nic

1.Which tool is used for analysis of the results?

2.Which trajectory formats are supported by AQUA-DUCT?

3.Can other programs than PyMOL be used for visualisation of AQUA-DUCT results?

4.How are resids indexed in AQUA-DUCT calculations?

5.What is the difference between easy, normal and expert modes in AQUA-DUCT GUI? What do the respective modes offer and how can the chosen modes affect the results?

6.Can AQUA-DUCT be used to trace small drug-like molecules either directly or extrapolated from solvent results?

7.How is AQUA-DUCT different from other tools, such as GIST?

8.Does AQUA-DUCT restrain the macromolecule?

9.How to define Scope and Object in a system of interest?

10.My system of interest is actually a transmembrane pore made of separate peptides. How do I define Scope here? the oligomer only or the whole membrane in which it is embedded?

11.Have you been able to validate AQUA-DUCT results experimentally?

12.Why does AQUA-DUCT functionality surpass that of other software?

13.Why and when should you try AQUA-DUCT?

14.How can I validate a dynamic when there is no native bound, that is, co-crystallized with the protein?

nic

nic

nic

1.Which tool is used for analysis of the results?

The analysis itself is all done in AQUA-DUCT. After the calculations you get the result files: 5_analysis_results.txt, 5_analysis_results.txt.csv, 6_visualize_results.py and 6_visualize_results.tar.gz. Two of the first files can be used for quantitative analysis and visualisation of the results as plots in Kraken, which is a small and smart GUI application that helps to gather all quantitative results from AQUA-DUCT. The third file is a Python script which when run, launches the 3D visualisation in PyMOL which is saved in 6_visualize_results.tar.gz archive.
For advanced analysis, pond module in AQUA-DUCT can be used. The results from pond are also visualised in PyMOL.

2. Which trajectory formats are supported by AQUA-DUCT?

AQUA-DUCT supports any trajectory format that is supported by MDAnalysis library. This includes (but does not limit to) coordinate and trajectory files from AMBER, GROMACS and NAMD. For more information visit MDAnalysis documentation: https://www.mdanalysis.org/docs/documentation_pages/coordinates/init.html.

If you work on an MDAnalysis incompatible format, you can easily convert them to a compatible format using one of a few available tools (for example CPPTRAJ, catdcd).

3.Can other programs than PyMOL be used for visualisation of AQUA-DUCT results?

Up until today, only PyMOL can be used to visualise the AQUA-DUCT results. As a result of calculations, a Python script is generated, which when run, launches AQUA-DUCT analysis in PyMOL. 

Other programs, such as VMD, can be used only to visualise MD simulations results prior to AQUA-DUCT calculations.

4.How are resids indexed in AQUA-DUCT calculations?

Resids are indexed exactly as they would be in your topology file, so it depends on how they would be marked in your topology file.

5.What is the difference between easy, normal and expert modes in AQUA-DUCT GUI? What do the respective modes offer and how can the chosen modes affect the results?

Modes don’t directly change your results but offer you incrementally more options to choose from. The Normal mode of editing provides you all the data to perform the basic solvent transportation analysis. The Expert mode extends the available options to calculations of global hotspots and more advanced paths smoothing. 

Modes have been made to facilitate gradual expansion of your options, so as to not overwhelm a new user. Should you be confident in what you are doing, you can use expert mode freely, as most options are not mandatory either way. Information on what options are and how they can be used is available within documentation on our website. Also, you can use valve_run –dump-template-config to output a template config featuring all available options.

6.Can AQUA-DUCT be used to trace small drug-like molecules either directly or extrapolated from solvent results?

You can track basically any molecules you wish, but the power of AQUA-DUCT comes from the concept of molecular probes and high sampling [see: https://academic.oup.com/bioinformatics/article/33/13/2045/3059930 and https://academic.oup.com/bioinformatics/article/36/8/2599/5682414]. So if you wish to track the drug-like compound you would probably need a lot of simulations to have good sampling. That is not a very efficient way. Of course you can, and since the events of such molecules entry are shorter, then you can use consolidator mode to merge parts of several simulations together.

7.How is AQUA-DUCT different from other tools, such as GIST?

GIST is similar to pond module, while it lacks the ability of valve module to trace small molecules within a selected space, and cannot say anything about for example the direction of solvent flow. GIST uses the Inhomogeneous Fluid Solvation Theory using water molecules to detect the water hydration sites. AQUA-DUCT’s pond uses the local solvent distribution approach, which generally yields best results for water molecules as well as some cosolvents molecules such as acetonitrile, phenol or any other small molecules. Although AQUA-DUCT will provide you much more information, for example, the energy profile along selected path, pockets – so the maximal available volume of the protein which can be penetrated by small molecules. AQUA-DUCT gives you also the precise information on the local density of particular sites of the macromolecule. As such we can say that AQUA-DUCT has a wider set of use cases and provides more types of information.

8.Does AQUA-DUCT restrain the macromolecule?

No, AQUA-DUCT does not restrain the macromolecule itself, but you need to remember to prepare  the trajectories and topologies in such a way that the protein needs to be centered inside the solvent box, you need to reduce the rotational movements and align the protein structure to some reference, e.g., the starting point of your simulation. This is ultimately because AQUA-DUCT takes the path as-is, and does not adjust coordinate trajectories to rotation.

9.How to define Scope and Object in a system of interest?

Running AQUA-DUCT calculations requires knowledge of your system of interest. Scope should be defined as the area in which transport of molecules you want to analyse, e.g. protein. Object should be defined as an area inside Scope of particular interest, e.g. protein’s active site. Only those molecules that have passed through Object will be traced in AQUA-DUCT, unless you add the option add_passing (then all the molecules found in Scope will be traced). Object and Scope should be defined in accordance with selection commands in MDAnalysis. The details on how to define Object and Scope can be found in AQUA-DUCT documentation [https://www.aquaduct.pl/documentation/ or https://tunneling-group.github.io/aqua-duct/valve/valve_manual.html#object-definition] or selection section in MDAnalysis documentation [https://docs.mdanalysis.org/1.0.0/documentation_pages/selections.html].

Please keep in mind, that you can use different combination of Object and Scope to address different questions, e.g., when your system has an active site and cofactor binding cavity you can set once Object within the active site and second time within the cofactor cavity. Also the main purpose of Scope is to speed-up calculations by limiting analysis to the relevant space, thus in some circumstances you can limit Scope for example to one of the subunits of the protein complex etc. 

10.My system of interest is actually a transmembrane pore made of separate peptides. How do I define Scope here? the oligomer only or the whole membrane in which it is embedded?

If you analyze a more complex system, such as a transmembrane system the flexibility of the Object and Scope definition gives you a tool to find answers for any questions focused on transportation. For instance, when you are interested in transportation only through the channel of transmembrane protein, you can limit Scope to protein and set Object in the middle of the channel, when you are interested in the flow through the membrane itself, you can use different definitions of Scope and Object. You can also use option scope_convexhull_inflate which allows you to increase or decrease the Scope area around the already defined definition of Scope

11.Have you been able to validate AQUA-DUCT results experimentally?

Up until today, we have not been able to directly verify AQUA-DUCT results experimentally, since we are not able to experimentally track solvent molecules. Definitely such a validation would be extremely useful and would provide an answer for the best settings for MD simulations (such as force field, and water model) that should be used to analyse protein interior. 

12.Why does AQUA-DUCT functionality surpass that of other software?

There are a lot of other tools that incorporate solvent molecules as molecular probes but AQUA-DUCT is the (only) one that allows to study properties of macromolecules from the intramolecular voids perspective. AQUA-DUCT’s functionality surpasses the one of other software since it provides a lot of types of information, such as analysis of transport of molecules along with detection of tunnels, energy profiles or sites of highest local density (hot spots). AQUA-DUCT calculations give a thorough image of the system. The main advantages of the AQUA-DUCT are:

  • system independent – you can analyse MD simulations from any software you are using (Amber, GROMACS, NAMD, and others)
  • ligands independent – you can analyse flow of water, co-solvents, small ligands, or ions, also within one calculation
  • high flexibility – you can adjust the area of interests, determine which tunnels entrances to inspect, analyse single or multiple simulations, analyse whole simulations or time evolution of the system
  • time dependent/independent analysis – you can use one software to get both the global statistic of your system as well as time dependent analysis (e.g., the information about your system in different conformational states) together with the information about system evolution
  • as for our knowledge, the only software that during single analysis can provide an access to information vital for protein engineering and drug design

13.Why and when should you try AQUA-DUCT?

Basic analysis: tracking small molecules flow through the active site of an enzyme, or a cavity of a macromolecule of interest; comparison of the flow of traced molecules through the active site of multiple variants of an enzyme of interest; analysis of the small molecules flow through different cavities of a macromolecule of interest. AQUA-DUCT provides the user with multiple statistical information on the small molecules (such as water molecules, ligands, ions, small-molecules or co-solvents) flow through a user-defined cavities. The user is then able to compare the flow through different entrances, analyse the direction of the traced molecules flow, etc. The basic analysis provides the user also very important information on the interior of the macromolecule of interest, such as which cavities are penetrable for a particular types of molecules, which are inaccessible and which are accessible but rarely.

Advanced analysis: identification of the most “attractive” regions within the macromolecule of interest as well as those which are able to trap a particular types of molecules (i.e., hot-spots), possible for multiple types of molecules which were present during the MD simulation (similar as a MixMD approach); identification of energetical barriers along a selected tunnel (i.e., energy profile); time-window analysis which enables the analysis of time evolution of the system. AQUA-DUCT provides the user with tools which are really helpful during protein engineering or drug design processes. The user is able to identify hot-spots for small molecules of different physico-chemical properties, which can then be utilized as starting points for drug design. Moreover, they gain access to the information on the energetic barriers along the selected tunnel, which is definitely helpful in protein re-design to modify the access to a particular cavity of interest. As a bonus – they can also access the statistical data provided during the basic analysis.

Still not convinced? Please check the answer to “Why does AQUA-DUCT functionality surpass that of other software?”

14.How can I validate a dynamic when there is no native bound, that is, co-crystallized with the protein?

We are still looking for an answer to this question. As for now, we were unable to verify the AQUA-DUCT’s results experimentally. However, we have shown that in the case of hot-spots, they can be used to detect the position of the catalytic water molecule in the SARS-CoV-2 Mpro [https://www.mdpi.com/1422-0067/21/9/3099/htm], and the buried water molecules in Solanum tuberosum epoxide hydrolase [https://www.mdpi.com/2218-273X/8/4/143]. Please take into account that the accuracy of the results strictly depend on the quality of your MD simulations. Therefore, we strongly recommend using dedicated software [https://www.sciencedirect.com/science/article/pii/S2001037019304556] to place water molecules properly inside your system prior to your MD simulation run, especially when poor quality crystal structures are provided.