Intrdocution
Installation
The latest stable release is available on PyPi (and on GitHub), and you can install it by saying
(.venv) $ pip install bacteria
To build Bacteria from source, say python setup.py build.
Then, to install Bacteria, say python setup.py install.
If all went well, you should be able to execute the demo scripts (read docs)
(OS X users should follow the installation guide given below).
Alternatively, you can download or clone the repository and use pip to handle dependencies:
(.venv) $ unzip bacteria.zip
(.venv) $ pip install -e bacteria
or
(.venv) $ git clone https://github.com/tuliofalmeida/bacteria
(.venv) $ pip install -e bacteria
By calling pip list you should see bacteria now as an installed package:
bacteria (1.x.x, /path/to/bacteria)
IO
The library’s input is the SuperSegger data, the clist, and the organization used in pandas DataFrame. Its possible to load one clist and extract the 2D and 3D data and also concatenate clists from different fields of view into a single dataframe.
Single data
To load data from a single field of view
>>> import bacteria as bac
>>> path_2d = "data/2d_clist.mat"
>>> path_3d = "data/3d_clist.mat"
>>> df2d = bac.df_data2d(path_2d,fps = 3)
>>> df3d = bac.df_data3d(path_3d,fps = 3)
Note
The fps parameter is to create the time column in minutes.
Concatenate data
To load and concatenate multiple field of view data from a single experiment, if the data is organized inside the SuperSegger folder, use:
>>> path = "/SuperSegger Data/Exp001/"
>>> df2d,df3d,_ = bac.concatenate_clist(path)
Note
By concatenating the data, it is possible to identify the cells of each FOV and the Cell ID is modified to remain unique, the IDs are a continuous count at each FOV.
Or you can pass it the path to a folder with all the clists of the experiment
>>> path_clists = "/SuperSegger Data/clists folder/"
>>> df2d,df3d,_ = bac.concatenate_clist(path_clists, direct = True)
Note
Is it possible to save this arrangement in a .mat file (this arrangement is different from the concatenation done by SuperSegger)
Filters
The library has 2 filters implemented along with the SuperSegger filter. The data is filtered using the “stat0” column of SupperSegger (applied in the functions of reading data directly from the .mat file), size filter (excludes very long cells) and the mother and daughter filter, which will exclude cells that were born smaller than 40% the mother’s size or greater than 60% of the mother’s size.
>>> df2d_f,df3d_f,_ = bac.combined_filters(df2d,df3d)
Note
The function returns a histogram with the effect of each filter on the data and it is possible to adjust some parameters and select whether all filters will be applied.
Hint!
After loading, concatenating and filtering the data, it is interesting to save them in a “.csv” file to optimize the analysis, leaving the original data intact and making data loading faster. For that use:
>>> import pandas as pd
>>> df2d_f.to_csv('/data/"2D_filtered.csv",index=False)
>>> df3d_f.to_csv('/data/"3D_filtered.csv",index=False)
Colab Tutorial
Check out the tutorial on Google Colab!
