Plotting

PySB could inform the results of a simulation to dataframes (See Simulation) and visualization of results could be done with matplotlib or seaborn even (See more here). To access the data, the dataframes columns reproduce the names of the Observables. The following example could be adapted to show the dynamics of any Observable.

Note

Importantly, PySB allows the inspection of the model to find which Monomers (and complexes of monomers) exists in the model, but as the simulation is network-free, the possible formed complexes are up to the user concern.

Note

Atlas produces automatically Observables for metabolites, and other components and complexes could also be observed and plotted, but their declaration in the model is entirely up to the user.

fig, ax = plt.subplots(1, 2, figsize = (4*2, 3*1), dpi = 100)

# ax[0].plot(data1.index, data1['obs_alpha_L_arabinopyranose_cyt'], label = '__NOLABEL__', color = palette[0])
# ax[0].plot(data1.index, data1['obs_ATP_cyt'], label = '__NOLABEL__', color = palette[3])
ax[0].fill_between(avrg.index,
        avrg['obs_alpha_L_arabinopyranose_cyt'] + stdv['obs_alpha_L_arabinopyranose_cyt'],
        avrg['obs_alpha_L_arabinopyranose_cyt'] - stdv['obs_alpha_L_arabinopyranose_cyt'],
        label = r'$\alpha$-arabinopyranose', **{'color' : palette[0], 'alpha' : 0.5})
ax[0].fill_between(avrg.index,
        avrg['obs_ATP_cyt'] + stdv['obs_ATP_cyt'],
        avrg['obs_ATP_cyt'] - stdv['obs_ATP_cyt'],
        label = r'ATP', **{'color' : palette[3], 'alpha' : 0.5})

# ax[1].plot(data1.index, data1['obs_PROTON_cyt'], label = '__NOLABEL__', color = palette[0])
# ax[1].plot(data1.index, data1['obs_XYLULOSE_5_PHOSPHATE_cyt'], label = '__NOLABEL__', color = palette[3])
ax[1].fill_between(avrg.index,
        avrg['obs_PROTON_cyt'] + stdv['obs_PROTON_cyt'],
        avrg['obs_PROTON_cyt'] - stdv['obs_PROTON_cyt'],
        label = r'H$^+$', **{'color' : palette[0], 'alpha' : 0.5})
ax[1].fill_between(avrg.index,
        avrg['obs_WATER_cyt'] + stdv['obs_WATER_cyt'],
        avrg['obs_WATER_cyt'] - stdv['obs_WATER_cyt'],
        label = 'WATER', **{'color' : palette[1], 'alpha' : 0.5})
ax[1].fill_between(avrg.index,
        avrg['obs_XYLULOSE_5_PHOSPHATE_cyt'] + stdv['obs_XYLULOSE_5_PHOSPHATE_cyt'],
        avrg['obs_XYLULOSE_5_PHOSPHATE_cyt'] - stdv['obs_XYLULOSE_5_PHOSPHATE_cyt'],
        label = r'xylulose 5-phosphate', **{'color' : palette[3], 'alpha' : 0.5})

ax[0].set_xlabel('Time [A.U.]')
ax[0].set_ylabel('Concentration [A.U.]')
# ax[0].set_xlim(left = 0, right = 100)
ax[0].set_ylim(bottom = 0, top = 200)

ax[1].set_xlabel('Time [A.U.]')
# ax[1].set_xlim(left = 0, right = 100)
ax[1].set_ylim(bottom = 0, top = 200)

ax[0].legend(frameon = False)
ax[1].legend(frameon = False)

seaborn.despine()
plt.savefig('Fig_Arabinose.png', format = 'png', bbox_inches = 'tight', dpi = 350)
# for publication
# plt.savefig('Fig_Arabinose.pdf', format = 'pdf', bbox_inches = 'tight', dpi = 350)

plt.show()

And the results is

Note

See the Arabinose Model to inspect the rules and reproduce (at some extent because of stochasticity) the plot showed in this Manual.