Repository providing python code to load, visualize, analyze, and plot Map Manager files. This can be done in three ways

• Pure Python code using the bMapManager.py class library
• Using the GUI of stackbrowser.py
• Creating web based plots and analysis using bokeh.py

Map Manager class library

See [bMapManager][bMapManager]. Include the following in a Python script...

from bMapManager import bMap
from bMapManager import bStack
from bMapManager import bStackPlot

To run inside an ipython/jupiter notebook, also include

%matplotlib inline
from plotly.offline import init_notebook_mode
init_notebook_mode() # run at the start of every ipython notebook to use plotly.offline

Requires numpy, pandas, tifffile

See [map.ipynb][1] for more sample code.

Python based stack browser

A full GUI stack browser implemented in Python using [pyqt][pyqt]. Run the browser with...

python stackbrowser2.py

The next image shows the stack browser interface. A single image plane of a 3D stack is displayed with an overlay of stack db points (red spines). The bottom table shows the stack db.

Eventual features are:

• Load a single timepoint from Map Manager
• 3D stack can be scrolled (image slices), panned, and zoomed
• Image contrast can be adjusted
• Display stack db points overload on image and as a table
• Clicking a point in the image will select it in the table and visa-versa
• Provide x/y plots. Clicking on a point in the plot will propogate to other views
• Provide statistics (mean, standard error, and n) for any plot.

Required libraries

numpy
pandas
pyqt #see below
tifffile #see below

Install tifffile

pip install tifffile

Install pyQT on OSX

brew install pyqt

Installing additional image readers

The QT .jpg reader is not installed by default. Just use .png until this is figured out.

Web browser plotting of Map Manager files using bokeh

Use the [bokeh][bokeh] python library to generates html pages with a table and x/y plots from a Map Manager file. Selecting points in one plot/table will propagate the selected points to other plots/tables.

python bokeh1.py

Eventual features are:

• Make a Flask app where a Map Manager map or single timepoint file can be loaded.
• Interface to select x/y data from table and then make a new plot.
• On selection, report mean/sd/se/n for selection in each plot
• Provide more advanced plotting and statistics using pandas pivot tables. For example, plot all sLen2d and then create a visual mask and output statistics for each segment.

[bokeh]: http://bokeh.pydata.org/en/latest/ [bMapManager]: https://github.com/cudmore/mapmanager_ipynb/blob/master/bMapManager.py [pyqt]: https://riverbankcomputing.com/software/pyqt/intro [1]: https://github.com/cudmore/mapmanager_ipynb/blob/master/map.ipynb