- Contents
This repository contains the data files (see releases for a data.zip with tracings annotations) as well as source code required for the analysis. In case of questions or difficulties running the code, please contact joergen.kornfeld@bi.mpg.de
- Provided data
2.1 Registered raw EM dataset
A streaming configuration file (j0256.conf) is provided for KNOSSOS (www.knossostool.org) that can be used to easily access all image data and perform your own tracings in the dataset. Additionally raw data hostings (e.g. on neurodata.io) will also be listed here as soon as they are available.
The configuration file can also be used together with the knossosDataset class provided in the public knossos_utils package to directly work with the dataset (e.g. download it, convert to an image stack, etc).
2.2 Skeleton files
All annotations are provided in a zip file, that contains the necessary data to run the functions of HVC_RA_analysis.py. The .k.zip files can also be opened directly in KNOSSOS for further analysis. Note that all postsynaptic tracings are contained within the axon k.zips.
- Source code
3.1 License
All source code is licensed under the GPL v2 license. Copyright, J. Kornfeld, Max Planck Society
3.2 Dependencies
The provided Python source code requires some additional libraries to be installed to run.
-
knossos_utils (see https://github.com/knossos-project/knossos_utils for installation instructions)
-
numpy (pip install numpy)
-
matplotlib (pip install matplotlib)
-
networkx (pip install networkx)
-
pandas (pip install pandas)
-
mayavi (pip install mayavi)
All pip-installable libraries come with anaconda, it is therefore advised to install the anaconda Python distribution.