Highly-efficient quantitative fluorescence resonance energy transfer measurements based on deep learning

• Main Authors: Lin Ge, Fei Liu, Jianwen Luo
• Format: Article
• Language: English
• Published: World Scientific Publishing 2020-11-01
• Series: Journal of Innovative Optical Health Sciences
• Subjects: resonance energy transfer; fluorescence; living cells; photoactivatable; deep network
• Online Access: http://www.worldscientific.com/doi/epdf/10.1142/S1793545820500212

Intensity-based quantitative fluorescence resonance energy transfer (FRET) is a technique to measure the distance of molecules in scale of a few nanometers which is far beyond optical diffraction limit. This widely used technique needs complicated experimental process and manual image analyses to obtain precise results, which take a long time and restrict the application of quantitative FRET especially in living cells. In this paper, a simplified and automatic quantitative FRET (saqFRET) method with high efficiency is presented. In saqFRET, photoactivatable acceptor PA-mCherry and optimized excitation wavelength of donor enhanced green fluorescent protein (EGFP) are used to simplify FRET crosstalk elimination. Traditional manual image analyses are time consuming when the dataset is large. The proposed automatic image analyses based on deep learning can analyze 100 samples within 30s and demonstrate the same precision as manual image analyses.