Simulation of fluorophore brightness

class fret_tester.PolyLnBrightness(parms, max_mean=None, precision=0.0)

LognormalBrightness subclass with polynomial mean-vs.-std relation

This is a subclass of LognormalBrightness where the relation between mean and standard deviation is described by a polynomial.

Parameters:

• parms (list of float) – Polynomial coefficients in decreasing order
• max_mean (float or None) – If given (and precision > 0), precalculate parameters of the lognormal distributions (corresponding to means and matchings stds) from 0 to max_mean in steps of precision for a performance gain (at cost of accuracy). Defaults to None, i.e. no pre-computation.
• precision (float or None) – Precision of pre-computed lognormal parameters. The smaller, the more accurate, as long as it is > 0. Defaults to 0, i.e. no pre-computation.

std_from_mean(m)

Get std. deviation for given mean according to polynomial relation

Parameters:m (array-like) – Mean values to calculate standard deviations for. Returns:Standard deviations corresponding to means. Return type:numpy.ndarray

class fret_tester.LognormalBrightness(max_mean=None, precision=0.0)

Fluorophore brightness modeled by a lognormal distribution

This is intended as a base class. One needs to implement the std_from_mean() in a subclass which gives the lognormal distribution standard deviation for a certain mean.

Parameters:

• max_mean (float or None) – If given (and precision > 0), precalculate parameters of the lognormal distributions (corresponding to means and matchings stds) from 0 to max_mean in steps of precision for a performance gain (at cost of accuracy). Defaults to None, i.e. no pre-computation.
• precision (float or None) – Precision of pre-computed lognormal parameters. The smaller, the more accurate, as long as it is > 0. Defaults to 0, i.e. no pre-computation.

parameters(m)

Calculate lognormal parameters from desired mean

Compute \(\mu\) and \(\sigma\) of the Gaussian distribution that the lognormal distribution is based on such that the lognormal distribution will have mean m and standard deviation std_from_mean(m) (see __init__()).

Parameters:m (array_like) – Mean value for which the lognormal parameters are desired Returns:

• mu (numpy.ndarray) – \(\mu\) of the underlying Gaussian
• sigma (numpy.ndarray) – \(\sigma\) of the underlying Gaussian

generate(m)

Draw random value from the brightness distribution with given mean

Parameters:m (array_like) – Mean values. For each entry, draw a random number from the brightness distribution whose mean value is this entry. Returns:Random brightness values Return type:numpy.ndarray

__call__(m)

Synonymous for generate()

std_from_mean(m)

Get standard deviation for given mean

This needs to be implemented in a subclass.

Parameters:m (array-like) – Mean values to calculate standard deviations for. Returns:Standard deviations corresponding to means. Return type:numpy.ndarray