.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "examples/2-structure_map.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        :ref:`Go to the end <sphx_glr_download_examples_2-structure_map.py>`
        to download the full example code.

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_examples_2-structure_map.py:


Structure-property maps
=======================

This example demonstrates the visualization of structures (or environments) using
data-driven descriptors of their geometry, to cluster together similar motifs. Here the
geometric descriptors have been computed by PCA starting from SOAP representations, but
are provided as text files to avoid external dependencies for the example.

The same parameters can be used with `chemiscope.show` to visualize an interactive
widget in a Jupyter notebook.

.. GENERATED FROM PYTHON SOURCE LINES 16-22

.. code-block:: Python


    import ase.io
    import numpy as np

    import chemiscope








.. GENERATED FROM PYTHON SOURCE LINES 23-24

Load structures

.. GENERATED FROM PYTHON SOURCE LINES 24-27

.. code-block:: Python


    frames = ase.io.read("data/trajectory.xyz", ":")








.. GENERATED FROM PYTHON SOURCE LINES 28-31

Load the SOAP-PCA descriptors. chemiscope does not provide analysis routines, but you
can look up for instance scikit-matter as a package to do dimensionality reduction
analyses.

.. GENERATED FROM PYTHON SOURCE LINES 32-36

.. code-block:: Python


    pca_atom = np.loadtxt("data/trajectory-pca_atom.dat")
    pca_structure = np.loadtxt("data/trajectory-pca_structure.dat")








.. GENERATED FROM PYTHON SOURCE LINES 37-44

When both environments and structure property are present, a toggle allows you to
switch between both modes.

.. note::

    if there are properties stored in the ASE frames, you can extract them with
    chemiscope.extract_properties(frames)

.. GENERATED FROM PYTHON SOURCE LINES 45-60

.. code-block:: Python


    properties = {
        # concise definition of a property, with just an array and the type
        # inferred by the size
        "structure PCA": pca_structure,
        "atom PCA": pca_atom,
        # an example of the verbose definition
        "energy": {
            "target": "structure",
            "values": [frame.info["energy"] for frame in frames],
            "units": "eV",
            "description": "potential energy, computed with DFTB+",
        },
    }








.. GENERATED FROM PYTHON SOURCE LINES 61-62

Environment descriptors have only been computed for C and O atoms.

.. GENERATED FROM PYTHON SOURCE LINES 63-71

.. code-block:: Python

    environments = []
    cutoff = 4.0
    for frame_i, frame in enumerate(frames):
        for atom_i, atom in enumerate(frame.numbers):
            if atom == 6 or atom == 8:
                environments.append((frame_i, atom_i, cutoff))









.. GENERATED FROM PYTHON SOURCE LINES 72-73

Create a visualization and save it as a file that can be viewed at chemiscope.org

.. GENERATED FROM PYTHON SOURCE LINES 74-102

.. code-block:: Python


    chemiscope.write_input(
        "trajectory-pca.json.gz",
        # dataset metadata can also be included, to provide a self-contained description
        # of the data, authors and references
        meta={
            "name": "Allyl alcohol PCA map",
            "description": (
                "This dataset contains a PCA map of the C and O environments "
                "from a few frames out of a MD simulation of allyl alcohol, C3H5OH."
            ),
            "authors": ["The chemiscope developers"],
            "references": [
                (
                    "G. Fraux, R. Cersonsky, and M. Ceriotti, "
                    '"Chemiscope: interactive structure-property explorer for materials '
                    'and molecules," JOSS 5(51), 2117 (2020).'
                )
            ],
        },
        frames=frames,
        properties=properties,
        environments=environments,
        settings={  # these are reasonable settings for trajectory visualization
            "structure": [{"keepOrientation": True, "playbackDelay": 100}]
        },
    )








.. GENERATED FROM PYTHON SOURCE LINES 103-105

The file can also be viewed in a notebook. Use `chemiscope.show` above to bypass the
creation of a JSON file and directly create a viewer.

.. GENERATED FROM PYTHON SOURCE LINES 106-108

.. code-block:: Python


    chemiscope.show_input("trajectory-pca.json.gz")


.. chemiscope:: _datasets/fig_2-structure_map_003.json.gz
            :mode: default
        


.. raw:: html

    <div class="output_subarea output_html rendered_html output_result">

    </div>
    <br />
    <br />


.. rst-class:: sphx-glr-timing

   **Total running time of the script:** (0 minutes 0.020 seconds)


.. _sphx_glr_download_examples_2-structure_map.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: 2-structure_map.ipynb <2-structure_map.ipynb>`

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: 2-structure_map.py <2-structure_map.py>`

    .. container:: sphx-glr-download sphx-glr-download-zip

      :download:`Download zipped: 2-structure_map.zip <2-structure_map.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_