Skip to content

csoeller/PYME-extra

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

586 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

README

Plugins and associated code for PYME (AKA python-microscopy).

This is a mixed bag of extensions/additions to standard PYME and also used as a testbed/platform for

  • recipe based processing of SMLM data
  • implementing I/O for new formats (especially MINFLUX data in .npy and (now preferred) .zarr.zip format)
  • new ideas for data processing

Installation

PYME-test-env

These days we recommend for any development install of PYME-extra the installation as part of a PYME-test-env controlled install. All further details please see there.

pip install

For users not interested in following development updates and mainly interested in the stable release for usage as is we now recommend a pip based install.

We highly recommend installing into a fresh virtual environment as can be generated with conda and related tools:

  1. if you don't yet have it, download and install Miniconda or miniforge.

  2. create and activate a new conda environment with python 3.10 to 3.13 (3.10 and 3.11 are probably the most well tested at present)

    conda create -n pyme-pip python=3.10
    conda activate pyme-pip

Now you are ready to use pip to install python-microscopy and PYME-extra (# is the comment sign on macOS/posix shells, just leave out that text when pasting to windows command prompts):

	# possibly install python-microscopy first and check that the install succeeds
	pip install python-microscopy
	pip install PYME-extra # installation from PyPi
	pymex_install_plugins # important final step: register the plugins systemwide

This should be all, at this stage you can launch the main applications, i.e. visgui (AKA PymeVis or PYMEVisualize) and dh5view (AKA PYMEImage) to open and process microscopy data. See also Verify Installation in the PYME docs.

Installing from source

As another alternative you can install directly from the source (e.g. as obtained from github). Still achieved with pip but now called from within the source directory in which you unpacked PYME-extra (typically done when you cloned the git repository). A plain install from source is done with

	pip install .

A development install can be achieved by requesting an install in "editable mode". When a package is installed in editable mode, edits to the project source code become effective without the need of a new installation step.

	pip install --no-build-isolation -e . # install in in “development mode”

In either case (plain or development install), you may need to register the various plugins to implement the extra functionality provided by PYME-extra. This is achieved with the plugin installer that will have been installed with PYME-extra. You register with the command

	pymex_install_plugins

By default it registers the plugins systemwide but you can supply the --user option to register only for the current user:

	pymex_install_plugins --user # for further details see also pymex_install_plugins -h

Requirements

External modules required for full functionality currently include

python-microscopy
statsmodels # for FRC smoothing with the lowess filter
roifile     # to allow using info from ImageJ/Fiji ROIs
colorcet    # add some colorcet colour tables in PYME
circle-fit  # needs pip install to get recent version; for 2D NPC analysis
alphashape # for cluster area and densities in clusters
zarr>=2,<3 # for MINFLUX I/O
seaborn # for some prettier plots
mrcfile # to output 3D data for FSC from a EM FSC server

These should be installed by the pip based install automatically.

We also often use a couple more dependencies in notebooks, but strictly speaking no functionality in PYME-extra depends directly on these:

openpyxl
tabulate

Issues

Note that the showErrs modules rely on a mac installation which uses bash scripts and the platypus app app to capture STDERR into a temporary file which these modules access.

Bottom line is that these two error display modules will likely not work on anything but a mac with my PYMEapps wrappers. On other systems they will just generate a message that this functionality is not supported.

The PYME mac app wrappers are available at the PYME-apps repository.

Author

Christian Soeller

About

extensions and plugin additions for python-microscopy (PYME)

Topics

Resources

License

Stars

Watchers

Forks

Packages

 
 
 

Contributors

Languages