Skip to content

MarcoSchaeferT/InVADo_setup

BranchesTags

Repository files navigation

InVADo: Interactive Visual Analysis of Molecular Docking Data

paper doi: 10.1109/TVCG.2023.3337642

video doi: 10.1109/TVCG.2023.3337642/mm1

video in Suppl. Mat.: https://ieeexplore.ieee.org/document/10334517/media#media

Replicability Stamp form the Graphics Replicability Stamp Initiative :
http://www.replicabilitystamp.org#https-github-com-marcoschaefert-invado-setup

InVADo_setup - Repository

contains the installation instructions and additional data to complement the main InVADo repository

main repository of InVADo: https://github.com/MarcoSchaeferT/megamol-prolint-InVADo

  • InVADo is available under the repository link above
  • download the repo and run CMake to prepare InVADo for building
    (see Requirements Building InVADo and Building InVADo)
  • running CMake will download the additional content of this repository
  • additional content is automatically saved to: InVADo_MAIN_REPO\plugins\prolint\InVADo_setup

additional data includes:

  • PLIP installation script
  • figures of functional groups (needed for web dashboard)
  • small precalculated data set (for web part development or when running the web dashboard without InVADo desktop)
  • small molecular docking pipeline
  • InVADo config file
  • graphics settings file

Requirements Building InVADo:

Building InVADo:

  • run CMake for MegaMol/InVADo (configure)

    • set visual studio 2022
    • set x64

  • ignore warnings during the 'configure' process

  • check/do the following boxes/steps:

    • ENABLE_MPI
    • set 'MPI_GUESS_LIBRARY_NAME' to 'MSMPI'
    • ENABLE_CUDA
    • BUILD_PROLINT_PLUGIN
    • configure again
    • generate
    • open project
    • (may need to install .NET Framework 4.8.1 SDK)

  • set the 'INSTALL' target of 'CMakePredefinedTargets' as the start project

  • build and install it as "RELEASE" not "DEBUG"

  • "DEBUG" is possible as well, but if the docking data set is not already preprocessed by InVADo as a RELEASE version it will fail to start as a DEBUG version

  • the docking data set will be preprocessed one time with the first start of InVADo as RELEASE version

InVADo Configuration

PATH = _InVADo_MAIN_REPO\plugins\prolint\InVADo_setup\

  • the InVADo config file InVADoConfig.mmprj is located in the folder 'PATH/InVADo_config'

  • in the InVADo config file, the path for the ligand list and the target protein file needs to be updated

  • this can be done manually in the XML-based file or graphically as follows:

  • manually:

    • open InVADoConfig.mmprj with a text editor and change the path stored in 'value' to e.g.:
    • <param name='pdbqtListFilename' value='C:\PATH\prepare_docking_data_scripts\results\results_list.txt' />
    • <param name="pdbFilename" value="C:\PATH\prepare_docking_data_scripts\7nn9_autoDockTools.pdbqt" />

  • graphically:

  • the InVADo config can be adjusted with the configurator PATH/InVADo_config/MegaMolConf.exe

  • drag InVADoConfig.mmprj and drop it onto MegaMolConf.exe

  • an error message will appear -> click ok -> then the following window will be shown:

  • click on the first fix button and navigate to _InVADo_MAIN_REPO/build/install/bin/ and select 'mmconsole.exe'

  • then click on the fix button of 'Core and Plugin State'

  • a new window appears -> there click the button 'Execute'

  • after the plugin scan is finished click on 'Use this State file'

  • a window appears where a path for saving the state file is already suggested -> use this suggested path and click 'Save'

  • close the configurator

  • again drag InVADoConfig.mmprj and drop it onto MegaMolConf.exe -> the following view will appear:

  • update the paths for the ligand list and the target protein as follows:

    • Module MultiPDBQTLoader1 Parameter: pdbqtListFilename: must be set to your path of a ligand PDBQT file list (*.txt)

      [e.g.:PATH\prepare_docking_data_scripts\results\results_list.txt*]

      ==> please note: 'results' is created after the test docking (see Create a New Data Set)

    • Module PDBLoader1: Parameter: pdbFilename: must be set to your path of a protein file (*.pdb)

      [e.g.: PATH\prepare_docking_data_scripts/7nn9_autoDockTools.pdbqt]

  • for easier development set 'INSTALL' target settings of the Visual Studio project as follows:

Create a New Data Set

PATH = _InVADo_MAIN_REPO\plugins\prolint\InVADo_setup\prepare_docking_data_scripts

  • the folder 'prepare_docking_data_scripts' contains a pipeline
  • this is an exemplary, simple pipeline for processing a molecular docking
  • a small example data set is included
    • PATH/data/AAABMN.xaa.pdbqt (includes the ligands)
    • PATH/7nn9_autoDockTools.pdbqt (is the target protein)
  • running the script PATH\run_full_automated_docking.cmd automatically performs a molecular docking of the data above
  • the docked data is then stored in the folder PATH/results
  • the example ligands are from the ZINC database: https://zinc15.docking.org/
  • the docking tool is AutoDock Vina: https://vina.scripps.edu/
  • if own ligands are used they should follow the ZINC naming scheme

Run InVADo:

PATH = _InVADo_MAIN_REPO\plugins\prolint\InVADo_setup\

  • InVADo can be started with mmconsole.exe

  • mmconsole.exe is stored after building in "_InVADo_MAIN_REPO/build/install/bin/"

  • example command for execution (program + config file): C:\Projects\InVADo\build\install\bin\mmconsole.exe -p "C:\PATH\InVADo_config\InVADoConfig.mmprj" -i Project_1 inst

  • the first start can take a while because InVADo preprocesses the data (loading bars in the cmd window will indicate the progress)

  • the processed data will be stored in [e.g.: PATH\prepare_docking_data_scripts\results\InVADo_tmpFiles]

Starting only the Web Dashboard

PATH = _InVADo_MAIN_REPO\plugins\prolint\server\

  • there is the possibility to run only the web part of InVADo without the 3D visualization
  • it will use the provided test data set
  • build the web app: run 'PATH\build_app.cmd'
  • start the web app: run 'PATH\run_app.cmd'

optional: improve rendering quality

PATH = _InVADo_MAIN_REPO\plugins\prolint\InVADo_setup\

  • got to folder 'graphicSettings' located in PATH
  • run setGraphics.cmd to set an Nvidia profile for InVADo

Trouble Shooting

  • if the build fails with: can not find _Py_wfopen() make sure Python 3.10 is installed and no other Python version is set for the Windows PATH variable

Acknowledgments

This work was partially funded by the German Research Foundation (DFG) as part of the Project No. 437702916 "Illustrative Visualization of Uncertainty in Dynamic Molecular Structures". It was also supported by the Czech Ministry of Education (grants ESFRI RECETOX RI LM2018121, ESFRI ELIXIR CZ LM2018131, e-INFRA LM2018140) and by the National Institute for Cancer Research (Programme EXCELES, ID Project No. LX22NPO5102) - Funded by the European Union – Next Generation EU.

Special thanks go to the experts from Loschmidt Laboratories, Masaryk University, Brno, for their participation in the evaluation and thorough and valuable feedback. We also thank Frank Böckler and his team from the Molecular Design and Pharmaceutical Biophysics group, University of Tübingen for providing task descriptions and feedback on the initial design of InVADo.

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

1 star

Watchers

1 watching

Forks

1 fork
Report repository

Releases

No releases published

Packages

No packages published

Languages