Request

Make it dead easy to try panel-chemistry. I would like a CLI tool panel-chemistry with the commands

• hello which panel serves the example notebooks
• examples which copies the example notebooks to /examples/panel-chemistry.

Motivation

Currently there is some friction to try out panel-chemistry. You have to open Binder (slow) or download the notebooks from github. It takes some time and some users might now how to do or understand its possible. You can also git clone the repo. There are 4 commands you have to execute and it might fail.

Drawbacks

It would add dependencies like Typer and Rich to the package if we implement it like the awesome-panel-cli.

Alternatives

• Not do it.
• I've also considered whether the awesome-panel-cli should enable plugins. So having one cli to collect everything. But I think

Additional Context

We should do this in a general way that is consistent with awesome-panel-cli, panel-highcharts etc.

We should deploy a first, simple version of the package in order to "reserve" the package name on PyPi.

The plan is to use a setup.py file, twine and manual upload similar to what is done at panel-highcharts.

Chemical gallery viewer seems very interested.
Example: https://github.com/cbouy/mols2grid

Do you think it is possible to add it to panel-chemistry?

PyHDX has an existing NGL Molecule Viewer which is ok to copy or get inspiration from.

https://github.com/Jhsmit/PyHDX/blob/master/dev/bokeh_extension.py

https://twitter.com/Jhsmit_/status/1398233190519693317

We also need to setup the project to support pre-built Bokeh Extensions.

This is described in https://awesome-panel.readthedocs.io/en/latest/guides/awesome-panel-extensions-guide/bokeh-extensions.html#prebuilt-bokeh-extensions. If possible we should record and distribute this as a video to make it easier for the Panel Community to build new packages/ extensions in the future.

The JSME Test Page show additional parameters that could be added. These are marked with a dot below.

Hi, I've been working with the pdbe molstar code and I think it's fantastic. I'm a little confused about how static files would work for panel in a jupyter notebook though. Is there a way to specify an assets directory in a notebook without running panel serve --static-dirs from the command line?

When testing on panel 0.13.0rc5 I get

2022-04-07 21:06:14,769 Error running application handler <bokeh.application.handlers.script.ScriptHandler object at 0x0000029BC5655100>: cannot import name 'string_types' from 'panel.util' (c:\repos\private\awesome-panel\.venv\lib\site-packages\panel\util.py)
File 'ngl_viewer.py', line 15, in <module>:
from panel.util import lazy_load, string_types Traceback (most recent call last):
  File "c:\repos\private\awesome-panel\.venv\lib\site-packages\bokeh\application\handlers\code_runner.py", line 231, in run
    exec(self._code, module.__dict__)
  File "C:\repos\private\awesome-panel\awesome_panel\apps\py3dmol.py", line 4, in <module>
    from panel_chemistry.pane import Py3DMol
  File "c:\repos\private\awesome-panel\.venv\lib\site-packages\panel_chemistry\pane\__init__.py", line 6, in <module>
    from .ngl_viewer import NGLViewer
  File "c:\repos\private\awesome-panel\.venv\lib\site-packages\panel_chemistry\pane\ngl_viewer.py", line 15, in <module>
    from panel.util import lazy_load, string_types
ImportError: cannot import name 'string_types' from 'panel.util' (c:\repos\private\awesome-panel\.venv\lib\site-packages\panel\util.py)

Description

See holoviz/panel#3997

I was following the instructions from video 0.1:

git clone https://github.com/MarcSkovMadsen/panel-chemistry.git
 cd panel-chemistry
 conda create --name panel-chemistry
 conda activate panel-chemistry
 pip install -e .[all]

this resulted in an error for pywinpty

 Cargo, the Rust package manager, is not installed or is not on PATH.
    This package requires Rust and Cargo to compile extensions.

As chemo-informaticians nowadays who use Python, are mainly relied on RDKit to compute about molecule information, it is preferable that that RDKit could be easily bundled in panel-chemistry. But, since RDKit installation with pip is really tedious, registering panel-chemistry in Anaconda repository could help users to use it easily.

Request

It would be nice to have Ketcher component in Panel, with bidirectional communication of SMILES strings (Python -> Editor and Editor -> Python), as well as editor's Selection and Highlight.

Motivation

An easy-to-use editor could enable some nice applications: calculate properties as user draws molecules (e.g. GT4SD), enable AI copilot-style applications where AI proposes modifications, or collect human feedback on which parts of molecules would be "flagged" by chemists for undesirable properties.

Drawbacks

No response

Alternatives

Ketcher is quite similar to JSME, however, JSME interface "feels" quite outdated. There is Ketcher in Streamlit and Ketcher in Gradio (and Gradio implementation looks surprisingly short).

Additional Context

I got a few questions. Would panel-chemistry be a suitable "home" for Ketcher Panel component, or should it have its own repo? From PR #41 it looks like panel-chemistry does not yet support the latest Panel version, if that is correct, how difficult is it to upgrade?

Request

Align the panel-chemistry intro video

with the awesome-panel-cli video

Motivation

• I think the awesome-panel-cli video communicates better.
• I would like the same look and feel across awesome-panel projects.

Drawbacks

Takes time.

Alternatives

Keep it as it is.

Additional Context

The panel-chemistry video is

https://github.com/awesome-panel/panel-chemistry/blob/main/assets/panel-chemistry-teaser.gif

The awesome-panel-cli video is

https://github.com/awesome-panel/awesome-panel-cli/blob/main/assets/videos/awesome-panel-cli-intro.gif

The structure of the project will be the same as the panel-highcharts project and package until something else is decided. I.e. we use

• binder to provide an easy to use environment for trying, learning, exploring and using the packing. See.
• Python Invoke to easily run build, test etc tasks. See
  • This includes includes running isort, autoflake, black, pylint, mypy and pytest and making sure all test should pass before merging to main branch. This will keep the code and package working and maintainable.
• Provide documentation via this README and an examples folder of notebooks. See

Maybe later we improve it further

• Add CI/ CD on Github to build, test and deploy the package. (I have not tried this before)
• Add various badges (I have not tried this before).
• Add documentation on read the docs. But not for now in order to keep things simple.