Introduction to MD Trajectory Analysis with Desmond

Tutorial Created with Software Release: 2024-3
Topics: Small Molecule Drug Discovery, Structure Prediction & Target Enablement
Products Used: Desmond, Maestro

Tutorial files

0.8 GB
This tutorial is written for use with a 3-button mouse with a scroll wheel.
Words found in the Glossary of Terms are shown like this: Workspacethe 3D display area in the center of the main window, where molecular structures are displayed.

 

Tip: You can hover over a glossary term to display its definition. You can click on an image to expand it in the page.
Abstract:

 

In this tutorial, you will learn how to analyze an unrestrained all-atom molecular dynamics (MD) trajectory of a protein-ligand complex generated with Desmond. This entails the general and automated analysis of the protein-ligand interactions and the properties of protein and ligand, as well as more specific investigation of the binding site to enhance your structural understanding of the system. A detailed statistical analysis of quantitative ensemble properties is beyond the scope of this tutorial.

Tutorial Content

  1. Introduction to MD Trajectory Analysis

  1. Creating Projects and Importing Structures

  1. Simulation Interactions Diagram

  1. Plotting with the Trajectory Player

  1. Conclusion and References

  1. Glossary of Terms

1. Introduction to MD Trajectory Analysis

All-atom MD simulations describe the motion of each individual atom in a chemical system over time based on the interactions between them. The result of a simulation is a sequence of snapshots or frames, logging the atoms’ positions in space after very short time intervals. This is called a trajectory, and it can reveal the dynamic behavior of a chemical system and its associated structural, dynamical, and thermodynamic properties.

This tutorial will introduce you to the basics of MD trajectory analysis, with the goal to gain new structural and dynamical knowledge of your system. You will analyze the results of a Desmond generated unrestrained all-atom MD simulation, in particular you will study protein-ligand interactions and a specific part of the protein backbone in the binding pocket.

This tutorial does not cover structure preparation, how to set up a model system and how to run an MD simulation with Desmond. For further information on these topics see the Introduction to Structure Preparation and Visualization and Introduction to All-Atom Molecular Dynamics Simulations with Desmond tutorials. Additionally, a detailed statistical analysis of the trajectory to derive quantitative insights from the structural ensemble is beyond the scope of this tutorial.

The protein investigated in this tutorial is aldose reductase, an enzyme that catalyzes the reduction of aldehydes and carbonyls, primarily the reduction of glucose to sorbitol. It is associated with complications arising from diabetes, such as blindness, renal failure and heart disease. In the complex used here it is co-crystallized with the inhibitor zopolrestat. You will use the MD simulation to better understand the complementarity between ligand and receptor with the goal of deriving insights for further optimization of the ligand.

2. Creating Projects and Importing Structures

At the start of the session, change the file path to your chosen Working Directorythe location where files are saved in Maestro to make file navigation easier. Each session in Maestro begins with a default Scratch Projecta temporary project in which work is not saved, closing a scratch project removes all current work and begins a new scratch project, which is not saved. A Maestro project stores all your data and has a .prj extension. A project may contain numerous entries corresponding to imported structures, as well as the output of modeling-related tasks. Once a project is created, the project is automatically saved each time a change is made.

Structures can be imported from the PDB directly, or from your Working Directorythe location where files are saved using File > Import Structures, and are automatically added to the Entry Lista simplified view of the Project Table that allows you to perform basic operations such as selection and inclusion and Project Tabledisplays the contents of a project and is also an interface for performing operations on selected entries, viewing properties, and organizing structures and data. The Entry Lista simplified view of the Project Table that allows you to perform basic operations such as selection and inclusion is located to the left of the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed. The Project Tabledisplays the contents of a project and is also an interface for performing operations on selected entries, viewing properties, and organizing structures and data can be accessed by Ctrl+T (Cmd+T) or Window > Project Table if you would like to see an expanded view of your project data.

  1. Double-click the Maestro icon.

Figure 2-1. Changing the Working Directory via the option in the main window.

  1. Go to File > Change Working Directory.
  2. Find and choose the directory you want to use as your Working Directorythe location where files are saved.
  3. Pre-generated input and result files are included for running jobs or examining output. Download the .zip file here: https://www.schrodinger.com/sites/default/files/s3/release/current/Tutorials/zip/desmond_trj_analysis.zip.
  4. After downloading the zip file, unzip the contents into your Working Directorythe location where files are saved for ease of access throughout the tutorial.

Figure 2-2. Saving the project via the Save Project As dialog box.

  1. Go to File > Save Project As.
  2. Change the File name to Aldose_Reductase_Analysis,
    click Save.
    • The project is now named Aldose_Reductase_Analysis.prj and is saved in your Working Directorythe location where files are saved.

3. Simulation Interactions Diagram

Non-covalent protein-ligand interactions are of paramount importance in the drug design process as they, along with steric and thermodynamic contributions, determine the interaction strength and binding affinity of a ligand for a receptor.

In this section, you will analyze protein-ligand interactions as well as protein and ligand properties in graphical form. You will do this general analysis of a protein-ligand complex in an automated way with the Simulation Interactions Diagram (SID) panel.

You can generate reports from the panel to export your data as a PDF summary of the analysis, the plots as single images and the raw data as plain text to perform further and customized analysis.

Please note that calculations may have been performed with an earlier version of the software and the results may not be exactly the same as those you produced in this tutorial. Especially for methods with a statistical component, such as MD simulations, an exact reproduction of a trajectory and its properties is only possible if the same random seed, software and hardware configuration are used. With sufficient sampling, all properties of the ensemble should be reproducible, while the properties of individual frames and the exact course of a trajectory may differ.

3.1 Generating an event analysis file or loading an existing file

To obtain the necessary analysis results for visualization in the SID panel, you can either activate the option Run interactions analysis when the simulation completes in the Molecular Dynamics panel in the analysis section as shown in the Introduction to All-Atom Molecular Dynamics Simulation with Desmond tutorial. Alternatively, you can import any Desmond trajectory (via the -out.cms file) and perform the analysis directly from the SID panel. The results will be stored in a generated event analysis file (.eaf) in both cases.

Figure 3-1. Loading a file into the Simulation Interactions Diagram panel.

If you are continuing from the Introduction to All-Atom MD Simulation with Desmond tutorial and have run the MD yourself, you can use the file obtained there or use the file provided in the .zip file coming with this tutorial (desmond_md_2DUX.eaf).

  1. Go to Tasks > Browse > Classical Simulation > Simulation Interactions Diagram.
    • The Simulation Interactions Diagram panel opens.
  1. Click Load….
    • The Select Event Analysis or Trajectory File dialog box opens.

You will now be prompted to provide both the event analysis file and the trajectory file, one after the other.

  1. Find and choose the desmond_md_2DUX.eaf file from the zip archive.
  2. In the second dialog box, find and choose the desmond_md_2DUX-out.cms file from the zip archive.

You can now continue with section 3.2 to explore the results of the analysis. The following steps are optional instructions for generating the .eaf file for an existing trajectory.

Figure 3-2. Setting up the SID analysis the SID Analysis Setup dialog box.

  1. Click Load….
  2. Find and choose the desmond_md_2DUX-out.cms file.
    • The SID Analysis Setup dialog box opens.
  3. Click Select All, to choose all available analysis types.

 

Note that frame zero is the default reference structure. In principle you could select any frame or import a structure from an external file.

 

  1. Click Run.
    • A window opens to select the ligand molecule.

Figure 3-3. Choosing the zopolrestat ligand.

  1. Click on the zopolrestat (ZST) ligand on the right to select it.
    • The chosen ligand is highlighted in green.
  2. Confirm by clicking OK.
    • The SID - Job Settings window opens.

Figure 3-4. Naming and running the SID analysis job.

  1. For Name choose desmond_md_2DUX_SID.
  2. Click Start to run the job.
    • Depending on your hardware this job will take ~1 hour.
    • After completion, the results of the analysis are directly loaded into the panel.

3.2 Analyzing the SID results

The results of the SID analysis are displayed in eight different tabs, showcasing different protein and ligand properties, as well as their interactions: Protein and ligand root-mean-square displacement (RMSD), protein secondary structure elements (SSE), protein root-mean-square fluctuation (RMSF), ligand RMSF, protein-ligand contacts, ligand-protein contacts, ligand torsions and ligand properties.

Figure 3-5. The PL-RMSD tab of the SID panel with all supported protein RMSDs displayed over the course of the trajectory.

  1. In the PL-RMSD tab, select the options for C-alphas, Side chains, Backbone and Heavy atoms one by one to show the RMSD of the selected atoms of the protein over the simulation time.

 

All frames are first aligned on the backbone of the reference structure, then the RMSD is calculated for the selected group of atoms.

 

  1. Hover over the plot to show the frame number/simulation time and corresponding RMSD values for a particular frame.

 

Figure 3-6. The PL-RMSD tab of the SID panel with all supported ligand RMSDs displayed over the course of the trajectory.

  1. Unselect the options for the protein RMSDs.
  2. For the ligand, choose the options Aligned on Ligand and Aligned on Protein to display the RMSDs of the ligand over the simulation time.

 

Aligned on Ligand: The ligand is aligned on its reference frame and the RMSD is calculated for heavy atoms.

 

Aligned on Protein: The protein-ligand complex is aligned on the protein backbone and the RMSD of the ligand’s heavy atoms is calculated.

The RMSD describes the change in displacement of a selection of atoms for a given frame with respect to a reference frame, which is typically the first frame. In other words, it is a measure of structural similarity for all frames compared to the starting structure of the simulation.

A stabilized RMSD value that only fluctuates around a mean value confirms the equilibration of the system and the convergence of the simulation. An initial increase in RMSD is expected, as the conformation adjusts to the solvated environment. In particular, flexible loops can be constrained in unfavorable conformations in the crystal environment and need some time to fully relax. If the RMSD continues to increase throughout the simulation, there may be other issues with the structure, in which case you should visually inspect the simulation, dive deeper into the results, and have another look at the choices made during protein preparation.

 

Additionally, the RMSD can give insights into the structural conformations during the simulation. Fluctuations in the order of 1 to 3 Å are expected for small globular proteins, while larger fluctuations may indicate conformational changes.

Here, both the protein and ligand are sufficiently equilibrated and no conformational changes occur. The protein RMSD still increases during the first 50 ns, but then stabilizes on average. Naturally, the protein side chains show higher RMSD values, because they are more mobile compared to the protein backbone. The ligand RMSD aligned to the ligand itself shows the internal fluctuations of the ligand structure, which do not indicate any conformational changes here. The ligand RMSD aligned to the protein shows the movements of the ligand relative to the protein. The ligand does not move significantly relative to the protein. Only at about 85 to 110 ns there is a slight increase in the RMSD that warrants a closer look as it could indicate a small change in binding mode.

Figure 3-7. The P-SSE tab of the SID panel with the time per residue spent in a SSE as well as the total and per residue SSE along the trajectory.

  1. Switch to the P-SSE tab to display the protein secondary structure elements (beta-strand, alpha-helix, loop).

 

The top plot shows the distribution of SSE as a percentage of time every residue spends in a structural element.

 

The middle plot shows the total (helix plus strand) percentage of elements for the whole protein along the simulation time.

 

The bottom plot shows the elements as a function of residue index along the trajectory.

The tertiary structure of aldose reductase consists of several alpha-helices and beta-strands in between loop regions. All SSE appear to be mostly stable throughout the simulation, as is the total occurrence of helices and strands. An interesting exception are the residues around index 300, which spend only a small percentage of time in a helical conformation, distributed over the course of the trajectory. This observation is worth keeping in mind for further analysis.

Figure 3-8. The P-RMSF tab of the SID panel with all supported protein RMSFs as well as B factors, ligand interactions and secondary structure highlighted for all residues.

  1. Switch to the P-RMSF tab and select the options for C-alphas, Side chains, Backbone and Heavy atoms.
  2. Choose the options B factor (PDB), Ligand contacts and Secondary structure, set the Smooth value to 5.

 

Protein-ligand interactions are highlighted with green vertical lines. Secondary structure regions, which persist for more than 70% of simulation time, are marked in light red (alpha-helices) or light blue (beta-strands) vertical bars.

 

  1. Hover over the plot to show the chain, residue index, residue name, RMSF and B factor values for particular residues.

While the RMSD is a measure of structural similarity as a function of simulation time, the RMSF measures structural fluctuations as a function of residue or atom. In other words, it describes the local flexibility broken down to individual residues or atoms. Alpha-helices and beta-strands are more rigid and therefore fluctuate less than loop regions and especially the termini. The crystallographic B factors, which are a measure of positional uncertainty in the crystal, are expected to correlate well with the RMSF by definition. A poor correlation could indicate structural issues, such as a bad crystallographic fit.

For aldose reductase, the termini and the large unstructured region around residues 210 to 230 show large RMSF values. The B factor correlates well with the RMSFs. Most interactions with the ligand occur in the region around residue 300, which also shows some flexibility. This suggests that this region actually is part of the binding site of the ligand and is another reason to look at this region more closely with an emphasis on the interactions with the ligand.

Figure 3-9. The L-RMSF tab of the SID panel with all supported ligand RMSFs displayed per ligand heavy atom.

  1. Switch to the L-RMSF tab and choose the options Aligned on protein and Aligned on ligand.

 

Aligned on Ligand: The ligand is aligned on its reference frame and the RMSF is calculated for heavy atoms.

 

Aligned on Protein: The protein-ligand complex is aligned on the protein backbone and the RMSF of the ligands heavy atoms is calculated.

 

  1. Hover over the plot to show the atom number, name and type as well as the RMSF values for particular atoms. The atom is additionally highlighted in the 2D representation of the molecule in the top right corner.

The ligand RMSF gives you insights into the local flexibility of individual atoms. Here, atoms with a large RMSF are the fluorine atoms at the tail of the ligand, which are able to rotate freely around the C-C bond. All other atoms have small RMSF values. Comparing the ligand RMSF in water and in the binding pocket can indicate the entropic role in the binding event.

Figure 3-10. The PL-Contacts tab of the SID panel with the fraction of interaction time as well as the total and per residue interaction counts along the simulation time.

  1. Switch to the PL-Contacts tab and select the options for all available interaction types (H-bonds, Hydrophobic, Ionic, Water bridge and Halogen bonds).

 

The top plot shows the average fraction of simulation time per residue in which they are interacting with the ligand. The value can be higher than one, as a residue can have multiple interactions with the ligand. Only choosing a single interaction type option at the top shows a breakdown of the interaction type into more specific subcategories.

 

  1. Hover over a particular bar to display the residue name, contact fraction and frame count.

 

The middle plot shows the total number of interactions for the selected types as a function of simulation time. The bottom plot displays the same but broken down to specific residues.

 

  1. Hover over the contact options at the top to reveal a tooltip with the definitions for the interaction types.

The protein-ligand interactions are categorized as hydrogen bonds, hydrophobic interactions, ionic interactions, water bridges and halogen bonds. For the aldose reductase-zopolrestat complex, the overall number of interactions stays fairly constant. This indicates sufficient equilibration of the ligand in the binding pocket and no change in binding mode.

The interactions mostly consist of hydrogen bonds and hydrophobic interactions. There are strong and stable hydrogen bonds to residues TYR 48 and TRP 111, which are stable throughout most of the simulation time. TRP 111 also shows a strong and stable hydrophobic interaction, as well as a very short water bridge at about 85 ns, the time at which the increase in ligand RMSD occurs. You will have a closer look at this in the following section.

LEU 300, which alternates between a loop and alpha-helical conformation, switches between a hydrogen bond and a hydrophobic interaction. In the following section, you will investigate a potential correlation.

Figure 3-11. The LP-Contacts tab of the SID panel shows a 2D schematic of the ligand and the types and points of interaction with the protein.

  1. Switch to the LP-Contacts tab.
    • A schematic 2D diagram of the ligand is displayed, highlighting the ligand interactions with specific residues.
  2. Change the Minimum contact strength to 20%.

 

Note that interactions that occur for less than the specified percentage of the Minimum contact strength will not be displayed.

Here you can see the points and types of interaction of the ligand and the protein binding pocket in a 2D schematic diagram. In addition, the solvent exposed regions of the ligand are highlighted.

The TYR 48 residue has a stable hydrogen bond to the carboxylate group of the ligand. TRP 111 shows a hydrophobic π-π stacking interaction with either of the two rings of the benzothiazole group and an additional hydrogen bond to the carbonyl oxygen of the carboxylate group. For LEU 300 there is a less stable hydrogen bond to the nitrogen atom of the benzothiazole group. You will investigate this further in the 3D structure in the next section.

Figure 3-12. The L-Torsions tab of the SID panel displays the five torsion distributions of the ligand.

  1. Switch to the L-Torsions tab.

 

The plots display the torsions for all rotatable bonds of the ligand as polar plots and bar charts.

 

  1. Click on Legend to open a panel explaining the plots.

Investigating the ligand torsions can give you insights into the conformational stability and strain in the protein bound conformation. The green, red and yellow torsions show a very narrow and defined region of density only in regions of low torsion potential. In contrast, the lilac torsion shows a generally lower potential curve and the density is distributed over all possible torsional angles. The trifluoride group essentially freely rotates in the simulation.

The blue torsion of the carboxylate group shows density in one single region of high torsional potential, which indicates two things in the protein bound conformation: No free rotation of the group, which indicates an unfavorable entropic contribution, as well as conformational strain, indicating an unfavorable enthalpic contribution.

So why does the ligand nevertheless adopt this conformation? Looking back at figure 3-11, you see that the carboxylate group shows hydrogen bonds to three different residues (TYR 48, HIS 110 and TRP 111). These favorable enthalpic contributions must outweigh all other contributions. To further improve the ligand, you could try to minimize these unfavorable contributions while maintaining the hydrogen bonds, e. g. by cyclization of the ligand.

Figure 3-13. The L-Properties tab of the SID panel.

  1. Switch to the L-Properties tab to show some properties of the ligand.
  2. Hover over one of the plots to display a vertical line in all plots and the respective values of the properties.

Six ligand properties are shown as a function of simulation time and bar charts, respectively: RMSD, radius of gyration, intramolecular hydrogen bonds, molecular surface area (MSA), solvent accessible surface area (SASA) and polar surface area (PSA).

All properties are relatively constant throughout the simulation, fluctuating around a mean value. No intramolecular hydrogen bonds are detected. In principle, these properties could indicate problems in the simulation, such as ligand unbinding.

Figure 3-14. Exporting the analysis data by generating a report.

You can export a PDF report of the entire analysis as well as all the plots and data for further analysis and use in manuscripts or presentations.

 

  1. Click Generate Report…
    • The Export Options menu opens
  2. Select all three options for PDF Report, Plots and Data, then confirm with OK.
    • The files are written to your Working Directorythe location where files are saved.
  3. Close the SID panel.

4. Plotting with the Trajectory Player

In addition to the general and automated analysis with the SID panel, you can perform more specific and customized analysis in Maestro with the Trajectory Player. Geometric measurements, such as atomic distances, angles and dihedral angles, ring distances and angles can be calculated and plotted over the course of the trajectory. All interaction counts available via the Interactions Toolbox and a variety of descriptors (RMSD, RMSF, PSA, SASA, MSA, etc) and energetic properties can be plotted as well. Note that the calculation of energetic properties requires access to a Linux host with a GPU. The measurements are available for predefined and custom groups of atoms and account for the periodic boundary conditions for measurements that span the boundary of the simulation box.

Depending on the system under investigation, different properties for customly selected atom groups can yield insightful information to address your research question. Here, you will analyze the trajectory to drill down into some of the observations and questions raised when you analyzed the SID. Specifically, you will examine the interactions between the residue TRP 111 and the ligand in the context of an observed increase in ligand RMSD and short lived water bridges. Additionally, you will study the protein backbone of residues ALA 299 and LEU 300 in the binding site, where you observed a helix motif which was only transiently present during the course of the trajectory.

Please note that If you have run the MD simulation yourself, your results may not be exactly the same as those reported in this tutorial. Especially for methods with a statistical component, such as MD simulations, an exact reproduction of a trajectory and its properties is only possible if the same random seed, software and hardware configuration are used. With sufficient sampling, all properties of the ensemble should be reproducible, while the properties of individual frames and the exact course of a trajectory may differ.

Figure 4-1. Importing the structure.

  1. Go to File > Import Structures….
    • The Import dialog box opens.
  2. Navigate to your Working Directorythe location where files are saved, select the file desmond_md_2DUX-out.cms and click Open.
    • A banner appears and a group is added to your Entry Lista simplified view of the Project Table that allows you to perform basic operations such as selection and inclusion.

Figure 4-2. Loading the trajectory via the workflow action menu.

  1. Load the trajectory by either double-left-clicking on the small T button in the Title column of the Entry Lista simplified view of the Project Table that allows you to perform basic operations such as selection and inclusion, or by right-clicking on the T button to open the workflow action menu and selecting Load Trajectory.
    • The Trajectory Player opens with the loaded trajectory in the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed and the toolbar at the bottom.

Optional: Choose a visualization style that's most helpful to you. Since you are interested in visualizing the ligand in the binding site, it may be useful to hide all atoms beyond a certain distance from the binding site. You can easily do this with the option Hide atoms: Beyond binding site in the Playback Settings of the Trajectory Player in the bottom toolbar.

Figure 4-3. There is a hydrogen bond between the TRP 111 residue and the ligand in the first frame.

  1. Have a look at the binding pocket.
  2. Locate the residue TRP 111 in the binding pocket (e.g. the Search Bar in the Structure Hierarchy), zoom to this residue plus the ligand and observe the interactions between them.

Figure 4-4. There is a water bridge between the TRP 111 residue and the ligand in the frame 871.

 

  1. Play through the frames of your trajectory and keep an eye on changes of the interactions between residue TRP 111 and the ligand as well as the surroundings, especially in the time period from 85 to 110 ns.

In the first part of the trajectory, there is a stable hydrogen bond between the residue TRP 111 and the carbonyl atom of the carboxylate group of the ligand. Both are not part of the solvent exposed region, but are buried relatively deep inside of the binding pocket. At around 85 ns, a water molecule diffuses into the binding pocket and replaces the hydrogen bond between TRP 111 and the ligand with a water bridge. After the water molecule diffuses out of the binding pocket again at around 110 ns of simulation time, the initial hydrogen bond is reestablished.

This observation correlates well with the increase in ligand RMSD relative to the protein between 85 and 110 ns.  The RMSD shift is likely caused by small rearrangements within the binding pocket to accommodate for this water molecule.

WaterMap is a versatile tool for studying the energetics and thermodynamics of water molecules in the binding pocket and can indicate whether displacement of this water molecule would be worthwhile in terms of drug potency of the ligand. Learn more in the Target Analysis with SiteMap and WaterMap tutorial.

Figure 4-5. Measuring a dihedral angle from the favorites toolbar.

  1. Locate the residues ALA 299 and LEU 300 in the binding pocket (e.g. via the Search Bar in the Structure Hierarchy) and zoom to both residues plus the ligand.
  2. Click Measure in the favorites toolbar.
    • A banner at the top of the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed appears.
  3. For Measure, select Dihedrals.

Figure 4-6. The measured dihedral angle along the backbone of the residues ALA 299 and LEU 300 in the first frame.

  1. Click on the following atoms in the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed: ALA299:N - ALA299:Cɑ - ALA299:C - LEU300:N
    • The selected dihedral angle is shown in the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed.

Figure 4-7. The measured dihedral angle for the backbone of the residues ALA 299 and LEU 300 in frame 1613.

  1. Play through some frames of your trajectory and keep an eye on the dihedral angle and the conformation of the residues ALA 299 and LEU 300.

Figure 4-8. Plotting the measurement, which is currently in the workspace, along the simulation time.

  1. In the Trajectory Player toolbar, click Plot > Measurements > Currently in Workspace.

Figure 4-9. The plotted dihedral angle for the backbone of the residues ALA 299 and LEU 300 over the simulation time.

  1. Click different points on the plot.
    • On click, the corresponding trajectory frame is displayed in the Workspacethe 3D display area in the center of the main window, where molecular structures are displayed and the view zooms to the atoms involved in the property.

The plot shows two distinct conformations or states along the trajectory for the backbone of the residues ALA 299 and LEU 300, with several transitions between them. The backbone conformation where the dihedral angle is roughly 70 degrees can lead to an alpha helical secondary structure across this and the neighboring residues. The time spent in the helix-inducing conformation can be estimated to roughly 3 ns, which corresponds to the percentage shown in the P-SSE tab of the interaction analysis. The non-helical conformation seems to be more probable, but to analyze this quantitatively, more sampling would be necessary.

The effect of the fluctuation of this dihedral angle goes beyond inducing a change in secondary structure. Changing the dihedral angle means that the oxygen atom of the carbonyl group of ALA 299 and the nitrogen atom of LEU 300 flip by 90 degrees relative to the ligand. This essentially leads to the nitrogen of LEU 300 to turn towards the ligand, forming a hydrogen bond to the nitrogen atom in the benzothiazole group, or to turn away by 90 degrees, thereby not allowing a hydrogen bond to be formed. This affects the hydrogen bond network in the pocket, which is relevant when targeting the binding pocket with potential drugs.

Experimental evidence validates this finding. Steuber et al. (J. Mol. Biol. 2006, 363, 174–187) found both conformations of this part of the backbone under different crystallization conditions (PDB codes 2DUX and 2DV0). They measured the dihedrals to be approximately 170 and 70 degrees, which agrees well with the observed values in your MD simulation.

 

Both conformations should be considered when designing new drugs for this specific target, e.g. by using both structures for docking in virtual screens.

5. Conclusion and References

In this tutorial, you learned how to analyze MD trajectories with Desmond. You started with the systematic and automated analysis of protein-ligand interactions as well as protein and ligand properties over the simulation time with the SID panel. Then, you investigated two key regions of the binding pocket with the Trajectory Player to follow up on observations from the automated analysis. You were able to enhance your structural knowledge of the system and found potential next steps to improve the binding of the ligand.

Finally, you switched from GUI-based tools to a command line approach for manipulating your trajectory with Desmond utilities.

In addition to running and analyzing Desmond MD simulations from the Maestro GUI, you can use the command line to perform more advanced and flexible types of simulations and trajectory manipulations and analyses via the multisim workflow and Desmond utilities. This includes enhanced sampling methods such as Metadynamics, Replica Exchange and Simulated Annealing, which can be run in a basic variant from the GUI as well. The Python API allows for more customized analysis and can be tailored to specific and complex research questions.

6. Glossary of Terms

Entry List - a simplified view of the Project Table that allows you to perform basic operations such as selection and inclusion

Included - the entry is represented in the Workspace, the circle in the In column is blue

Project Table - displays the contents of a project and is also an interface for performing operations on selected entries, viewing properties, and organizing structures and data

Scratch Project - a temporary project in which work is not saved, closing a scratch project removes all current work and begins a new scratch project

Selected - (1) the atoms are chosen in the Workspace. These atoms are referred to as "the selection" or "the atom selection". Workspace operations are performed on the selected atoms. (2) The entry is chosen in the Entry List (and Project Table) and the row for the entry is highlighted. Project operations are performed on all selected entries

Working Directory - the location where files are saved

Workspace - the 3D display area in the center of the main window, where molecular structures are displayed