Mifobio 2023 - Thorsten WOHLAND - Computational Advances in Imaging Fluorescence Correlation Spec... скачать в хорошем качестве

Mifobio 2023 - Thorsten WOHLAND - Computational Advances in Imaging Fluorescence Correlation Spec...

Thorsten WOHLAND - Computational Advances in Imaging Fluorescence Correlation Spectroscopy. The characterization of a biological system requires the determination of its structure and dynamics. Microscopy techniques to determine structure in complex systems even below the diffraction limit are widely accessible. But the measurement of dynamics is more difficult. One technique that has emerged over the last years is Imaging Fluorescence Correlation Spectroscopy (Imaging FCS) which provides 2D and 3D biomolecular dynamics and interactions in live systems. However, it faces several challenges that need to be addressed to make it readily usable. First, the large amount of data recorded by array detectors requires efficient approaches for data analysis. Second, new illumination geometries require solutions for data fitting as often analytic solutions cannot be found for the theoretical correlation functions that are required for data fitting. Third, measurement times are on the order of minutes, precluding the reliable detection of changes that occur below that limit.Here, we address these issues by employing GPUs to accelerate real-time evaluations and deep learning to improve Imaging FCS time resolution and data fitting. GPUs accelerate data evaluation between 10-40 times depending on the evaluation modality if more than ~100 pixels are evaluated. Below that limit of 100 pixels, the overhead of transferring data to the GPUs makes them less efficient than the inbuilt CPUs. In deep learning we developed two convolutional neural networks (CNNs), FCSNet and ImFFSNet, which use either correlation functions or raw intensity traces as input, respectively. Both networks are trained on simulated, synthetic data, as it is difficult to obtain ground truth data for FCS over wide parameter ranges. These approaches allow us to a) use arbitrary experimental geometries, as analytic fit models are no longer required, b) reduce the amount of data required by about one order of magnitude to achieve the same precision as conventional nonlinear least-squares data fitting, and c) reduce the evaluation time by about two orders of magnitude. Finally, we will present several applications of these computational approaches in cells and on data recorded on total internal reflection and light sheet microscopes. Thorsten Wohland - Department of Biological Sciences and NUS Centre for Bio-Imaging Sciences, National University of Singapore https://www.dbs.nus.edu.sg/staffs/tho...