Python module with C-level implementation of point-process correlations functions. These functions were primarily designed for computation of correlation functions for Fluorescence Correlation Spectroscopy (FCS)1 and derivative methods.
Correlations can be computed easily
import numpy as np
import pyFCS as fcs
# Further importats and data processing
# should produce timesT and timesU as
# monotonicaly increasing non-gegative integer arrays
corrl = fcs.correlate(times, times, bins)Beyond standard auto and cross-correlation of FCS data, pyFCS also implements calculation of correlation with weights for:
- Fluorescence Lifetime Correlation Spectroscopy (FLCS)2
- purified FCS (pFCS)3
- filtered FCS (fFCS)4 (the combination of FLCS and pFCS)
All of these will have the same basic form:
import numpy as np
import pyFCS as fcs
# Code gets data for photons and nanotimes,
# Generates the appropriate filter functions
# And provides the times in times
# (in FLCS usually T/U arrays are the same)
# and weights
corrl = fcs.correlate(times, times, bins, weightsT=weights, weightsU=weights)Most of the code is implemented in C for maximal efficiency. The algorithms are all adapted from1. If any issues are encountered that cannot be solved, especially segfaults/kernel crashes, please open an issue in github, providing the minimal code necessary to reproduce the crash.
Footnotes
-
Laurence et. al. (2006) Fast, flexible algorithm for calculating photon correlations. Optics Letters , 31 (6), 829–831 ↩ ↩2
-
Bohmer et. al. (2002) Time-resolved fluorescence correlation spectroscopy. Chem. Phys. Let. 353 (5-6) 439-445 ↩
-
Laurence et. al.(2007) Correlation Spectroscopy of Minor Fluorescent Species: Purification and Distribution Analysis. Biophysical Journal. 92, (6), 2184-2198 ↩
-
Felekyan et. al. (2012) Filtered FCS: Species Auto- and Cross-Correlation Functions Highlight Binding and Dynamics in Biomolecules. ChemPhysChem, 13 (4) 1036-1053 ↩