Autocalibration in the Quantitative Three-Image FRET method enables simplified and quantitative live-cell biosensor imaging

Genetically-encoded FRET biosensors enable real-time measurement of biochemical activity in living cells with the spatio-temporal resolution given by optical microscopy. However, their widespread use is hindered by challenges in obtaining quantitative, instrument- and expression-level-independent results. To address this, we recently developed Quantitative Three-Image FRET (QuanTI-FRET), which provides absolute FRET probabilities but requires extra calibration constructs and calibration experiments. Here, we leverage the constant stoichiometry of intramolecular FRET biosensors to calibrate the system directly from the dataset of interest. We validated this idea by comparing the standard and autocalibration results obtained on live-cell images of the FAK biosensor, demonstrating robust performance even in low-signal or noisy conditions.  By simplifying the calibration process, the QuanTI-FRET method facilitates quantitative comparisons across conditions or laboratories and enables high-throughput protocols. We provide open-source Python software and a Napari plug-in to make advanced FRET analysis widely available.