

This exercise demonstrates how to model an antibody by homology modeling, using a databaseof antibody structures to locate homologs, and then humanize the antibody. It uses a knownantibody structure, 1fsk, which is removed from the antibody database so that it can be treatedas an unknown antibody, and the model can be compared with the crystal structure.

1. Choose Tasks → Antibody Modeling → Prediction in the main window.

The Antibody Prediction panel opens. On the left is a labeled diagram of an antibody; onthe right are three tabs for setting up and performing the modeling.

2. Click the heavy variable region of the diagram and choose From PDB ID.

This region is marked when you pause the pointer over it. When you click, a menu is dis-played—see Figure 9.1. When you choose the menu item, the Enter PDB ID dialog boxopens.

3. Type 1fsk in the text box and click OK.

The protein is imported and analyzed. The PDB structure is a dimer. After a while, theChoose Heavy Region dialog box opens, prompting you to choose the chain to use.

4. Choose chain C from the menu and click OK.

The heavy variable region is colored green to indicate that it has been assigned.

5. Click the light variable region of the diagram and choose From Selected Entries in the

Project Table.

The PDB sequence and structure has already been imported, so you can use it for the lightchain. It is selected in the Project Table by default.

The protein is analyzed for light chains. After a while, the Choose Light_kappa Region

dialog box opens, prompting you to choose the chain to use.

6. Select chain B and click OK.

The light variable region is colored blue to indicate that it has been assigned, and the textprompting you to import the regions is removed—see Figure 9.1.

Figure 9.1. (next page) The Antibody Prediction panel. Top: Diagram showing chain label. Center: Import menu. Bottom: After importing both chains.



48


The next task is usually to select the database to use in the search for homologs for the frame-work region. In this tutorial, the default database from the installation is used, so no action isneeded to select a database. However, the antibody you imported must be removed from thedatabase search, so the modeling of an unknown antibody can be simulated.

7. Click Define Search Limits.

The Limit Criteria Definition dialog box opens. It allows you to limit the search by filteringthe database on the values of one or more properties. The property you will use for thisexercise is the PDB ID.

8. In the Property text box, type P, then choose PDB ID from the completion list that is dis-played.

9. From the option menu next to it, choose Not =.

10. In the next text box, type 1FSK.

Make sure you use upper case for the PDB ID here.

11. Click Add.

The Filtering definitions and criteria area displays the criterion you just defined, and alsoreports the number of matches to the filter, 1130, which is one fewer than the 1131 struc-tures in the full antibody database.

12. Click OK.

Figure 9.2. The Limit Criteria Definition dialog box.



50

You can now proceed with the search for a template for the framework region.

13. Click Search.

When the search finishes, the table is filled in with the results. 1fsk is not in the list,because it was removed by the filter. The highest-scoring result is 1i3g, which is selectedby default, and will be used for this exercise.

14. Click Accept.

The template for the selected row in the results table is imported into the project. After ashort while, the Basic Loop Model tab is displayed.

Figure 9.3. The Antibody Prediction panel after searching for templates.

Figure 9.4. The Antibody Prediction panel, Basic Loop Model tab.



In this tab you can choose whether to model the loops using the antibody database or usethe input coordinates for the loops. If you model the loops, you can select the cluster ofloop conformers you want to use for each loop that you model by clicking View Clusters,and choosing a cluster in the Loop Clusters panel. By default the largest cluster of loopsof the appropriate length is chosen automatically, and the loop from this cluster that ismost similar to the query is used. In this exercise you will use the defaults, whichincludes generating a single model.

15. Click Generate Initial Models.

After a few minutes, the models are generated, and the Model to view menu is populated.This model can be used as it is, even though it is called an initial model. In real situations,you might want a more accurate prediction of the H3 loop, which you can set up and runin the Advanced Loop Model tab.

To check the accuracy of the homology model, you will compare this structure with the X-raystructure. To do this, the X-ray structure needs to be pruned down to the size of the modeledstructure. This will be done using the Workspace sequence viewer.

16. If the Workspace is not empty, in the All row of the Toggle Table panel choose A →Remove Everything from Workspace.



18. Choose the first 1FSK entry and click OK.

The X-ray structure is now in the Workspace.

19. If the sequence viewer is not already displayed, choose Edit → Settings → Show

Sequence Viewer on the main menu bar.

20. In the sequence viewer, click the first residue in chain B, then scroll down and shift-clickthe last residue in chain B.

Chain B is now selected. Yellow markers are displayed in the Workspace on the chain andthe chain is highlighted in the sequence viewer.

21. Control-click the last residue in chain C, then scroll up and shift-control-click the firstresidue in chain C.

Both chains B and C should now be selected.

22. Right-click on the selection in the sequence viewer and choose Invert Selection.

The selection now includes all chains but B and C.



52

23. Press the DELETE key. (Small keyboard Mac: FN+DELETE.)

The selected chains are deleted and only chains B and C remain. These are the chains thatwere modeled, but they are longer than the chains in the model.

24. In the Toggle Table panel, click Include in Workspace.

25. Choose the Antibody prediction (model 1) entry and click OK.

Both structures are now in the Workspace.

26. In the sequence viewer, scroll to the last residue in chain L of the model.

This should be Ile106. Chain L has the same sequence as chain B in the X-ray structureup to this point. Chain B has more residues, which will be deleted.

27. Click the next residue in chain B (Lys107).

28. Scroll to the end of chain B and shift-click the last residue.

29. Press the DELETE key.

30. Scroll to the last residue in chain H of the model.

This should be Ser112. Chain H has the same sequence as chain C in the X-ray structure,but the numbering is slightly different because chain H comes from the homology model,and uses the numbering of the template, with insertion codes for any insertions. The cor-responding residue in chain C is Ser117.

31. Click the next residue in chain C (Val118).

32. Scroll to the end of chain C and shift-click the last residue.

33. Press the DELETE key.

The X-ray structure is now pruned down to the same size as the model, and you can nowperform a structural alignment, to see how much the two structures differ. To facilitate thecomparison, some changes must be made to the display.

34. In the All row of the Toggle Table, choose C → Auto → All Atoms by Object.

35. In the All row, choose S → As → Cartoon.


The Protein Structure Alignment panel opens below the Toggle Table. The default settingsshould be appropriate for this task:

• Use proteins from is set to Workspace (included entries).• The text box in the Reference residues section contains all.• Use same ASL as reference residues is selected under Residues to align.



37. Click Align.

After a short while, the proteins are aligned. The alignment score and the RMSD are reportedat the bottom of the Results text area, and both indicate that the alignment is good. The Work-space should look something like Figure 9.5. Note the patchy appearance of the ribbons: this isbecause they are almost in the same location, indicating that the model is very good. Likewise,the loops of the model follow the loops of the X-ray structure closely.

38. Close the Protein Structure Alignment panel and the Antibody Modeling panel.

39. If it is still displayed, select the legend in the Workspace and click DELETE.

The final exercise is to humanize the antibody, which involves selection of residues to mutate.The residues can be selected on the basis of homology or structural criteria.

Figure 9.5. Results of the comparison.



54

If you do not want to continue, you can close the project and discard it. Otherwise, continuewith the following steps.

40. In the 1FSK row of the Toggle Table, choose A → Remove from Workspace.

The truncated X-ray structure is removed, and the antibody model remains.

41. Choose Tasks → Antibody Modeling → Humanization.

The Humanize Antibodies panel opens, with the Humanization Criteria tab displayed.


After a short delay, the light chain is displayed in the sequence viewer, annotated with itsdisulfide bonds and secondary structure assignment. If you want to limit the analysis to apart of the antibody, you can select the relevant part in the Workspace first, and thenselect Analyze only selected Workspace residues before clicking Analyze Workspace.

Figure 9.6. The Humanize Antibodies panel, Humanization Criteria tab.



43. Click Search Antibody Database for Homologs.

The progress of the search is shown at the bottom of the panel. When it is done, thesequence viewer includes the results of the search. Next, the homologs must be aligned sothat selection of residues can be done on the basis of related residue positions.

44. Click Align Homologs.

The homologs are aligned. You may see gaps in the sequences in the sequence viewer.These are added as part of the alignment, but they do not affect the original sequences orthe structures.

The lower half of the tab contains tools for selecting the regions that are searched for pos-sible mutations and criteria for selecting residues by variability or conservation amongthe homologs, or by solvent accessibility or contact with other residues. In this exercise,the default regions (CDR) and selection criteria will be used.

45. Click Select in the Residues Tab.

The residues that match all the criteria are selected for mutation, and a message is dis-played indicating how many residues were selected.

46. In the Residues tab, scroll through the table to locate the selected residues.

Five residues are selected, four in the heavy chain and one in the light chain. Three of theresidues selected in the heavy chain are adjacent.

For each selected residue, mutations are specified based on the variability in the homolo-gous proteins. The total number of mutations is reported lower in the panel. You canchange them by clicking in the table cell, and selecting the residues from the menu that isdisplayed in the cell. Click again in the table (e.g. on the row number) to dismiss themenu. For this exercise, no changes will be made to the mutations.

The mutated structures are refined as part of the mutation job. You can specify the rangeof the refinement around the mutation and the refinement method. For this exercise, thedefaults will be used.

47. Deselect Calculate binding affinity.

Normally, this would be used to calculate the binding affinity of the mutant antibodies tothe antigen, which must be part of the Workspace structure. In this exercise, the antigen isabsent, so it does not make sense to calculate binding affinities.

48. Click Start to open the Start dialog box.

You do not need to change any of the settings.



56

49. Click Start in the dialog box to run the job.

The job starts, and progress is reported at the bottom of the panel. The job takes abouthalf an hour. When it finishes, the Humanize Antibody Viewer opens to display the results,and the mutated structures are added to the project. The results viewer lists the mutationsin a table, with changes in various properties due to the mutation. It also displays a plot ofone of these properties as a function of the table row number.

50. Choose Δ Stability from the Graph property option menu.

A plot of the change in stability with respect to the parent structure is displayed. One ofthe mutations significantly increases the stability, one significantly decreases the stability,and the rest have small or moderate effects.

51. Close the Humanize Antibody Viewer panel.

Figure 9.7. The Humanize Antibodies panel, Residues tab.





Figure 9.8. The Humanize Antibody Viewer panel.


58


Chapter 10

Chapter 10: Protein-Protein Docking

This chapter provides exercises that demonstrate how to run protein-protein docking calcula-tions. One protein is treated as the “receptor” and the other as the “ligand”. In the general case,it does not matter which protein is treated as the receptor and which protein is treated as theligand. For antibody-antigen docking, the receptor is the antibody and the ligand is the antigen.The docking is performed as a rigid-body optimization: there is no subsequent minimization ofthe interfacial region.

Two examples are given here: a general docking example, and an antibody-antigen dockingexample. Structures are taken from the PDB. Protein-protein docking calculations take severalhours to run.

If the Workspace is not empty, in the Toggle Table panel click the A button in the All row andchoose Remove Everything from Workspace.

10.1 General Protein-Protein Docking

The receptor used in this example is 1qqu, which is a single chain. The ligand is chain B of1ba7. The reference structure is 1avx, whose chains differ a little from those of the receptorand the ligand, but not in the binding region. 1qqu differs from chain A of 1avx by two pointmutations on the opposite side of the protein from the binding region. Chain B of 1ba7 differsfrom chain B of 1avx by two small insertions, one in each structure, but not in the bindingregion. The conformations of chains A and B of 1avx differ a bit from the free proteins due tothe formation of the complex. The exercises will compare docking the “native” ligand, chain Bof 1avx, to docking the 1ba7 ligand.

As the jobs take 4 hours to run, you should first create a project for the jobs, so that you canclose the BioLuminate interface and return to the project later to view and analyze the results,if you want.

1. Choose File → Save Project As.

2. Navigate to a location, and enter a name in the File name text box, such as1avx_docking_exercise.



60

10.1.1 Docking the Ligand Protein As Is

In this part of the exercise, you will dock the ligand protein to the receptor protein without anymodification other than what is done by the Protein-Protein Docking panel.

1. Choose Tasks → Protein-Protein Docking.

The Protein-Protein Docking panel opens.

2. In the Mode section, check that Standard is selected.

3. In the Protein structures section, click Receptor, then click From PDB ID.

4. Enter 1qqu in the Enter PDB ID dialog box, and click OK.

A question box opens, prompting you to add hydrogens to the structure.

5. Click Add Hydrogens.

The protein is displayed in the Workspace and the box next to the receptor is filled in withthe identity of the receptor. Chain A is selected, as this is the only chain available. TheView button next to this box is enabled: this button allows you to view the protein andzoom to it in the Workspace.

6. In the Protein structures section, click Ligand, then click From PDB ID.

7. Enter 1ba7 in the Enter PDB ID dialog box, and click OK.

An information box opens, indicating that problems were found when importing thestructure. These are structural defects that could affect modeling. In the case of protein-protein docking, missing side chains on the surface of the protein might be a cause forconcern, and then you should fix the structure before using it for docking. The protein-protein docking program can run with incomplete structures, but the results might not be

Figure 10.1. The Protein-Protein Docking panel with receptor and ligand defined.



what you want. This protein has a number of missing side chains, two of which are in thebinding region. It also has two short missing loops, which are not in the binding region.The missing side chains will be fixed in a later part of the exercise. In this part, the rawstructure is docked.

8. Click OK.



A dialog box opens, prompting you to select the chains to dock. In general, you canchoose as many chains as you like: there is no restriction on the number of chains.

10. Select chain B, and click OK.

The entire protein is displayed in the Workspace and the box next to the ligand is filled inwith the identity of the ligand. The View button next to this box is enabled.

The Constraints section is also enabled. In this exercise, no constraints are applied.


A Start dialog box opens, in which you can choose a host and name the job.

12. Enter 1qqu-1ba7-raw in the Name text box.

13. Ensure that Append new entries is chosen from the menu in the Output section.

14. (Optional) Choose a host.

15. Click Start.

The job starts. It takes about 4 hours to run, and runs on a single processor. If you have suffi-cient product licenses and processors available to run more than one job at a time, you canproceed immediately to the next part of the exercise, where the missing side chains will beadded to 1ba7 chain B and the resulting structure will be docked to 1qqu. Otherwise you canproceed when the job finishes.

10.1.2 Docking the Prepared Ligand Protein

In this part of the exercise, you will prepare the ligand protein, chain B of 1ba7, by adding themissing side chains. The protein also has two short loops that could be built, but this exercisedoes not include building them.

1. In the 1QQU row of the Toggle Table panel, choose A → Remove from Workspace.

1qqu is removed from the Workspace, leaving only 1ba7. The side chains will now beadded to this structure.



62



3. Select Remove original hydrogens.

4. Select Fill in missing side chains using Prime.


A dialog box opens, asking you to confirm the addition of missing side chains.

6. Click Continue.

The structure is processed, removing the hydrogens that were added on import and re-adding them later, and the side chains are added. The addition of the side chains takes afew minutes. When it is complete, a new entry is added and shown in the Workspace, andyou can now use this entry for docking.

7. Close the Protein Preparation Wizard panel.

Figure 10.2. Adding side chains.






8. In the Protein-Protein Docking panel, click Ligand, then click From the Workspace.

A dialog box opens, prompting you to select the chains to dock. In general, you canchoose as many chains as you like: there is no restriction on the number of chains.

9. Select chain B, and click OK.

The ligand has been replaced with the new structure that has side chains.



11. Enter 1qqu-1ba7-prep in the Name text box.



Since you already have a job running, you might want to choose a host. If you are runninglocally on a machine that has more than two cores, you need not select a different host, asthere will be room for both jobs. (The number of cores is reported in parentheses after thehost name.) Otherwise, choose a host for the job.

14. Click Start.

10.1.3 Docking the Native Ligand to the Native Receptor

To find out what sort of accuracy can be expected for protein-protein docking, it is a good ideato run a self-docking experiment, if possible. The reference protein 1avx is the complex of twochains that are very close in sequence to the receptor and ligand that we have been using, 1qquand 1ba7 chain B. The chains of 1avx adopt the optimal conformation in the complex, so if weuse chain B of 1avx as the ligand and chain A of 1avx as the receptor, we should get the bestpossible docking result. This result can be used as a standard for comparison of the results ofthe two docking runs performed so far.

To compare the “native” ligand (1avx chain B) to the docked poses, the receptor in 1avx (chainA) needs to be aligned to 1qqu, so that they are in the same frame of reference. This is notnecessary for the docking run, but only for comparison of the results. The alignment also tellsus how closely 1avx chain A matches the receptor protein 1qqu.

1. In the All row of the Toggle Table panel, click the A button in and choose Remove Every-

thing from Workspace.





64

3. Choose the 1QQU entry and click OK.

The receptor structure is now in the Workspace.


The Get PDB FIle panel opens.

5. Enter 1avx in the PDB ID text box, and click Download.

The reference structure is added to the Workspace, with the receptor in the background.


The Protein Structure Alignment panel opens below the Toggle Table. All atoms in bothstructures are marked in yellow.

7. Ensure that Use proteins from is set to Workspace (included entries).

8. Click the Clear (X) button in the Reference residues section.

The yellow markers are removed.

9. From the Pick menu, choose Chains.

10. Pick an atom in the receptor structure 1qqu (not the reference structure).

Yellow markers are added to the receptor structure and chain A of the reference structure.The text chain.name A is added to the ASL text box.

Figure 10.3. The Protein Structure Alignment panel with alignment results.



11. Ensure that Use same ASL as reference residues is selected under Residues to align.

12. Click Align.

After a short while, the proteins are aligned. The alignment score and the RMSD arereported in the Protein Structure Alignment Results dialog box. The score should be 0.010and the RMSD should be 0.486 (you might get slightly different results on differentsystems).

13. Close the Protein Structure Alignment panel.

The yellow markers are removed.

14. Click the Reset button in the Toggle Table to center the proteins.

15. In the 1QQU row of the Toggle Table panel, choose A → Remove from Workspace.

1qqu is removed from the Workspace, leaving only 1avx, which will now be used to setup the self-docking run.

16. In the Protein-Protein Docking panel, click Receptor, then click From the Workspace.



A dialog box opens, prompting you to select the chains to dock.

18. Select chain A, and click OK.

19. In the Protein-Protein Docking panel, click Ligand, then click From the Workspace.

Chain B is automatically selected because it is the only available chain.

20. Click OK.



22. Enter 1avx-self-dock in the Name text box.



If you are running the jobs simultaneously, and you already have two jobs running, youmight want to choose a host for the third job, depending on the resources you haveavailable.

25. Click Start.



66

The job starts. If you have worked through these exercises (and are not running other jobs), youshould see 3/3 on the Jobs button at the bottom of the Workspace, indicating that three jobs arerunning out of a total of three. The jobs take about 4 hours each. However, if your resources arelimited (either by licenses or by processors), you may see 1/3 on the Jobs button, as the jobsare being run sequentially. Then the total run time will be 12 hours.

10.1.4 Comparing the Top Poses

When the jobs finish, you should restart BioLuminate if you closed the interface, and reopenthe project if you closed it (File → Open Project or Open Recent Project). When the projectopens, you may be prompted to incorporate the results of the jobs, which you should do.

In the first part of the analysis, the top poses will be compared to the reference pose.

1. In the All row of the Toggle Table panel, choose A → Remove Everything from Workspace.

This ensures that you are starting with nothing displayed.


3. Select 1AVX and the three Ligand_1.000.00 lines, and click OK.

The ligand lines should be the first line under each receptor. Make sure that you selectthem in order—raw, then prepared, then self-dock—because they all have the same title,and you will not otherwise be able to tell which is which.

The reference structure and the best ligand pose for each docking run is included in theWorkspace. The Workspace looks a bit cluttered, but we will look at each pose in turn.

4. In the All row of the Toggle Table panel, choose S → As → Cartoon.

The structures are shown in cartoon representation. This makes visual comparison of theposes easier: showing the atoms gives too much detail.

5. In the 1AVX row of the Toggle Table panel, choose C → Color by Chain → by Chain.

The reference structure chains are colored, which distinguishes them from the ligands.

6. Click the rows in the Toggle Table panel for the first and second ligands.

These ligands are hidden, and you can now see the reference ligand and the pose from theself-docking run.

7. Rotate the structure to examine the match between the reference and the docked pose.

The docked pose is displaced and rotated somewhat from the reference. This is due to thedocking algorithm, because the pose should be exactly over the reference ligand if thedocking were perfect. This shows the level of accuracy you can expect from docking.



8. Click the rows in the Toggle Table panel for the second and third ligands.

The second ligand is shown and the third is hidden, and you can now see the referenceligand and the pose from the docking run with the prepared ligand.


The agreement is similar to that for the self-docked ligand.

10. Click the rows in the Toggle Table panel for the first and second ligands.

The first ligand is shown and the second ligand is hidden, and you can now see the refer-ence ligand and the pose from the docking run with the raw ligand.


The agreement is not as good as that for the prepared ligand. This is to be expected,because there are missing side chains in the binding region.

10.1.5 Calculating the RMSD Between the Top Poses

To quantify the difference, the RMSD between the docked ligands and the reference ligand canbe calculated. To avoid problems with missing side chains in the raw pose and sequence iden-tity, the RMSD will be calculated between the alpha carbons of the first 124 residues.

1. Choose Tools → Superposition.

The Superposition panel opens below the Toggle Table.

2. Ensure that Included entries is selected under Entries to superimpose.

Figure 10.4. Docked poses (gray) with the 1avx protein (receptor green, ligand blue). From the left: raw, prepared, self-docked.



68

3. Select Calculate 'in place' (no transformation).

This choice just calculates the RMSD without superimposing the structures.

4. Select Create RMSD property.

5. In the ASL tab, click Select.

The Atom Selection dialog box opens.

6. In the Chain tab, choose B from the Chain name list, and click Add.

7. In the Atom tab, choose PDB type from the list on the left.

The list of PDB atom types is displayed in the center, with the heading PDB type.

8. Choose CA from the PDB type list, and click Intersect.

9. In the Residue tab, choose Residue number from the list on the left.

10. In the Residue number text box, enter the text 1-124,501-624, and click Intersect.

The reason for the two residue ranges is that the reference residue chain B is numberedstarting at 501 rather than at 1.

11. Click OK.

Figure 10.5. The Atom Selection dialog box for superposition of poses.



The calculation of the RMSD is performed, and the results appear in the RMSD text box. Theyare also added to the project as a property.

The RMSD values are about 4.7 for the self-docked pose, 5.4 for the prepared ligand pose, and8.8 for the raw ligand pose. The difference between the self-docked pose and the preparedligand pose is relatively small, and could possibly be explained mainly by the fact that theligand from 1ba7 does not superimpose exactly on the reference ligand, even with the restric-tion to the common part of the chain. The RMSD for doing this superposition is about 0.5.(You can verify this by including only the reference ligand and the prepared ligand, and dese-lecting Calculate 'in place' in the Superposition panel, so that the ligands are actually moved.)

The difference between the raw ligand and the prepared ligand poses, on the other hand, issignificant, and demonstrates the importance of preparing the proteins. Many proteins havemissing side chains or loops on their surface, because these regions are more mobile and aremore likely to be poorly or incompletely modeled in the PDB structure. Using the Protein

Preparation Wizard and adding both side chains and filling short loops with Prime is recom-mended. If the X-ray data is available, you can consider using PrimeX to obtain a better struc-ture, guided by the density. Long loops can also be modeled by Prime, but in the absence ofstructural information it might be better to disregard poses that include long missing loops inyour docking runs.

12. Close the Superposition panel.

10.1.6 Examining the Poses and the Binding Sites

The final part of this exercise is to examine the poses that are generated, to see where on theprotein the ligand is likely to dock. For this purpose, it is better to use the Project Table, whichenables you to easily step through the display of a set of entries in the Workspace.

1. Choose File → Project Table, or press CTRL+T (T).

The Project Table panel opens.

2. Click the row that contains 1qqu-1ba7-raw in the title.

This row should have a number in square brackets in the Row column. It is the row for an“entry group”, which is a collection of project entries. This group contains the dockingresults for the “raw” run.

3. Choose Entry → Include Only.

After a short while, all the entries in the group are included in the Workspace.

4. In the All row of the Toggle Table, choose S → As → Cartoon.

The entries are all changed to cartoon representation.



70

5. In the Project Table panel, choose Entry → Exclude.

The entries are removed from the Workspace again, but they are now set up in cartoonrepresentation so that they can be compared.

6. Click the + sign in the In column for the group row.

The group is expanded so you can see all the entries. All of the rows are highlighted inyellow to show that they are selected.

7. Select the receptor row.

The receptor row is now the only row highlighted

8. Choose Entry → Fix.

The receptor entry is included in the Workspace, as indicated by the box in the In column.There is also a padlock icon, to indicate that this entry will remain in the Workspace untilit is explicitly removed.

9. Zoom out in the Workspace so that you can see the receptor and enough space around itto fit a ligand of the same size.

Click the group row again, to select all its entries.

This is the row that contains 1qqu-1ba7-raw in the title.

10. With the pointer in the Workspace, press the RIGHT ARROW key.

The first ligand is displayed in the Workspace.

11. Press the RIGHT ARROW key repeatedly, to display the ligands in turn.

To go back, you can press the LEFT ARROW key.

The majority of the ligands bind in the native binding site, but a few bind in other sites. It isclear that there are a few sites that favor binding, and other regions that do not favor binding.

This concludes the exercise, and you can now clean up.

12. In the All row of the Toggle Table panel, choose A → Remove Everything from Workspace.

13. Close the Protein-Protein Docking panel, if it is still open.

14. Choose File → Close Project.



10.2 Antibody-Antigen Docking

This exercise uses the antibody-antigen docking mode, in which the non-CDR regions of theantibody are automatically detected and masked in the docking. The reference for this exerciseis 1fsk: the bound antigen is chain A and the antibody consists of chains B (the light chain) andC (the heavy chain). This exercise uses chains B and C for the receptor. The free antigen is1bv1, which is used for the ligand, and has the same sequence as chain A of 1fsk. The RMSDfor superposition of the free antigen on the bound antigen is 0.65 angstroms.

First, save the project so that you can come back to it later. The docking run takes over 5 hourson a single processor.

1. Choose File → Save Project As.

2. Navigate to a location, and enter a name in the File name text box, such asantibody_docking_exercise.

10.2.1 Preparing the Antibody

The 1fsk structure is actually a tetramer, with 12 chains, so a single antibody-antigen unit willbe extracted for use in this exercise. A convenient way of doing this is to use the Protein Prep-aration Wizard panel, even though the antibody does not need processing for this exercise.


2. Enter 1fsk in the PDB text box, and click Import.

The 1fsk structure is imported into the project and displayed in the Workspace.

3. In the Review and Modify tab, click Analyze Workspace.

The chains are listed in the table.

4. Select chains A, B, and C.

5. Click Invert Selection, then click Delete.

Chains D and higher are deleted, leaving chains A, B, and C. This is the reference struc-ture that will be used for comparing the results, and also for the antibody receptor.

6. Close the Protein Preparation Wizard panel.



72

10.2.2 Setting Up and Running the Job

With the reference structure imported, you can now set up the docking run.

1. Choose Tasks → Protein-Protein Docking.

The Protein-Protein Docking panel opens.

2. In the Mode section, select Antibody.

The Mask non-CDR region option is now available, and is selected by default. This optionrestricts the docking to the CDR region. For this exercise, the option should remainselected.

3. In the Antibody structures section, click Receptor, then click From the Workspace.


Figure 10.6. Deleting chains.



2. Select chains.





The antibody is imported and analyzed to determine which are the light and heavy chains,and to locate the CDR regions. This operation takes a minute, and a progress bar is dis-played at the bottom of the panel. If you had imported 1fsk directly, the analysis wouldhave taken four times as long, so it is worth trimming down the structure beforehand.

When the analysis finishes, a dialog box opens, prompting you to select the chains to usefor the antibody receptor. You must select two chains: a light chain and a heavy chain.

5. Select chains B and C, and click OK.

The text box for the antibody is filled in with the information on the structure used.

6. Click Antigen, then click From PDB ID.

7. Enter 1bv1 in the Enter PDB ID dialog box, and click OK.

An alert box opens, warning you about the chains. You can dismiss this box.




10. Enter 1fsk-1bv1 in the Name text box.

11. Click Start.

The job starts. It takes about 5 hours to run.

10.2.3 Comparing the Best Pose with the Reference

After the results are imported, you can compare the best pose with the reference structure. Botha visual and a quantitative comparison are done in this section.

1. Remove everything but 1fsk from the Workspace.

You can do this with A → Remove from Workspace for each row.

2. In the 1FSK row, choose C → Color by Chain → by Chain.

The three chains are colored with different colors.


4. Select Antigen_1.000.00 and click OK.

The first (best ranked) antigen pose is included in the Workspace.



74

5. In the All row of the Toggle Table, choose S → As → Cartoon.

The proteins are shown in cartoon representation. You can see that the docked posematches the native pose very well. To quantify the agreement, the RMSD will be calcu-lated between the alpha carbons.

6. Choose Tools → Superposition.

The Superposition panel opens below the Toggle Table.

7. Ensure that Included entries is selected under Entries to superimpose.

8. Select Calculate 'in place' (no transformation).

This choice just calculates the RMSD without superimposing the structures.

9. In the ASL tab, click Select.

The Atom Selection dialog box opens.

10. In the Chain tab, choose A from the Chain name list, and click Add.

Figure 10.7. The 1fsk reference structure with the docked antigen pose (in gray).



11. In the Atom tab, choose PDB type from the list on the left.

The list of PDB atom types is displayed in the center, with the heading PDB type.

12. Choose CA from the PDB type list, and click Intersect.

13. Click OK.

The calculation of the RMSD is performed, and the results appear in the RMSD text box.The RMSD is about 2.9 angstroms, which indicates the quality of the result. This value issmaller than obtained for the general docking exercise. Here, the docking is constrainedto the CDR region, which might account in part for the better result.

You can examine the other poses if you wish, using a procedure similar to that inSection 10.1.6 on page 69. Because of the constraint, the poses are all in the same region.

This concludes the exercise, and you can now clean up.

14. Close the Protein-Protein Docking panel, if it is still open.

15. Choose File → Close Project.

The project is closed.

Figure 10.8. The Atom Selection dialog box for antigen superposition


76


Getting Help

Information about Schrödinger software is available in two main places:

• The docs folder (directory) of your software installation, which contains HTML andPDF documentation. Index pages are available in this folder.

• The Schrödinger web site, http://www.schrodinger.com/, particularly the Support Center,http://www.schrodinger.com/support center, and the Knowledge Base, http://www.schro-dinger.com/kb.

Finding Information in the BioLuminate interface

The BioLuminate interface provides access to nearly all the information available onSchrödinger software.

To get information:

• Pause the pointer over a GUI feature (button, menu item, menu, ...). In the main window,information is displayed in the Auto-Help text box, which is located at the foot of themain window, or in a tool tip. In other panels, information is displayed in a tool tip.

If the tool tip does not appear within a second, check that Show tooltips is selected underGeneral → Appearance in the Preferences panel, which you can open with CTRL+, (,).Not all features have tool tips.

• Click the Help button in a panel or press F1 for information about a panel or the tab that isdisplayed in a panel. The help topic is displayed in your browser.

• Choose Help → Online Help or press CTRL+H (H) to open the default help topic in yourbrowser.

• When help is displayed in your browser, use the navigation links or search the help in theside bar.

• Choose Help → Manuals Index, to open a PDF file that has links to all the PDF docu-ments. Click a link to open the document.

• Choose Help → Search Manuals to search the manuals. The search tab in Adobe Readeropens, and you can search across all the PDF documents. You must have Adobe Readerinstalled to use this feature.


http://www.schrodinger.com

http://www.schrodinger.com/supportcenter

http://www.schrodinger.com/kb

http://www.schrodinger.com/kb

Getting Help

78

For information on:

• Problems and solutions: choose Help → Knowledge Base or Help → Advanced Help

Options → Known Issues → product.

• Software updates: choose Help → Check for Updates.

• New software features: choose Help → Advanced Help Options → New Features.

• Scripts available for download: choose Tasks → Scripts → Update.

• Python scripting: choose Help → Advanced Help Options → Python Module Overview.

• Utility programs: choose Help → About → About Utilities.

• Keyboard shortcuts: choose Help → Keyboard Shortcuts.

• Installation and licensing: see the Installation Guide.

• Running and managing jobs: see the Job Control Guide.

• Using Maestro: see the Maestro User Manual.

• Maestro commands: see the Maestro Command Reference Manual.

Contacting Technical Support

If you have questions that are not answered from any of the above sources, contact Schrödingerusing the information below.

E-mail: [email protected]: Schrödinger, 101 SW Main Street, Suite 1300, Portland, OR 97204Phone: (503) 299-1150Fax: (503) 299-4532WWW: http://www.schrodinger.comFTP: ftp://ftp.schrodinger.com

Generally, e-mail correspondence is best because you can send machine output, if necessary.When sending e-mail messages, please include the following information:

• All relevant user input and machine output• BioLuminate purchaser (company, research institution, or individual)• Primary BioLuminate user• Installation, licensing, and machine information as described below.


mailto:[email protected]

http://www.schrodinger.com

Getting Help

Gathering Information for Technical Support

This section describes how to gather the required machine, licensing, and installation informa-tion, and any other job-related or failure-related information, to send to technical support.

For general enquiries or problems:

1. Open the Diagnostics panel.

• BioLuminate Interface: Help → Diagnostics • Command line: $SCHRODINGER/diagnostics

2. When the diagnostics have run, click Technical Support.

A dialog box opens, with instructions. You can highlight and copy the name of the file.

3. Attach the file specified in the dialog box to your e-mail message.

If your job failed:

1. Open the Monitor panel in the BioLuminate interface.

Use Tasks → Job Monitor.

2. Select the failed job in the table, and click Postmortem.

The Postmortem panel opens.

3. If your data is not sensitive and you can send it, select Include structures and deselectAutomatically obfuscate path names.

4. Click Create.

An archive file is created in your working directory, and an information dialog box withthe name of the file opens. You can highlight and copy the name of the file.

5. Attach the file specified in the dialog box to your e-mail message.

6. Copy and paste any log messages from the window used to start Maestro (or the job) intothe email message, or attach them as a file.

• Windows: Right-click in the window and choose Select All, then press ENTER tocopy the text.

• Mac: Start the Console application (Applications → Utilities), filter on the applica-tion that you used to start the job (Maestro, BioLuminate, Elements), copy the text.


80

120 West 45th Street 17th FloorNew York, NY 10036

101 SW Main StreetSuite 1300Portland, OR 97204

245 First StreetRiverview II, 18th FloorCambridge, MA 02142

8910 University Center LaneSuite 270San Diego, CA 92122

155 Gibbs StSuite 430Rockville, MD 20850-0353

Dynamostraße 13D-68165 MannheimGermany

Zeppelinstraße 73D-81669 MünchenGermany

Potsdamer Platz 11D-10785 BerlinGermany

Quatro HouseFrimley RoadCamberley GU16 7ERUnited Kingdom

8F Pacific Century Place1-11-1 MarunouchiChiyoda-ku, Tokyo 100-6208Japan

No. 102, 4th Block3rd Main Road, 3rd StageSharada ColonyBasaveshwaranagarBangalore 560079, India

SCHRÖDINGER ®

bioluminate quick start guide

Documents