Resolution of heterogeneous fluorescence by the combined use of indirect excitation decay-associated spectral analysis and principal factor analysis

Abstract
Heterogeneous fluorescence of a cholesterol analogue probe is resolved by employing both model-dependent and model-independent curve resolution methods to arrive at pure-component excitation spectra. The model-dependent approach involved performing an indirect excitation decay-associated spectral (IEDAS) analysis which, because of the triple-exponential fluorescence decay kinetics of the cholesterol analogue probe, yielded three component spectra. The model-independent approach involved the use of principal factor analysis (PFA), coupled with either self-modelling (SM) or entropy minimization (EM) routines, and yielded two component spectra. The discrepancy in the number of components obtained by the two different data analysis methods can be taken as an indication that the model incorporated in the IEDAS analysis is not appropriate to describe the photophysics of the cholesterol analogue probe. A modified IEDAS analysis, in which the underlying model allowed for an irreversible excited-state reaction, arrived at two component spectra which were in excellent agreement with those derived by the model-independent PFA/EM analysis. The limitations of self-modelling analysis, as a method to arrive at pure-component spectra from the eigenvectors generated by PFA, were very evident in this case. This work demonstrates the value of a multifaceted approach to the problem of resolving pure-component spectra from heterogeneous fluorescence spectra.