API

This section has the code examples to do a complete analysis using the library.

Imports and Data

For execution we will use the following imports:

>>> import bacteria as bac
>>> import pandas as pd
>>> import matplotlib.pyplot as plt

We will use the example data available on GitHub:

>>> # we use '!' to execute linux commands in notebooks
>>> !git clone https://github.com/tuliofalmeida/bacteria
>>> df2d = pd.read_csv('/content/bacteria/data_tutorial/df2d_filtered.csv')
>>> df3d = pd.read_csv('/content/bacteria/data_tutorial/df3d_filtered.csv')

Note

This data is not representative about any experiment, it has a shift in the growth medium.

Experiment in time

Calculating the instantaneous volume change (Growth Rate) and fluorescence.

>>> gr_df,_ = bac.derivative(df3d,column='Volume')
>>> fluo_df,_ = bac.derivative(df3d,column='Fluor1 sum')

Calculating the average of all cells for different columns from derivative and 3D data

>>> time_gr,mean_gr,ci_gr = bac.column_mean(gr_df,column='Derivative')
>>> time_fluo,mean_fluo,ci_fluo = bac.column_mean(fluo_df,column='Derivative')
>>> time_div,mean_div,ci_div = bac.column_mean(df3d,column='Long axis (L)')
>>> time_ratio,mean_ratio,ci_ratio = bac.column_mean(df3d,column='F/V')

Rearranging 2D data based on division time to plot the data over time

>>> df_2d_temp = pd.melt(df2d, id_vars=['Cell ID','Cell age','Area death','Time Division'], value_vars=['Vd-Vb']).sort_values(by=['Time Division'])

Entire time plot

Note

The plot codes are available on Google Colab.

Pre and pos shift plots

Transition plots

Cell Cycle Analysis

First we will extract the Cell ID’s pre and pos shift

>>> cells_pre = bac.cells_pre_shift(df3d,pre=600)
>>> cells_pos = bac.cells_pos_shift(df3d,pos=800)

Bin the data for all the entire time

>>> # Volume derivative
>>> bins_fluo_vol_norm,ci_fluo_vol_norm = bac.derivative_binning(fluo_df,derivative_column='Derivative/V',sort_by='Volume',print_bins=True)
>>> bins_fluo_vol,ci_fluo_vol = bac.derivative_binning(fluo_df,derivative_column='Derivative',sort_by='Volume',print_bins=False)
>>> # Fluorescence derivative
>>> bins_fluo_cc_norm,ci_fluo_cc_norm = bac.derivative_binning(fluo_df,derivative_column='Derivative/V',sort_by='Cell Cycle',print_bins=False)
>>> bins_fluo_cc,ci_fluo_cc = bac.derivative_binning(fluo_df,derivative_column='Derivative',sort_by='Cell Cycle',print_bins=False)
>>> # Fluorescence/Volume Ratio
>>> bins_ratio_vol, ci_ratio_vol = bac.bin_column(df3d,column = 'F/V',sort_by = 'Volume',print_bins=False)
>>> bins_ratio_cc, ci_ratio_cc = bac.bin_column(df3d,column = 'F/V',sort_by = 'Cell Cycle',print_bins=False)

Bin plots entire time

Pre plots

>>> # Volume derivative pre
>>> bins_fluo_vol_norm,ci_fluo_vol_norm = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pre)],derivative_column='Derivative/V',sort_by='Volume',print_bins=False)
>>> bins_fluo_vol,ci_fluo_vol = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pre)],derivative_column='Derivative',sort_by='Volume',print_bins=False)
>>> # Fluorescence derivative pre
>>> bins_fluo_cc_norm,ci_fluo_cc_norm = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pre)],derivative_column='Derivative/V',sort_by='Cell Cycle',print_bins=False)
>>> bins_fluo_cc,ci_fluo_cc = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pre)],derivative_column='Derivative',sort_by='Cell Cycle',print_bins=False)
>>> # Fluorescence/Volume Ratio pre
>>> bins_ratio_vol, ci_ratio_vol = bac.bin_column(df3d[df3d['Cell ID'].isin(cells_pre)],column = 'F/V',sort_by = 'Volume',print_bins=False)
>>> bins_ratio_cc, ci_ratio_cc = bac.bin_column(df3d[df3d['Cell ID'].isin(cells_pre)],column = 'F/V',sort_by = 'Cell Cycle',print_bins=False)

Pos plots

>>> # Volume derivative pos
>>> bins_fluo_vol_norm,ci_fluo_vol_norm = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pos)],derivative_column='Derivative/V',sort_by='Volume',print_bins=False)
>>> bins_fluo_vol,ci_fluo_vol = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pos)],derivative_column='Derivative',sort_by='Volume',print_bins=False)
>>> # Fluorescence derivative pos
>>> bins_fluo_cc_norm,ci_fluo_cc_norm = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pos)],derivative_column='Derivative/V',sort_by='Cell Cycle',print_bins=False)
>>> bins_fluo_cc,ci_fluo_cc = bac.derivative_binning(fluo_df[fluo_df['Cell ID'].isin(cells_pos)],derivative_column='Derivative',sort_by='Cell Cycle',print_bins=False)
>>> # Fluorescence/Volume Ratio pos
>>> bins_ratio_vol, ci_ratio_vol = bac.bin_column(df3d[df3d['Cell ID'].isin(cells_pos)],column = 'F/V',sort_by = 'Volume',print_bins=False)
>>> bins_ratio_cc, ci_ratio_cc = bac.bin_column(df3d[df3d['Cell ID'].isin(cells_pos)],column = 'F/V',sort_by = 'Cell Cycle',print_bins=False)

Lineages Analysis

Calculate all lineages in the dataset

>>> reverse_lineages, large_lineages, large_nodes , all_nodes = bac.lineages(df2d)

Extract the longer lineages in dict and array

>>> filtered_lineage = bac.lineage_corr_filter(df3d,reverse_lineages)

Extract lineages start pre and pos the shift

>>> lineage_pre = bac.lineages_pre_shift(df3d,reverse_lineages, shift = 700)
>>> lineage_pos = bac.lineages_pos_shift(df3d,reverse_lineages, shift = 600)

Plot one cell mother lineage and the correlation between the nodes

>>> bac.plot_corr_lineage(df3d,reverse_lineages,filtered_lineage[5][0])

The order is a paramater that will influence the amount of noise accepet to estimate the smoothed derivative Fluorescence concentration pre shift (using order = 5)

Fluorescence concentration pos shift (using order = 5)

Fluorescence rate pre shift (using order = 5)

Fluorescence rate pos shift (using order = 5)

Minima Analysis

>>> diff_dict = bac.diff_minima(df3d,df2d,pre = 600,pos = 800,order = 3, adjust = False)

Plot Minima Volume by Volume at birth

Plot Daughther Minima Volume - Mother Minima Volume

>>> diff_dict = bac.plot_diff_minima(diff_dict,color='purple')

Plot minima lineage by volume (order = 3)

Colab Tutorial

All codes and plots are available on a notebook at Google Colab!

Here is a template for performing this analysis

Here is an example of estimating the mean and CI of a column from 2D data