Tutorial about analyzing localization properties#
from pathlib import Path
%matplotlib inline
import numpy as np
import matplotlib.pyplot as plt
import locan as lc
lc.show_versions(system=False, dependencies=False, verbose=False)
Locan:
version: 0.20.0.dev41+g755b969
Python:
version: 3.11.6
Load rapidSTORM data file#
Identify some data in the test_data directory and provide a path using pathlib.Path (returned by lc.ROOT_DIR)
path = lc.ROOT_DIR / 'tests/test_data/rapidSTORM_dstorm_data.txt'
print(path, '\n')
dat = lc.load_rapidSTORM_file(path=path, nrows=1000)
/home/docs/checkouts/readthedocs.org/user_builds/locan/envs/latest/lib/python3.11/site-packages/locan/tests/test_data/rapidSTORM_dstorm_data.txt
Jupyter environment detected. Enabling Open3D WebVisualizer.
[Open3D INFO] WebRTC GUI backend enabled.
[Open3D INFO] WebRTCWindowSystem: HTTP handshake server disabled.
Print information about the data:
print(dat.data.head(), '\n')
print('Summary:')
dat.print_summary()
print('Properties:')
print(dat.properties)
position_x position_y frame intensity chi_square local_background
0 9657.40 24533.5 0 33290.10 1192250.0 767.732971
1 16754.90 18770.0 0 21275.40 2106810.0 875.460999
2 14457.60 18582.6 0 20748.70 526031.0 703.369995
3 6820.58 16662.8 0 8531.77 3179190.0 852.789001
4 19183.20 22907.2 0 14139.60 448631.0 662.770020
Summary:
identifier: "1"
comment: ""
source: EXPERIMENT
state: RAW
element_count: 14
frame_count: 1
file {
type: RAPIDSTORM
path: "/home/docs/checkouts/readthedocs.org/user_builds/locan/envs/latest/lib/python3.11/site-packages/locan/tests/test_data/rapidSTORM_dstorm_data.txt"
}
creation_time {
2024-03-14T11:07:36.400872Z
}
Properties:
{'localization_count': 14, 'position_x': 15873.847142857145, 'uncertainty_x': 2361.4490857013648, 'position_y': 17403.909285714286, 'uncertainty_y': 1803.9975262697349, 'intensity': 183987.66999999998, 'local_background': 675.0614, 'frame': 0, 'region_measure_bb': 730882123.3259, 'localization_density_bb': 1.915493559521281e-08, 'subregion_measure_bb': 108337.2}
Visualization#
lc.render_2d(dat, bin_size=1000, rescale=(0,100));
Analyze a localization property#
We have a look at a certain localization property in locdata.
The analysis class LocalizationProperty provides a dataframe with the property as function of another property (index), and a plot or histogram of this property.
lprop = lc.LocalizationProperty(loc_property='intensity', index='frame')
lprop.compute(dat)
print(lprop.results.head())
intensity
frame
0 33290.10
0 21275.40
0 20748.70
0 8531.77
0 14139.60
The plot shows results smoothed by a running average according to the specified window.
lprop.plot(window=100);
The histogram shows the probability density function of results.
lprop.hist(fit=False);
Per default the distribution is fitted to an exponential decay.
lprop.hist();
Fit results (as derived using the lmfit library) are provided in the distribution_statistics attribute.
lprop.distribution_statistics.parameter_dict()
{'intensity_loc': 5730.75, 'intensity_scale': 7411.226428571428}
lprop.results.min()
intensity 5730.75
dtype: float64
Fitting different distribution models#
Per default the ‘with_constraints’ flag is True to apply standard fit constraints. This can be set to false and other parameters can be passed to the fit function.
lprop.fit_distributions(with_constraints=False, floc=0)
lprop.distribution_statistics.parameter_dict()
{'intensity_loc': 0.0, 'intensity_scale': 13141.976428571428}
lprop.hist(fit=True)
print(lprop.distribution_statistics.parameter_dict())
{'intensity_loc': 0.0, 'intensity_scale': 13141.976428571428}
Showing correlations between two properties#
By setting the index to another localization property correlations can be shown.
lprop = lc.LocalizationProperty(loc_property='intensity', index='local_background').compute(dat)
lprop.plot(marker='o', linestyle="", alpha=0.1);
Correlation coefficients can be investigated in more detail using the LocalizationPropertyCorrelation class that is just a visualization of pandas.DataFrame.corr().
lpcorr = lc.LocalizationPropertyCorrelations(loc_properties=['intensity', 'local_background']).compute(dat)
lpcorr
LocalizationPropertyCorrelations(loc_properties=['intensity', 'local_background'])
lpcorr.plot();
2-dimensional distribution of localization properties#
In order to investigate a certain localization property in 2D you can just print the image with a color code representing the mean value of the chosen localization property in each bin.
lc.render_2d_mpl(dat, other_property='local_background', bin_size=500);
Otherwise use a specific class to analyse localization properties in 2d. Per default a bimodal normal distribution is fitted. This can e.g. help to check on even illumination during the recording.
lprop2d = lc.LocalizationProperty2d(loc_properties=None, other_property='local_background', bin_size=500).compute(dat)
lprop2d.plot_deviation_from_mean();
lprop2d.plot(colors="r");
lprop2d.report()
Fit results for:
[[Model]]
Model(model_function)
[[Fit Statistics]]
# fitting method = leastsq
# function evals = 43
# data points = 12
# variables = 5
chi-square = 75179.6882
reduced chi-square = 10739.9555
Akaike info crit = 114.912757
Bayesian info crit = 117.337290
R-squared = 0.25569453
[[Variables]]
amplitude: 732.468389 +/- 47.0045236 (6.42%) (init = 875.461)
center_x: 13822.4142 +/- 14800.9733 (107.08%) (init = 18245.59)
center_y: 19401.6868 +/- 3825.17126 (19.72%) (init = 13803.61)
sigma_x: 42169.8059 +/- 52664.9262 (124.89%) (init = 7125)
sigma_y: 15484.6869 +/- 8624.45209 (55.70%) (init = 6250)
[[Correlations]] (unreported correlations are < 0.250)
C(center_y, sigma_y) = +0.7364
C(center_x, sigma_x) = -0.6514
C(amplitude, sigma_y) = -0.6030
C(amplitude, sigma_x) = -0.5552
C(amplitude, center_y) = -0.4075
C(center_y, sigma_x) = +0.3845
C(amplitude, center_x) = +0.3313
C(sigma_x, sigma_y) = +0.3140
C(center_x, sigma_y) = -0.2882
C(center_x, center_y) = -0.2544
Maximum fit value in image: 731.334
Minimum fit value in image: 580.351
Fit value variation over image range: 0.21