bundle_info.xml

<BundleInfo name="ChimeraX-LigandRecognizer"
	    version="0.2" package="chimerax.ligand_recognizer"
  	    minSessionVersion="1" maxSessionVersion="1">

  <!-- Additional information about bundle source -->
  <Author>Witold Taisner, Anna Przybyłowska, Jacek Karolczak, Dariusz Brzezinski, John M. Heumann</Author>
  <Email>taisner.witold@gmail.com</Email>
  <URL>https://github.com/wtaisner/chimerax-density-map-classification</URL>

  <!-- Synopsis is a one-line description is a full multi-line description -->
  <Synopsis>This tool uses a PDB model (.pdb or .cif) and corresponding density map (.cpp4) to validate/identify ligands.</Synopsis>
  <Description>
    This tool implements the functionality proposed in the publication "Ligand Identification in CryoEM and X-ray Maps Using Deep Learning"
    (https://doi.org/10.1093/bioinformatics/btae749) that uses a deep learning model to predict ligand types.
    The prediction is performed on a server (https://ligands.cs.put.poznan.pl/) based on a preprocessed map fragment sent from ChimeraX.

    Usage:
    This tool assumes that both a .pdb/.cif model and a .ccp4 map are loaded into ChimeraX.
    A key requirement is that the map is a map-versus-model difference map in which high density suggests unmodeled ligands.
    For cryoEM, the difference map can be the map_maodel_difference_1.ccp4 created with the Phenix command
    phenix.real_space_diff_map $MODEL $MAP resolution=$RES, where $MODEL is the ligand-free model, $MAP is the experimental map, and $RES
    is the model resolution in Angstroms. Alternatively, in Linux or a Linux-like environment, the provided computeMapModelDifference.sh
    script can be invoked as computeMapModelDifference.sh $DIR $MAP $MODEL, which will extract the resolution from the model, run
    phenix.real_space_diff_map, and then delete map_model_difference_2.ccp4 which is not needed. For X-ray crystallography data, the
    Fo-Fc map can be used. In either case, the model used for creating the difference map not contain the ligands of interest.

    The tool implements three basic commands ([] signifies optional arguments):
    1. blob autothreshold [map_id] [style] [density_std_threshold]: set a surface-defining nthreshold for the map / volume and choose display options..
    2. blob recognize [map_id] [surface_id] [pdb_id] [flg_xray] [density_threshold]: tries to identify a ligand from a selected map fragment (surface).
    3. blob validate res_id [map_id] [pdb_id] [flg_xray] [density_threshold]: validates a previously modeled ligand at residue res_id.

    Blob_recognize and blob_validate are alisses for blobus validatus, blobus recognitus, and blobus autothresholdus, respectively. Each command
    runs the same prediction algorithm; they differ only in how the blob to be analyzed is identified.
    
    Examples:
    - blob autothreshold style 2 (sets an automatic threshold on the displayed map and changes the display style of the map and model)
    - blob validate /A:1401 (validates a ligand at residue 1401 in chain A using the default map and model)
    - blob recognize #2 #4.1 (recognize a ligand using map #2 and a user-selected blob-limiting surface with id #4.1)
    - blob validate /A:1401 flg_xray 1 (validates a ligand at residue 1401 in chain A using the default map and model and treats the map as an X-ray difference map)

    For more details, please see tutorial available at https://github.com/wtaisner/chimerax-ligand-recognizer/blob/main/tutorial.pdf.
  </Description>

  <!-- Categories is a list where this bundle should appear -->
  <Categories>
    <Category name="General"/>
  </Categories>


  <!-- DataFiles is a list of additional files to include in bundle -->
  <DataFiles>
    <DataFile>docs/user/commands/blob.html</DataFile>
    <DataFile>docs/user/commands/blobus.html</DataFile>
  </DataFiles>

  <!-- Dependencies on other ChimeraX/Python packages -->
  <Dependencies>
    <Dependency name="ChimeraX-Core" version="~=1.1"/>
    <Dependency name="numpy"/>
    <Dependency name="scipy"/>
  </Dependencies>

  <!-- Python and ChimeraX-specific classifiers
       From https://pypi.python.org/pypi?%3Aaction=list_classifiers
       Some Python classifiers are always inserted by the build process.
       These include the Environment and Operating System classifiers
       as well as:
         Framework :: ChimeraX
         Intended Audience :: Science/Research
         Programming Language :: Python :: 3
         Topic :: Scientific/Engineering :: Visualization
         Topic :: Scientific/Engineering :: Chemistry
         Topic :: Scientific/Engineering :: Bio-Informatics
       The "ChimeraX :: Bundle" classifier is also supplied automatically.  -->
  <Classifiers>
    <!-- Development Status should be compatible with bundle version number -->
    <PythonClassifier>Development Status :: 3 - Alpha</PythonClassifier>
    <PythonClassifier>License :: Freeware</PythonClassifier>
    <!-- ChimeraX classifiers describe supplied functionality -->
    <ChimeraXClassifier>ChimeraX :: Command :: blob validate :: General ::
	predicts class of a modeled ligand </ChimeraXClassifier>
    <ChimeraXClassifier>ChimeraX :: Command :: blobus validatus :: General ::
	predicts class of a modeled ligand </ChimeraXClassifier>
    <ChimeraXClassifier>ChimeraX :: Command :: blob recognize :: General ::
	predicts a ligand for a selected map fragment </ChimeraXClassifier>
    <ChimeraXClassifier>ChimeraX :: Command :: blobus recognitus :: General ::
	predicts a ligand for a selected map fragment </ChimeraXClassifier>
    <ChimeraXClassifier>ChimeraX :: Command :: blob autothreshold :: General ::
	sets a threshold on the displayed map (volume) </ChimeraXClassifier>
    <ChimeraXClassifier>ChimeraX :: Command :: blobus autothresholdus :: General ::
	sets a threshold on the displayed map (volume) </ChimeraXClassifier>
  </Classifiers>

</BundleInfo>