PYME.Analysis.BleachProfile.rollins module

Additional kinetic fitting stuff from Ben Rollins.

TODO - move out of core to plugin??? Esp. as technical accuracy of some fits debatable and duplicates more accurate functions in photophysics / kinModels.

The following need to be resolved to ensure it’s useful in the core:

• Fitting a simple exponential to a histogram of lifetimes behaves poorly for short lifetimes (< ~10 frames). The correct

model function needs to take bin width into account. - Using unweighted least squares to fit an exponential will lead to undue influence of the very short time bins, and generally poor performance.

PYME.Analysis.BleachProfile.rollins.FitModel_N(res_fxn, startParameters, x_data, y_data)

PYME.Analysis.BleachProfile.rollins.FitModel_NB(res_fxn, startParameters, x_data, y_data, prob)

PYME.Analysis.BleachProfile.rollins.defm(x, a, b, c, d)

PYME.Analysis.BleachProfile.rollins.g_histfitting(colourFilter, metadata, cluster_ids, blink_ids, N_bins)

PYME.Analysis.BleachProfile.rollins.gen_nb_testdata(x, N, p, noise=0, n_outliers=1, random_state=0)

PYME.Analysis.BleachProfile.rollins.gfm(x, a)

PYME.Analysis.BleachProfile.rollins.grff(t, x, y)

PYME.Analysis.BleachProfile.rollins.histfitting(colourFilter, metadata, cluster_idxs, fit_order, num_bins, blink_on_label, blink_off_label, to_json=False, log_bins=False, n_on_bins=1, n_off_bins=1, fixed_on_max=-1, fixed_off_max=-1)

PYME.Analysis.BleachProfile.rollins.nbmf(x, N, p)

PYME.Analysis.BleachProfile.rollins.nbrff(t, x, y, p)

PYME.Analysis.BleachProfile.rollins.neg_binom_histfitting(colourFilter, metadata, cluster_ids, blink_ids, N_bins, P)

PYME.Analysis.BleachProfile.rollins.se_diff_l2_mean(p0, x, y)

PYME.Analysis.BleachProfile.rollins.sefm(x, a, b)

PYME.Analysis.BleachProfile.rollins.tfoef(t, x, y)

PYME.Analysis.BleachProfile.rollins.tsoef(t, x, y)