FRETBursts - ns-ALEX example

This notebook is part of a tutorial series for the FRETBursts burst analysis software.

For a step-by-step introduction to FRETBursts usage please refer to us-ALEX smFRET burst analysis.

In this notebook we present a typical FRETBursts workflow for ns-ALEX smFRET burst analysis.

While FRETBursts does not specifically includes functions for fitting TCSPC fluorescence decays, a fitting with exponential decays and IRF deconvolution can be easily performed using standard python libraries. For an example and a brief discussion see the notebook Lifetime decay fit.

Loading FRETBursts



In [ ]:

    
from fretbursts import *



In [ ]:

    
sns = init_notebook()

Downloading the sample data file

The full list of smFRET measurements used in the FRETBursts tutorials can be found on Figshare.

Here we download the ns-ALEX data-file and put it in a folder named data, inside the notebook folder. For this purpose we use the download_file function provided by FRETBursts:



In [ ]:

    
url = 'http://files.figshare.com/2182602/dsdna_d7_d17_50_50_1.hdf5'



In [ ]:

    
download_file(url, save_dir='./data')



In [ ]:

    
filename = './data/dsdna_d7_d17_50_50_1.hdf5'
filename

Selecting a data file

Alternatively you can use an open-file dialog in order to select a data file:



In [ ]:

    
# filename = OpenFileDialog()
# filename



In [ ]:

    
import os
if os.path.isfile(filename):
    print("Perfect, I found the file!")
else:
    print("Sorry, I can't find the file:\n%s" % filename)

Load the selected file

Here we load the file and we set the alternation parameters:



In [ ]:

    
d = loader.photon_hdf5(filename)
#d = loader.nsalex(fname)



In [ ]:

    
d.time_max



In [ ]:

    
d.det_t



In [ ]:

    
print("Detector    Counts")
print("--------   --------")
for det, count in zip(*np.unique(d.det_t, return_counts=True)):
    print("%8d   %8d" % (det, count))



In [ ]:

    
#d.add(A_ON=(200, 1500), D_ON=(1750, 3200), det_donor_accept=(4, 6))



In [ ]:

    
d.nanotimes_t



In [ ]:

    
bpl.plot_alternation_hist(d)

Execute the previous 2 cells until you get a satisfying selection for the excitation periods. Then run the following to apply the parameters:



In [ ]:

    
loader.alex_apply_period(d)

Burst search and selection



In [ ]:

    
d.calc_bg(fun=bg.exp_fit, time_s=30, tail_min_us='auto', F_bg=1.7)



In [ ]:

    
dplot(d, timetrace_bg)



In [ ]:

    
dplot(d, timetrace)
xlim(1, 2)
ylim(-50, 50)



In [ ]:

    
d.burst_search()



In [ ]:

    
ds = d.select_bursts(select_bursts.size, th1=30)



In [ ]:

    
alex_jointplot(ds)



In [ ]:

    
ds.leakage = 0.05



In [ ]:

    
alex_jointplot(ds)



In [ ]:

    
dplot(ds, hist_fret, show_kde=True)

Nanotimes

The nanotimes for the measurement is saved in the .nanotimes attribute of the Data() object (here either d or ds).

As an example here we get the array of nanotimes for all photons, donor emission and acceptor emission:



In [ ]:

    
d.nanotimes



In [ ]:

    
nanotimes = d.nanotimes[0]
nanotimes_d = nanotimes[d.get_D_em()]
nanotimes_a = nanotimes[d.get_A_em()]

We can plot the histogram for this 3 nanotimes:



In [ ]:

    
hist_params = dict(bins=range(4096), histtype='step', alpha=0.6, lw=1.5)
hist(nanotimes, color='k', label='Total ph.', **hist_params)
hist(nanotimes_d, color='g', label='D. em. ph.', **hist_params)
hist(nanotimes_a, color='r', label='A. em. ph.', **hist_params)
plt.legend()
plt.yscale('log')

We can also select only nanotimes of photons inside bursts. Here, as an example, we will use the ds variable that contains a selection of bursts.

First we compute a selection mask (a boolean array) for photons inside bursts:



In [ ]:

    
ph_in_bursts_mask = d.ph_in_bursts_mask_ich()

Then we apply this selection to the nanotimes array. To get the donor- and acceptor-emission nanotimes we combine the in-bursts selection mask (ph_in_bursts_mask) with the donor or acceptor emission mask (that we get with .get_D_em() and .get_D_em()):



In [ ]:

    
bursts_nanotimes_t = nanotimes[ph_in_bursts_mask]
bursts_nanotimes_d = nanotimes[ph_in_bursts_mask * d.get_D_em()]
bursts_nanotimes_a = nanotimes[ph_in_bursts_mask * d.get_A_em()]

And, as before, we can histogram the nanotimes:



In [ ]:

    
hist_params = dict(bins=range(4096), histtype='step', alpha=0.6, lw=1.5)
hist(bursts_nanotimes_t, color='k', label='Total ph.', **hist_params)
hist(bursts_nanotimes_d, color='g', label='D. em. ph.', **hist_params)
hist(bursts_nanotimes_a, color='r', label='A. em. ph.', **hist_params)
plt.legend()
plt.yscale('log')

Saving to a file

Saving some of all these arrays to file is straightforward.

Save array to txt comma-separed-values

To save a single array to a file we can use the .tofile method:



In [ ]:

    
nanotimes.tofile('nanotimes_t.csv', sep=',\n')  # save in CSV txt format

Save to legacy MATLAB format

To save a set of arrays in MATLAB format we can use the scipy.io.savemat function.

Here we save 3 arrays bursts_nanotimes_t, bursts_nanotimes_d and bursts_nanotimes_a to a file called bursts_nanotimes.mat:



In [ ]:

    
from scipy.io import savemat



In [ ]:

    
savemat('bursts_nanotimes.mat', 
        dict(bn_t=bursts_nanotimes_t, 
             bn_d=bursts_nanotimes_d, 
             bn_a=bursts_nanotimes_a,))

When loaded in MATLAB the arrays will be named bn_t, bn_d and bn_a.



In [ ]:



In [ ]:



In [ ]:



In [ ]: