import numpy as np
from dlsmicro.backend import analysis_tools
from dlsmicro.backend import io
from dlsmicro.backend import utils
import pandas as pd
import matplotlib.pyplot as plt
[docs]def analyze_conditions(csv_name, root_folder, condition_dir,
replicate_dict, T, r, erg, Laplace=False,
df_save_path=None, df_file_name=None,
save_as_text=True, save_as_df=True,
plot_corr=False, plot_msd=False, plot_G=False,
save_plots=False):
""" Analyze files exported from Zetasizer software for multiple
conditions and plot data per replicate of a condition.
Parameters
----------
csv_name : str
Name of every csv file (all the same name)
root_folder : str
Name of folder in which all files to be probed are
condition_dir : dictionary
Dictionary of conditions and respective folders
replicate_dict : dictionary
Dictionary of replicates for each condition
T : float or dictionary of float
Temperature of the experiment/condition in Kelvin
r : float or dictionary of float
Radius of particle in experiment/condition in nanometers
erg : boolean or dictionary of boolean
Ergodicity in experiment/condition
Laplace : boolean, `optional`
If `True`, use direct Laplace transform to find
shear modulus. This is useful because it can
smooth the data to noise.
df_save_path : str, `optional`
Path to Dataframe to be saved containing results
from DLS microrheology analysis.
If `None`, path will be set to root_folder.
df_file_name : str, `optional`
Name of Dataframe to be saved containing results
from DLS microrheology analysis.
If None, `condition_data.pkl` is default name
save_as_text : boolean, `optional`
If `True`, save each element of Dataframe separately
as a text file
save_as_df : boolean, `optional`
If `True`, save the Dataframe
plot_corr : boolean, `optional`
If `True`, show plot of the correlation function of
each replicate
plot_msd : boolean, `optional`
If `True`, show plot of the mean-squared displacement
of each replicate
plot_G : boolean, `optional`
If `True`, show plot of the shear modulus of each replicate
save_plots : boolean, `optional`
If `True`, saves plots of correlation function, MSD, G
"""
conditions = list(condition_dir.keys())
if type(T) is dict:
if T.keys() != condition_dir.keys():
raise Exception('Keys for temp dictionary do not match conditions')
T_dict = T
else:
T_dict = dict((k, T) for k, v in condition_dir.items())
if type(r) is dict:
if r.keys() != condition_dir.keys():
raise Exception('Keys for radius dictionary do not match conditions')
r_dict = r
else:
r_dict = dict((k, r) for k, v in condition_dir.items())
if type(erg) is dict:
if erg.keys() != condition_dir.keys():
raise Exception('Keys for ergodicity dictionary do not match conditions')
erg_dict = erg
else:
erg_dict = dict((k, erg) for k, v in condition_dir.items())
if df_save_path == None:
df_save_path = root_folder
if df_file_name == None:
df_file_name = 'condition_data.pkl'
# Define scattering vector parameters about solvent
n = 1.333 # index of refraction of water (default)
theta = 173.*np.pi/180.
lam = 633.
####################################################
# Analyze data
# Don't edit this unless you know what you're doing
####################################################
# Create pandas dataframe to organize replicate data
df = pd.DataFrame(columns=['t', 'msd_smooth', 'alpha', 'omega', 'G1', 'G2',
'replicate', 'condition', 'id', 'scattering', 'epos'])
idx = 0
for condition in conditions:
for replicate in replicate_dict[condition]:
file_path = '%s/%s/replicate%s/%s' % (root_folder,
condition_dir[condition],
replicate, csv_name)
# Read the data and truncate over the desired time windows
data_dict = io.read_zetasizer_csv_to_dict(file_path, 0)
[t, g, I, Ie, point_pos, epos] = list(data_dict.values())
# Figure out where data is no longer trustworthy (correlation function goes to zero)
tinds = [3, len(g)]
tinds[1] = np.argmax(g<0.05)
if tinds[1] == 0:
tinds[1] = len(g) - 1
g = g[tinds[0]:tinds[1]]
t = t[tinds[0]:tinds[1]]
# Get a DLS microrheology object that contains all raw data
# and analyzed results.
T = T_dict[condition]
r = r_dict[condition]
ergodic = erg_dict[condition]
q = analysis_tools.calc_q(n, theta, lam)
dlsmicro_df = analysis_tools.full_dlsur_analysis(t, g, ergodic, r, T, q,
I, Ie)
# Store the scattering vs. position data
scattering = np.zeros(len(dlsmicro_df['t']))
positions = np.zeros_like(scattering)
scattering[0:len(Ie)] = Ie
positions[0:len(epos)] = epos
# Laplace transformed modulus
if Laplace:
[omega_L, G1_L, G2_L] = analysis_tools.shear_modulus_laplace_transform(t,
dlsmicro_df['msd_smooth'], r, T)
dlsmicro_df['G1'], dlsmicro_df['G2'] = utils.laplace_merge(dlsmicro_df['t'],
dlsmicro_df['G1'],
dlsmicro_df['G2'],
G1_L,G2_L)
# Construct pandas dataframe for this replicate and append it to the
# master dataframe
dlsmicro_df['replicate'] = [replicate]*len(dlsmicro_df['t'])
dlsmicro_df['scattering'] = scattering
dlsmicro_df['epos'] = positions
dlsmicro_df['condition'] = condition
dlsmicro_df['id'] = idx
idx += 1
df = pd.concat((df,dlsmicro_df), axis=0, sort=True)
###############################################
# Plot the analyzed data for this replicate
###############################################
if plot_corr:
plt.plot(t, g, '-r')
plt.xscale('log')
plt.xlabel('$\\mathregular{time\ (\\mu s)}$')
plt.ylabel('g')
plt.tight_layout()
if save_plots:
save_path = '%s/%s/replicate%s/corr' % (root_folder,
condition_dir[condition],
replicate)
plt.savefig(save_path)
plt.show()
if plot_msd:
plt.plot(dlsmicro_df['t'], dlsmicro_df['msd_smooth'], '-r')
plt.xscale('log')
plt.yscale('log')
plt.xlabel('$\\mathregular{time\ (\\mu s)}$')
plt.ylabel('MSD')
plt.tight_layout()
if save_plots:
save_path = '%s/%s/replicate%s/msd' % (root_folder,
condition_dir[condition],
replicate)
plt.savefig(save_path)
plt.show()
if plot_G:
plt.plot(dlsmicro_df['omega'], dlsmicro_df['G1'], '-r')
plt.plot(dlsmicro_df['omega'], dlsmicro_df['G2'], '--r')
plt.ylabel('$\\mathregular{G^*\ (Pa)}$')
plt.xlabel('$\\mathregular{\\omega\ (s^{-1})}$')
plt.legend(['$\\mathregular{G^{\\prime}}$',
'$\\mathregular{G^{\\prime \\prime}}$'], frameon=False,)
plt.xscale('log')
plt.yscale('log')
plt.tight_layout()
if save_plots:
save_path = '%s/%s/replicate%s/G' % (root_folder,
condition_dir[condition],
replicate)
plt.savefig(save_path)
plt.show()
##############################################
# Save analysis results
##############################################
if save_as_text:
save_path = '%s/%s/replicate%s/' % (root_folder,
condition_dir[condition],
replicate)
for i in dlsmicro_df.columns[:-2]:
np.savetxt('%s/%s' % (save_path,i), dlsmicro_df[i].values)
#################################################
# Save the pandas dataframe
#################################################
if save_as_df:
save_path = df_save_path + '/' + df_file_name
df.to_pickle(save_path)