Implementation of Neural Networks for the Identification of Single Molecules

Abstract

The effectiveness of neural networks and the optimization of parameters for implementing neural networks were evaluated for use in the identification of single molecules according to their fluorescence lifetime. The best network architecture and training parameters were determined for both ideal and nonideal single-molecule fluorescence data. The effectiveness of the neural network is compared to that of the maximum likelihood estimator on the basis of its ability to correctly identify single molecules. For ideal single-molecule data, it was found that the neural networks and the maximum likelihood estimator perform approximately equally well. For nonideal single-molecule fluorescence data, neural networks were able to correctly identify a larger percentage of single-molecule events than the MLE method.

Single-Molecule Detection and Identification of Multiple Species by Multiparameter Fluorescence Detection

“…A neural network was constructed and trained to identify molecules based on fluorescence lifetime data and was shown to perform better than (nonideal more complex data) or equal to (ideal data) a maximum likelihood estimator (MLE) approach. 23 Here "ideal" means, that the data could be well described by the MLE by assuming, for example, a singleexponential decay. By use of identification procedures of this kind, applied in the multiparameter space, it is reasonable to believe that the number of species that can be separated from each other, and the specificity with which this can be done, can be increased even further.…”

“…11,17 In recent years, the characterization of molecules in singlemolecule fluorescence detection measurements has step by step been established for the different fluorescence parameters. Spectral properties of absorption and fluorescence, F(λ A , λ F ), 18 fluorescence brightness and quantum yield, Φ F , 19 fluorescence lifetime, τ, [20][21][22][23][24][25][26] and anisotropy, r, 27,28 are the five intrinsic properties of a fluorophore that are accessible in a multiparameter fluorescence detection (MFD) experiment (see Figure 1). These "chromophore parameters" can be deduced from the time-resolved detection of the five observables of the chromophore, which now serves as a tool to report on its local environment.…”

Some Recent Advances in Fluorescence Spectroscopy

“…Bowen et al (2004) examined their effectiveness in the identification of fluorescence events from single molecules Pulsed-source time-resolved fluorescence lifetime methods were enhanced by Dolenko et al (2002b), who showed that ANNs could help to resolve noninteracting twocomponent dye mixtures in cases where the lifetimes were shorter than both the exciting light pulse and the detector gate time. Overall it is clear that ANN methods have a great deal to offer to the variety of data-intensive fluorescence techniques now available to researchers, and this area will surely increase in importance, especially as ANN software is now readily available.…”

Analyzing complex single-molecule emission patterns with deep learning

“…Deep neural networks (DNN) extract features from the input and learns its connection to the output automatically 14 . A neural network in its basic form (1-2 fully-connected layers) has been utilized to identify fluorophore species in single-molecule fluorescence lifetime experiments 15 as well as to speed up dipole orientation estimation from an analytical approximation of the dipole PSF 16 . Through the deep architecture of DNN, the complex mapping between input and output is extracted from different levels of features hierarchically 14 .…”

“…Single-molecule emission patterns can be designed to evolve and encode molecular properties, such as, three-dimensional positions 2 , probe spectra 19 , identities 15 and orientations 3 . However, encoding two or more classes of information in the emission patterns will increase their dimensionality, which challenges the traditional decoding processes such as feature recognition and regression.…”