Super-resolution FRET image reconstruction method based on donor-acceptor intermolecular dipole resonance constraint

Abstract

The invention discloses a super-resolution FRET (Fluorescence Resonance Energy Transfer) image reconstruction method based on donor-acceptor intermolecular dipole resonance constraint. The method comprises the following steps: firstly, acquiring an FRET three-channel original image data set through a structured light illumination super-resolution microscopic imaging system, and carrying out image preprocessing; then, correlation constraint is carried out on the super-resolution SIM image based on donor-acceptor molecule dipole resonance prior information, an SI-FRETM unified forward model and an objective function are constructed, forward iteration solution is carried out, and reconstructed FRET three-channel fluorescence distribution is obtained; and finally, super-resolution FRET efficiency distribution and donor concentration ratio distribution are obtained through pixel-by-pixel calculation. According to the method, correlation constraint is carried out on super-resolution SIM imaging by using donor-acceptor molecule dipole resonance prior information, spectrum expansion is realized through forward iteration solution so as to carry out FRET reconstruction, and the fidelity of super-resolution quantitative FRET imaging and the accuracy of quantitative analysis are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of super-resolution fluorescence resonance energy transfer (FRET) imaging, and in particular relates to a super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules. Background Art

[0002] Fluorescence resonance energy transfer (FRET) is an important technical means to detect the structure and dynamic interactions of biomacromolecules in living cells. FRET can quantitatively and dynamically analyze the interactions and conformational changes of biomacromolecules at the scale of 1-10nm in real time in living cells. However, due to the limitation of spatial resolution by the optical diffraction limit, the spatial resolution of FRET imaging based on wide-field microscopy or confocal microscopy is greater than 200nm, which makes it difficult to reveal the structural information of biomacromolecules in subcellular structures. By integrating super-resolution structured illumination microscopy (SR-SIM) with FRET imaging technology, super-resolution FRET microscopy (SIM-FRET) based on structured light illumination is developed, the spatial resolution capability of FRET technology is improved, and FERT detection in subcellular localization is more accurate, which provides a new method for analyzing the dynamic process of biomacromolecule interactions in the subcellular organelle structure of living cells.

[0003] At present, super-resolution SIM-FRET imaging adopts a two-step image reconstruction strategy of SIM reconstruction followed by quantitative FRET analysis, and its fidelity is restricted by SIM reconstruction artifacts. The linear Wiener SIM reconstruction method can maintain the grayscale fidelity of the reconstructed image, but it will produce significant amplified noise artifacts, and the artifacts will be further accumulated in the FRET algebraic operation, thereby affecting the quantitative FRET analysis. Patent application number 2022103326816 proposes a method of using a co-localization mask to filter out random noise artifacts after SIM reconstruction and then quantitative FRET analysis. This method uses the spatial co-localization characteristics of FRET to generate a binary mask, and uses the mask to filter out false positive FRET signals caused by noise artifacts; however, the mask generation of this method depends on artificial threshold selection, and cannot completely eliminate the noise artifacts superimposed on the FRET signal. Another Chinese patent (application number: 202310581280.9) proposes a method for super-resolution SIM-FRET reconstruction after denoising the original image data through a self-supervised learning denoising neural network. This method performs denoising preprocessing on the original fluorescence image through a self-supervised learning denoising neural network.

[0004] The above-mentioned prior arts are all based on the super-resolution SIM-FRET two-step reconstruction strategy, but have the following disadvantages: First, since the first step of SIM image reconstruction lacks the constraint of FRET prior information, SIM reconstruction and FRET analysis are two independent processes, resulting in the infidelity of super-resolution FRET signals. Second, the accumulation of artifacts caused by the step-by-step reconstruction steps is difficult to completely eliminate, affecting the accuracy of quantitative FRET analysis. Summary of the invention

[0005] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide a super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules, utilize the prior information of the dipole resonance of donor-acceptor molecules to perform correlation constraints on super-resolution SIM imaging, and then construct a SI-FRETM unified forward model and reconstruction objective function, and realize the reconstruction of the super-resolution FRET sample fluorescence signal through forward iterative solution, and obtain the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution, thereby improving the fidelity of super-resolution quantitative FRET imaging and the accuracy of quantitative analysis.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The first object is to provide a super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules, comprising the following steps:

[0008] Acquire a FRET three-channel raw image data set of a living cell FRET sample by a structured light illumination super-resolution microscopy imaging system; the FRET three-channel includes a DD channel for donor excitation and donor emission, a DA channel for donor excitation and acceptor emission, and an AA channel for acceptor excitation and acceptor emission;

[0009] Performing image preprocessing on the FRET three-channel original image data set to obtain a preprocessed FRET three-channel image data set;

[0010] Based on the prior information of the dipole resonance of donor-acceptor molecules, the super-resolution SIM imaging is associated with constraints, and a SI-FRETM unified forward model is constructed; the SI-FRETM unified forward model includes FRET sample fluorescence distribution, FRET three-channel spectral mixing parameters, FRET three-channel cosine structured light field and FRET three-channel point spread function;

[0011] Taking the FRET sample fluorescence distribution as the reconstruction target, establishing the objective function according to the SI-FRETM unified forward model and performing forward iteration to obtain the reconstructed FRET sample fluorescence distribution; the FRET sample fluorescence distribution includes the total super-resolution donor fluorescence distribution, the total super-resolution acceptor fluorescence distribution and the super-resolution sensitized FRET fluorescence distribution;

[0012] Based on the reconstructed super-resolution donor fluorescence total distribution, super-resolution acceptor fluorescence total distribution and super-resolution sensitized FRET fluorescence distribution, the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution are obtained by pixel-by-pixel calculation.

[0013] As a preferred technical solution, the image preprocessing includes:

[0014] Taking the original image of the AA channel as a reference, the original images of the DD channel and the DA channel are aligned pixel by pixel by using an affine transformation to obtain a FRET three-channel registration image data set; the affine transformation is a linear transformation operation, including rotation, translation, scaling and reflection;

[0015] The grayscale value statistics of the FRET three-channel registered image data set are performed pixel by pixel, and the grayscale value corresponding to the first peak in the grayscale value histogram is used as the background grayscale value to perform pixel by pixel background correction on the FRET three-channel registered image data set to obtain the preprocessed FRET three-channel image data set.

[0016] As a preferred technical solution, the construction of the SI-FRETM unified forward model is specifically as follows:

[0017] The FRET three-channel spectral mixing parameter matrix is ​​obtained by measuring the FRET reference sample; the FRET three-channel spectral mixing parameter matrix is ​​used to characterize the mixing degree of spectral overlap and crosstalk between the three FRET channels caused by the physical process of the donor-acceptor molecule dipole resonance, specifically:

[0018]

[0019] Where m ij , i,j∈[1,3] represents the contribution of the donor fluorophore, acceptor fluorophore or sensitized FRET fluorophore in the FRET sample fluorescence distribution X to the fluorescence signal collected by the FRET three channels; the FRET reference sample usually includes a single-transfection donor sample, a single-transfection acceptor sample and a standard plasmid sample with a donor-acceptor ratio of 1:1;

[0020] The structured light field parameter estimation method is used to estimate the preprocessed FRET three-channel image data set to obtain the FRET three-channel cosine structured light field, which is expressed as:

[0021]

[0022] in, It is the FRET three-channel cosine structured light field. is the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel, r is the spatial coordinate of the fluorescence image pixel of the corresponding channel, θ and n represent the different direction angles and different phases of the structured light field, respectively. is the wave vector of the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel, is the modulation depth of the DD channel cosine structured light field, the DA channel cosine structured light field or the AA channel cosine structured light field, is the initial phase of the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel;

[0023] The FRET three-channel point spread function is obtained by actual measurement using a microscopic imaging system or simulation based on a theoretical model, and is expressed as:

[0024]

[0025] Among them, H FRET is the FRET three-channel point spread function, H DD (r) is the point spread function of the DD channel, H DA (r) is the point spread function of the DA channel, H AA (r) is the point spread function of the AA channel, and r is the spatial coordinate of the fluorescence image pixel of the corresponding channel;

[0026] According to the forward physical process of structured light excitation of FRET samples and acquisition of FRET three-channel images, the SI-FRETM unified forward model is constructed by combining the prior information of donor-acceptor molecular dipole resonance to perform correlation constraints on super-resolution SIM imaging.

[0027] The SI-FRETM unified forward model includes the FRET sample fluorescence distribution, the FRET three-channel spectral mixing parameters, the FRET three-channel cosine structured light field and the FRET three-channel point spread function, which is expressed as:

[0028]

[0029] in, is the unified forward model of SI-FRETM, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, M is the FRET three-channel spectral mixing parameter matrix, X = [X 1 ,X 2 ,X 3 ] T is the FRET sample fluorescence distribution, X 1 is the total distribution of super-resolution donor fluorescence, X 2 is the total distribution of super-resolution receptor fluorescence, X 3 It is the super-resolution sensitized FRET fluorescence distribution.

[0030] As a preferred technical solution, taking the FRET sample fluorescence distribution as the reconstruction target, the reconstruction objective function of the SI-FRETM unified forward model is established, which is expressed as:

[0031]

[0032] Where X = [X 1 ,X 2 ,X 3 ] T Reconstructed FRET sample fluorescence distribution for the target, X 1 is the total distribution of super-resolution donor fluorescence, X 2 is the total distribution of super-resolution receptor fluorescence, X 3 For super-resolution sensitized FRET fluorescence distribution, Unified forward model for SI-FRETM; is the data likelihood term, and the likelihood term is characterized in the form of least squares:

[0033]

[0034] Among them, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, respectively, and M is the FRET three-channel spectral mixing parameter;

[0035] According to the functional properties of the data likelihood term, a suitable optimization algorithm is designed to perform forward iterative solution on the reconstruction objective function of the SI-FRETM unified forward model to obtain the reconstructed FRET sample fluorescence distribution.

[0036] As a preferred technical solution, the regularization term constraint of the fluorescence image prior information is added to the reconstruction objective function of the SI-FRETM unified forward model, and the regularized reconstruction objective function of the SI-FRETM unified forward model is established, which is expressed as:

[0037]

[0038] Where X = [X 1 ,X 2 ,X 3 ] T Reconstructed FRET sample fluorescence distribution for the target, X 1 is the total distribution of super-resolution donor fluorescence, X 2 is the total distribution of super-resolution receptor fluorescence, X 3 For super-resolution sensitized FRET fluorescence distribution, Unified forward model for SI-FRETM; is the data likelihood term, R(X) is the regularization term, and λ is the weight factor for weighing the data likelihood term and the regularization term;

[0039] The data likelihood term The likelihood term is characterized in the form of least squares:

[0040]

[0041] Among them, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, respectively, and M is the FRET three-channel spectral mixing parameter;

[0042] The regularization term R(X) is used to add prior information of the fluorescence image;

[0043] According to the functional properties of the regularization term, a suitable optimization algorithm is designed to perform forward iterative solution on the regularized reconstruction objective function of the SI-FRETM unified forward model to obtain the reconstructed FRET sample fluorescence distribution.

[0044] As a preferred technical solution, the regularization term is spatial continuity regularization and / or non-negative regularization.

[0045] As a preferred technical solution, the calculation method of the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution is:

[0046]

[0047] Among them, E D is the apparent efficiency centered on the donor, E A is the apparent efficiency centered on the donor, R C is the acceptor-donor concentration ratio distribution, X 1 is the total distribution of super-resolution donor fluorescence reconstructed, X 2 is the total distribution of super-resolution receptor fluorescence reconstructed, X 3 Reconstructed super-resolution sensitized FRET fluorescence distribution.

[0048] The second purpose is to provide a super-resolution FRET image reconstruction device based on dipole resonance constraint between donor and acceptor molecules, which is applied to the above-mentioned super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules, including an image acquisition module, an image preprocessing module, a model building module, a target solving module and a FRET calculation module;

[0049] The image acquisition module is used to acquire a FRET three-channel raw image data set of a living cell FRET sample imaging through a structured light illumination super-resolution microscopy imaging system; the FRET three channels include a DD channel for donor excitation and donor emission, a DA channel for donor excitation and acceptor emission, and an AA channel for acceptor excitation and acceptor emission;

[0050] The image preprocessing module is used to perform image preprocessing on the FRET three-channel original image data set to obtain a preprocessed FRET three-channel image data set;

[0051] The model building module is used to perform association constraints on super-resolution SIM imaging based on prior information of donor-acceptor molecular dipole resonance, and to construct a SI-FRETM unified forward model; the SI-FRETM unified forward model includes FRET sample fluorescence distribution, FRET three-channel spectral mixing parameters, FRET three-channel cosine structured light field and FRET three-channel point spread function;

[0052] The target solving module is used to take the FRET sample fluorescence distribution as the target, establish a reconstruction target function according to the SI-FRETM unified forward model and perform forward iterative solution to obtain the reconstructed FRET sample fluorescence distribution; the FRET sample fluorescence distribution includes the total super-resolution donor fluorescence distribution, the total super-resolution acceptor fluorescence distribution and the super-resolution sensitized FRET fluorescence distribution;

[0053] The FRET calculation module is used to obtain the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution by pixel-by-pixel calculation according to the reconstructed super-resolution donor fluorescence total distribution, super-resolution acceptor fluorescence total distribution and super-resolution sensitized FRET fluorescence distribution.

[0054] The third object is to provide an electronic device, comprising:

[0055] at least one processor; and a memory in communication with the at least one processor; wherein,

[0056] The memory stores computer program instructions that can be executed by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can perform the above-mentioned super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules.

[0057] The fourth object is to provide a computer-readable storage medium storing a program, which, when executed by a processor, implements the above-mentioned super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules.

[0058] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0059] 1. The present invention introduces the FRET physical process information into the image reconstruction process by establishing a unified forward model of structured light excitation FRET based on dipole resonance constraints between donor and acceptor molecules (SI-FRETM unified forward model), thereby improving the fidelity of super-resolution FRET reconstruction results and the accuracy of quantitative analysis.

[0060] 2. The present invention establishes an objective function based on the SI-FRETM unified forward model, and realizes forward iterative reconstruction of the SI-FRETM unified forward model by optimizing and solving the objective function. It is not necessary to first perform FERT three-channel super-resolution SIM reconstruction, but directly use the structured light excitation FRET three-channel original image group to forward iteratively reconstruct the super-resolution quantitative FRET signal, avoiding the noise amplification of Wiener SIM reconstruction and the noise accumulation caused by FRET algebraic operations, so that noise artifacts can be more effectively suppressed. Super-resolution quantitative FRET analysis is achieved for living cell biological samples under low signal-to-noise ratio and weak light conditions.

[0061] 3. The present invention further provides a SI-FRETM regularized objective function, which has good regularized scalability and can introduce different fluorescence image prior information to further constrain the solution space. Super-resolution quantitative FRET analysis is achieved by optimizing and solving the SI-FRETM regularized objective function, further enhancing the applicability and scalability of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0063] Figure 1 4 is an overall flow chart of a super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules in an embodiment of the present invention.

[0064] Figure 2 It is a schematic diagram of the results of reconstructing the FRET efficiency and the acceptor-donor concentration ratio for the simulated FRET three-channel original image containing 30% Gaussian white noise in an embodiment of the present invention.

[0065] Figure 3 It is a schematic diagram of the FRET efficiency results reconstructed from the experimental data of MCF7 living cells expressing Acta-G17M in the embodiments of the present invention.

[0066] Figure 4 4 is a structural block diagram of a super-resolution FRET image reconstruction system based on dipole resonance constraint between donor and acceptor molecules in an embodiment of the present invention.

[0067] Figure 5 Schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0069] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments.

[0070] like Figure 1 As shown, the super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules in this embodiment includes the following steps:

[0071] S1. Acquire a FRET three-channel raw image data set of a living cell FRET sample by a structured light illumination super-resolution microscopy imaging system; wherein the FRET three channels include a DD channel of donor excitation donor emission, a DA channel of donor excitation acceptor emission, and an AA channel of acceptor excitation acceptor emission.

[0072] Specifically, the FRET three-channel raw image data set specifically includes the DD channel raw image data set of donor excitation and donor emission. DA channel raw image data set of donor excitation acceptor emission AA channel raw image data set of receptor excitation and receptor emission Where r = (x, y) is the spatial coordinate of the fluorescence image pixel, θ and n represent different direction angles and different phases of the structured light field respectively; in this embodiment, θ = (1, 2, 3) and n = (-1, 0, 1), so each channel contains three different directions, each direction contains three different phases, a total of 9 original images, and the three FRET channels obtain a total of 27 original images of structured light modulation. Figure 3GFP-mCherry fluorescent protein pair was selected as the FRET donor-acceptor pair, 488nm laser was used as donor excitation light, 525±15nm band was used as detection channel for donor fluorescence emission, 561nm laser was used as acceptor excitation light, 600±15nm band was used as detection channel for acceptor fluorescence emission.

[0073] S2. Perform image preprocessing on the FRET three-channel original image data set to obtain a preprocessed FRET three-channel image data set.

[0074] Further, the image preprocessing steps are:

[0075] S201, channel registration: taking the original image of the AA channel as a reference, the original images of the DD channel and the DA channel are aligned pixel by pixel through affine transformation to obtain a FRET three-channel registered image data set; wherein the affine transformation is a linear transformation operation, including linear transformation operations such as rotation, translation, scaling and reflection.

[0076] S202, background subtraction: performing pixel-by-pixel grayscale value statistics on the FRET three-channel registered image data set, using the grayscale value corresponding to the first peak in the grayscale value histogram as the background grayscale value to perform pixel-by-pixel background correction on the FRET three-channel registered image data set, and obtaining a pre-processed FRET three-channel image data set.

[0077] S3. Based on the prior information of the dipole resonance of donor-acceptor molecules, correlation constraints are imposed on super-resolution SIM imaging, and the SI-FRETM unified forward model (structured light excitation FRET unified forward model based on the dipole resonance constraints between donor and acceptor molecules) is constructed, including the FRET sample fluorescence distribution, the FRET three-channel spectral mixing parameters, the FRET three-channel cosine structured light field, and the FRET three-channel point spread function.

[0078] Furthermore, the construction steps of the SI-FRETM unified forward model are:

[0079] S301, measure the FRET three-channel spectral mixing parameter matrix through the FRET reference sample. In this application, the FRET three-channel spectral mixing parameter is used to characterize the mixing degree of spectral overlap and crosstalk between the three FRET channels caused by the physical process of the donor-acceptor molecule dipole resonance, specifically:

[0080]

[0081] Where m ij, i,j∈[1,3] represents the contribution of the donor fluorophore, acceptor fluorophore or sensitized FRET fluorophore in the FRET sample fluorescence distribution X to the fluorescence signal collected by the FRET three-channel. The FRET three-channel spectral mixing parameter matrix M requires additional FRET reference samples to be pre-determined; FRET reference samples include single-transfection donor samples, single-transfection acceptor samples, and standard plasmid samples with a donor-acceptor ratio of 1:1.

[0082] This embodiment provides a method for obtaining FRET three-channel spectral mixing parameters through a FRET reference sample, and the measurement steps are:

[0083] First, the calibration parameters a, b, c, and d of the fluorescence crosstalk between the donor and acceptor emission spectra were measured by single-transfection donor samples and single-transfection acceptor samples; among them, a and b were measured by single-transfection donor samples, and c and d were measured by single-transfection acceptor samples. Then, the FRET sensitization quenching conversion factor G (the ratio of the expression of sensitized acceptor emission to the quenching of donor emission) and the ratio k of the donor fluorescence intensity to the acceptor fluorescence intensity under equimolar concentration conditions without FRET were determined using a standard plasmid reference sample with a donor-acceptor ratio of 1:1; finally, the FRET three-channel spectral mixing parameter matrix was obtained, which is expressed as:

[0084]

[0085] Among them, a, b, c, d are calibration parameters for fluorescence crosstalk between the emission spectra of the donor and the acceptor, G is the FRET sensitization quenching conversion factor, and k is the ratio of the donor fluorescence intensity to the acceptor fluorescence intensity under equimolar concentration conditions in the absence of FRET. It should be noted that the parameters a, b, c, d, G, k, etc. are different when using different FRET donor-acceptor fluorescence pairs.

[0086] S302, using the structured light field parameter estimation method to estimate the FRET three-channel image data set after preprocessing to obtain the FRET three-channel cosine structured light field. Specifically, the structured light field parameter estimation method is used to estimate the preprocessed FRET three-channel image data set to obtain the wave vector, modulation depth and initial phase of the cosine structured light field, and then the FRET three-channel cosine structured light field is obtained, which is expressed as:

[0087]

[0088] The cosine structured light field of each channel is expressed as:

[0089]

[0090] in, It is the FRET three-channel cosine structured light field. is the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel, r is the spatial coordinate of the fluorescence image pixel after the corresponding channel is processed, θ and n represent the different direction angles and different phases of the structured light field, respectively. is the wave vector of the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel, is the modulation depth of the DD channel cosine structured light field, the DA channel cosine structured light field or the AA channel cosine structured light field, is the initial phase of the DD channel cosine structured light field, the DA channel cosine structured light field or the AA channel cosine structured light field. The method of estimating structured light parameters includes but is not limited to the peak phase method (POP), non-iterative autocorrelation reconstruction (ACR), image reconstructing transform (IRT), cross-correlation iteration method (COR), etc. In this example, the cross-correlation iteration method (COR) is used to estimate the structured light field parameters.

[0091] S303, obtaining a FRET three-channel point spread function according to actual measurement of the microscopic imaging system or based on a theoretical model.

[0092] Specifically, obtaining the point spread function includes but is not limited to using a microscopic imaging system to actually measure or generate a FRET three-channel point spread function based on a theoretical model simulation, which is expressed as:

[0093]

[0094] Among them, H FRET is the FRET three-channel point spread function, H DD (r) is the point spread function of the DD channel, H DA (r) is the point spread function of the DA channel, H AA (r) is the point spread function of the AA channel, and r is the spatial coordinate of the fluorescence image pixel after the corresponding channel processing.

[0095] In this embodiment, based on the parameters of the microscopic imaging system, including the numerical aperture of the objective lens, the magnification, the wavelength of the emission channel, etc., the optical transfer function OTF is generated using a theoretical model, and then the OTF is inverse Fourier transformed to obtain the FRET three-channel point spread function PSF. The optical transfer function OTF is:

[0096]

[0097] in, is the optical transfer function OTF, k c is the cutoff frequency of the microscopic imaging system, k is the spectrum coordinate;

[0098] Optical transfer function Perform inverse Fourier transform to obtain the FRET three-channel point spread function H CH (k),H=(DD,DA,AA).

[0099] S304, according to the forward physical process of structured light excitation of FRET sample and acquisition of FRET three-channel image, combined with the prior information of donor-acceptor molecule dipole resonance, the super-resolution SIM imaging is associated with constraints, and the SI-FRETM unified forward model is constructed, including the FRET sample fluorescence distribution, the FRET three-channel spectral mixing parameters, the FRET three-channel cosine structured light field and the FRET three-channel point spread function, which is expressed as:

[0100]

[0101] in, is the unified forward model of SI-FRETM, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, M is the FRET three-channel spectral mixing parameter, X = [X 1 ,X 2 ,X 3 ] T is the fluorescence distribution of the three channels of FRET, X 1 is the total distribution of super-resolution donor fluorescence, X 2 is the total distribution of super-resolution receptor fluorescence, X 3 It is the super-resolution sensitized FRET fluorescence distribution.

[0102] S4. Taking the FRET sample fluorescence distribution as the target, a reconstruction objective function is established according to the SI-FRETM unified forward model and a forward iterative solution is performed to obtain the reconstructed FRET sample fluorescence distribution; wherein the FRET sample fluorescence distribution includes the total super-resolution donor fluorescence distribution, the total super-resolution acceptor fluorescence distribution and the super-resolution sensitized FRET fluorescence distribution.

[0103] Furthermore, taking the FRET sample fluorescence distribution as the reconstruction target, the reconstruction objective function of the SI-FRETM unified forward model is established, which is expressed as:

[0104]

[0105] Where X = [X 1 ,X 2 ,X 3 ] T Reconstructed FRET sample fluorescence distribution for the target, X 1 is the total distribution of super-resolution donor fluorescence, X 2 is the total distribution of super-resolution receptor fluorescence, X3 For super-resolution sensitized FRET fluorescence distribution, Unified forward model for SI-FRETM; is the data likelihood term, which is used to measure the fidelity of the reconstructed image. The likelihood term is characterized in the form of least squares:

[0106]

[0107] Among them, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, respectively, and M is the FRET three-channel spectral mixing parameter;

[0108] According to the functional properties of the data likelihood term, a suitable optimization algorithm is designed to perform forward iterative solution on the reconstruction objective function of the SI-FRETM unified forward model to obtain the reconstructed FRET sample fluorescence distribution.

[0109] For the reconstruction objective function (6) in this embodiment, given the initial value X 0 , let k = k + 1, and generate the search direction d according to certain criteria k and step size α k , so there is an iterative format X k+1 =X k+1 +α k d k , gradually realize the forward iterative solution; further determine whether it converges to the optimal reconstruction result X based on the error k+1 , when the error is less than the threshold ε=1×10 -6 , that is, ||(X k+1 -X k ) / (X k )||<ε, the optimal reconstruction result X of the FRET sample fluorescence distribution is obtained k+1 , stop the iteration.

[0110] Furthermore, the regularization term constraint of the fluorescence image prior information is added to the reconstruction objective function of the SI-FRETM unified forward model, and the regularized reconstruction objective function of the SI-FRETM unified forward model is established, which is expressed as:

[0111]

[0112] Where X = [X 1 ,X 2 ,X 3 ] T Reconstructed FRET sample fluorescence distribution for the target, X 1 is the total distribution of super-resolution donor fluorescence, X 2is the total distribution of super-resolution receptor fluorescence, X 3 For super-resolution sensitized FRET fluorescence distribution, Unified forward model for SI-FRETM; the first is the data likelihood term, the second term R(X) is the regularization term, and λ>0 is the weight factor for weighing the data likelihood term and the regularization term;

[0113] Item 1 is the data likelihood term, which is used to measure the fidelity of the reconstructed image. The likelihood term is also characterized in the form of least squares:

[0114]

[0115] Among them, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, respectively, and M is the FRET three-channel spectral mixing parameter;

[0116] The second term R(X) is a regularization term, which is used to introduce prior information of the FRET three-channel fluorescence image to improve the well-posedness of the solution. Since the structure of the biological sample is continuously changing and the noise is random and discontinuous, and the minimum signal collected during the fluorescence imaging process is zero, based on the above prior, the regularization term in this application adopts spatial continuity regularization and / or non-negative regularization. In this embodiment, the spatial continuity regularization adopts TV regularization constraints, which are expressed as:

[0117]

[0118] in are the horizontal first-order gradient and the vertical first-order gradient of the fluorescence image, respectively, ‖·‖ 1 is the L1 norm;

[0119] The non-negative positive rule introduces the non-negative constraint of the fluorescence intensity of the fluorescence image, which is expressed as:

[0120]

[0121] Finally, according to the functional properties of the regularization term, a suitable optimization algorithm is designed to perform forward iteration to solve the regularized reconstruction objective function of the SI-FRETM unified forward model to obtain the reconstructed FRET sample fluorescence distribution. In this embodiment, the split Bregman method is used to introduce split variables and auxiliary variables into the regularized reconstruction objective function of the SI-FRETM unified forward model, decouple it into several sub-problems and solve them step by step iteratively to obtain the reconstruction result of the FRET sample fluorescence distribution.

[0122] S5. Based on the reconstructed total super-resolution donor fluorescence distribution, total super-resolution acceptor fluorescence distribution and super-resolution sensitized FRET fluorescence distribution, the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution are obtained by pixel-by-pixel calculation:

[0123]

[0124] Among them, E D is the apparent efficiency centered on the donor, E A is the receptor-centered apparent efficiency, R C is the acceptor-donor concentration ratio distribution, X 1 is the reconstruction result of the total distribution of super-resolution donor fluorescence, X 2 is the reconstruction result of the total distribution of super-resolution receptor fluorescence, X 3 This is the reconstruction result of super-resolution sensitized FRET fluorescence distribution.

[0125] In order to further demonstrate the effect of this method, this embodiment uses a simulation model for testing. It is pre-set that a=0.2, b=0, c=0, d=0.8, G=5, k=0.69. This simulation model assumes a fixed FRET efficiency of 0.3 and a radiation star FRET signal with a donor-acceptor concentration ratio of 1:1. The parameters consistent with the actual system are used to simulate the FRET three-channel spectral mixing degree, structured light modulation and point spread function PSF fuzzy simulation to generate a FRET three-channel raw data set. Figure 2 As shown, a schematic diagram showing the results of reconstructing FRET efficiency and acceptor-donor concentration ratio of a simulated FRET three-channel original image containing 30% Gaussian white noise using the method of this embodiment. Figure 2 (a) is the wide-field FRET result. Figure 2 (b) is the linear Wiener SIM-FRET reconstruction result. Figure 2 (c) is the iterative reconstruction result of the unregularized SI-FRETM objective function provided in this embodiment, Figure 2 (d) is the iterative reconstruction result provided by this embodiment containing TV and non-negative regular SI-FRETM objective function. Figure 2 It can be seen that the FRET efficiency histogram and the acceptor-donor concentration ratio histogram of linear Wiener SIM-FRET are widened due to noise artifacts, while the SI-FRETM model reconstruction without regularization can effectively suppress noise artifacts, and in the SI-FRETM reconstruction results that further introduce TV and non-negative regularization constraints, the standard deviation of the FRET efficiency histogram and the acceptor-donor concentration ratio histogram is further reduced. This shows that the super-resolution FRET image reconstruction method based on the dipole resonance constraint between the donor and acceptor molecules of the present invention is more robust to noise, can effectively suppress the false positive FRET signal caused by noise artifacts, and improve the fidelity of super-resolution FRET reconstruction and the accuracy of quantitative analysis.

[0126] In order to further demonstrate the effect of the super-resolution FRET image reconstruction method based on the dipole resonance constraint between the donor and acceptor molecules, this example also uses MCF7 living cells for experiments, and uses Acta-G17M plasmid to target the outer membrane of the mitochondria in MCF7 living cells for labeling. Through pre-measurement, a=0.045179054, b=0.000912663, c=0.001961017, d=0.054721272, G=0.381640369, k=2.40367731 were obtained. Figure 3 As shown, compared with linear Wiener SIM-FRET, the iterative reconstruction of the SI-FRETM model can resolve the FRET signal of the mitochondrial outer membrane region that is contaminated by noise in SIM-FRET imaging, which shows that the super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules of the present invention can effectively suppress noise artifacts in living cell FRET imaging.

[0127] It should be noted that, for the sake of convenience, the aforementioned method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the present invention is not limited to the described order of actions, because according to the present invention, certain steps can be performed in other orders or simultaneously.

[0128] Based on the same idea as the super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules in the above-mentioned embodiment, the present invention also provides a super-resolution FRET image reconstruction system based on dipole resonance constraint between donor and acceptor molecules, which can be used to perform the above-mentioned super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules. For ease of explanation, the structural schematic diagram of the embodiment of the super-resolution FRET image reconstruction system based on dipole resonance constraint between donor and acceptor molecules only shows the parts related to the embodiment of the present invention. Those skilled in the art can understand that the illustrated structure does not constitute a limitation on the device, and may include more or fewer components than shown in the diagram, or combine certain components, or arrange the components differently.

[0129] like Figure 4 As shown, another embodiment of the present invention provides a super-resolution FRET image reconstruction system based on dipole resonance constraint between donor and acceptor molecules, comprising an image acquisition module, an image processing module, a model building module, a target solving module and a FRET calculation module;

[0130] The image acquisition module is used to acquire a FRET three-channel raw image data set of the living cell FRET sample imaging through a structured light illumination super-resolution microscopy imaging system; the FRET three channels include a DD channel of donor excitation donor emission, a DA channel of donor excitation acceptor emission, and an AA channel of acceptor excitation acceptor emission;

[0131] The image processing module is used to perform image preprocessing on the FRET three-channel original image data set to obtain a preprocessed FRET three-channel image data set;

[0132] The model building module is used to perform correlation constraints on super-resolution SIM imaging based on the prior information of the donor-acceptor molecular dipole resonance, and to construct a unified forward model of SI-FRETM, including FRET three-channel spectral mixing parameters, FRET three-channel cosine structured light field, FRET three-channel point spread function and FRET sample fluorescence distribution;

[0133] The target solving module is used to take the FRET sample fluorescence distribution as the target, establish the reconstruction target function according to the SI-FRETM unified forward model and perform forward iterative solution to obtain the reconstructed FRET sample fluorescence distribution; the FRET sample fluorescence distribution includes the total super-resolution donor fluorescence distribution, the total super-resolution acceptor fluorescence distribution and the super-resolution sensitized FRET fluorescence distribution;

[0134] The FRET calculation module is used to obtain the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution by pixel-by-pixel calculation based on the reconstructed super-resolution donor fluorescence total distribution, super-resolution acceptor fluorescence total distribution and super-resolution sensitized FRET fluorescence distribution.

[0135] It should be noted that the super-resolution FRET image reconstruction system based on dipole resonance constraints between donor and acceptor molecules of the present invention corresponds one to one with the super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules of the present invention. The technical features and beneficial effects described in the embodiment of the super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules are applicable to the embodiment of the super-resolution FRET image reconstruction system based on dipole resonance constraints between donor and acceptor molecules. For specific contents, please refer to the description in the embodiment of the method of the present invention, which will not be repeated here. This is hereby declared.

[0136] In addition, in the implementation of the super-resolution FRET image reconstruction system based on dipole resonance constraints between donor and acceptor molecules in the above-mentioned embodiment, the logical division of each program module is only an example. In actual applications, the above-mentioned functions can be assigned to different program modules as needed, for example, for the convenience of corresponding hardware configuration requirements or software implementation. That is, the internal structure of the super-resolution FRET image reconstruction system based on dipole resonance constraints between donor and acceptor molecules is divided into different program modules to complete all or part of the functions described above.

[0137] See also Figure 5In one embodiment, an electronic device for implementing a super-resolution FRET image reconstruction method based on dipole resonance constraints between donor and acceptor molecules is provided. The electronic device may include a first processor, a first memory and a bus, and may also include a computer program stored in the first memory and executable on the first processor, such as a super-resolution FRET image reconstruction program based on dipole resonance constraints between donor and acceptor molecules.

[0138] Wherein, the first memory includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the first memory can be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the first memory can also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the first memory can also include both an internal storage unit of the electronic device and an external storage device. The first memory can not only be used to store application software and various types of data installed in the electronic device, such as the code of a super-resolution FRET image reconstruction program based on dipole resonance constraints between donor and acceptor molecules, but can also be used to temporarily store data that has been output or is to be output.

[0139] In some embodiments, the first processor may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The first processor is the control core (ControlUnit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the first memory (such as a super-resolution FRET image reconstruction program based on dipole resonance constraints between donor and acceptor molecules, etc.), and calls data stored in the first memory to perform various functions of the electronic device and process data.

[0140] Figure 5 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 5 The structure shown does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0141] The super-resolution FRET image reconstruction program based on dipole resonance constraint between donor and acceptor molecules stored in the first memory of the electronic device is a combination of multiple instructions, and when running in the first processor, can achieve:

[0142] A FRET three-channel raw image data set of the living cell FRET sample imaging is obtained by a structured light illumination super-resolution microscopy imaging system; wherein the FRET three channels include a DD channel of donor excitation donor emission, a DA channel of donor excitation acceptor emission, and an AA channel of acceptor excitation acceptor emission;

[0143] Performing image preprocessing on the FRET three-channel original image data set to obtain a preprocessed FRET three-channel image data set;

[0144] Based on the prior information of the dipole resonance of donor-acceptor molecules, the super-resolution SIM imaging is associated with constraints, and the SI-FRETM unified forward model is constructed, including the FRET three-channel spectral mixing parameters, the FRET three-channel cosine structured light field, the FRET three-channel point spread function and the FRET sample fluorescence distribution.

[0145] Taking the FRET sample fluorescence distribution as the target, the reconstruction objective function is established according to the SI-FRETM unified forward model and forward iteration is performed to obtain the reconstructed FRET three-channel fluorescence distribution; wherein, the FRET sample fluorescence distribution includes the total super-resolution donor fluorescence distribution, the total super-resolution acceptor fluorescence distribution and the super-resolution sensitized FRET fluorescence distribution;

[0146] Based on the reconstructed super-resolution donor fluorescence total distribution, super-resolution acceptor fluorescence total distribution and super-resolution sensitized FRET fluorescence distribution, the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution are obtained by pixel-by-pixel calculation.

[0147] Furthermore, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile computer-readable storage medium. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0148] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0149] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0150] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules, characterized in that: The steps include: Acquire a FRET three-channel raw image data set of a living cell FRET sample by a structured light illumination super-resolution microscopy imaging system; the FRET three-channel includes a DD channel for donor excitation and donor emission, a DA channel for donor excitation and acceptor emission, and an AA channel for acceptor excitation and acceptor emission; Performing image preprocessing on the FRET three-channel original image data set to obtain a preprocessed FRET three-channel image data set; Based on the prior information of the dipole resonance of donor-acceptor molecules, the super-resolution SIM imaging is associated with constraints, and a SI-FRETM unified forward model is constructed; the SI-FRETM unified forward model includes FRET sample fluorescence distribution, FRET three-channel spectral mixing parameters, FRET three-channel cosine structured light field and FRET three-channel point spread function; Taking the FRET sample fluorescence distribution as the reconstruction target, a reconstruction objective function is established according to the SI-FRETM unified forward model and forward iterative solution is performed to obtain a reconstructed FRET sample fluorescence distribution; the FRET sample fluorescence distribution includes a super-resolution donor fluorescence total distribution, a super-resolution acceptor fluorescence total distribution and a super-resolution sensitized FRET fluorescence distribution; Based on the reconstructed super-resolution donor fluorescence total distribution, super-resolution acceptor fluorescence total distribution and super-resolution sensitized FRET fluorescence distribution, the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution are obtained by pixel-by-pixel calculation.

2. The super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules according to claim 1, characterized in that: The image preprocessing comprises: Taking the original image of the AA channel as a reference, the original images of the DD channel and the DA channel are aligned pixel by pixel by using an affine transformation to obtain a FRET three-channel registration image data set; the affine transformation is a linear transformation operation, including rotation, translation, scaling and reflection; The grayscale value statistics of the FRET three-channel registered image data set are performed pixel by pixel, and the grayscale value corresponding to the first peak in the grayscale value histogram is used as the background grayscale value to perform pixel by pixel background correction on the FRET three-channel registered image data set to obtain the preprocessed FRET three-channel image data set.

3. The super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules according to claim 1, characterized in that: The construction of the SI-FRETM unified forward model is specifically as follows: The FRET three-channel spectral mixing parameter matrix is ​​obtained by measuring the FRET reference sample; the FRET three-channel spectral mixing parameter matrix is ​​used to characterize the mixing degree of spectral overlap and crosstalk between the three FRET channels caused by the physical process of the donor-acceptor molecule dipole resonance, specifically: Where m ij , i,j∈[1,3] represents the contribution of the donor fluorophore, acceptor fluorophore or sensitized FRET fluorophore in the FRET sample fluorescence distribution X to the fluorescence signal collected by the FRET three channels; the FRET reference sample usually includes a single-transfection donor sample, a single-transfection acceptor sample and a standard plasmid sample with a donor-acceptor ratio of 1:1; The structured light field parameter estimation method is used to estimate the preprocessed FRET three-channel image data set to obtain the FRET three-channel cosine structured light field, which is expressed as: in, It is the FRET three-channel cosine structured light field. is the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel, r is the spatial coordinate of the fluorescence image pixel of the corresponding channel, θ and n represent the different direction angles and different phases of the structured light field, respectively. is the wave vector of the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel, is the modulation depth of the DD channel cosine structured light field, the DA channel cosine structured light field or the AA channel cosine structured light field, is the initial phase of the cosine structured light field of the DD channel, the cosine structured light field of the DA channel, or the cosine structured light field of the AA channel; The FRET three-channel point spread function is obtained by actual measurement using a microscopic imaging system or simulation based on a theoretical model, and is expressed as: Among them, H FRET is the FRET three-channel point spread function, H DD (r) is the point spread function of the DD channel, H DA (r) is the point spread function of the DA channel, H AA (r) is the point spread function of the AA channel, and r is the spatial coordinate of the fluorescence image pixel of the corresponding channel; According to the forward physical process of structured light excitation of FRET samples and acquisition of FRET three-channel images, the SI-FRETM unified forward model is constructed by combining the prior information of donor-acceptor molecular dipole resonance to perform correlation constraints on super-resolution SIM imaging. The SI-FRETM unified forward model includes the FRET sample fluorescence distribution, the FRET three-channel spectral mixing parameters, the FRET three-channel cosine structured light field and the FRET three-channel point spread function, which is expressed as: in, is the unified forward model of SI-FRETM, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, M is the FRET three-channel spectral mixing parameter matrix, X = [X1, X2, X3] T is the FRET sample fluorescence distribution, X1 is the total super-resolution donor fluorescence distribution, X2 is the total super-resolution acceptor fluorescence distribution, and X3 is the super-resolution sensitized FRET fluorescence distribution.

4. The super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules according to claim 1, characterized in that: Taking the FRET sample fluorescence distribution as the reconstruction target, the reconstruction objective function of the SI-FRETM unified forward model is established, which is expressed as: Where X = [X1, X2, X3] T is the target reconstructed FRET sample fluorescence distribution, X1 is the total super-resolution donor fluorescence distribution, X2 is the total super-resolution acceptor fluorescence distribution, X3 is the super-resolution sensitized FRET fluorescence distribution, Unified forward model for SI-FRETM; is the data likelihood term, and the likelihood term is characterized in the form of least squares: Among them, H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, respectively, and M is the FRET three-channel spectral mixing parameter; According to the functional properties of the data likelihood term, a suitable optimization algorithm is designed to perform forward iterative solution on the reconstruction objective function of the SI-FRETM unified forward model to obtain the reconstructed FRET sample fluorescence distribution.

5. The super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules according to claim 1 or claim 4, characterized in that: The regularization term constraint of the fluorescence image prior information is added to the reconstruction objective function of the SI-FRETM unified forward model, and the regularized reconstruction objective function of the SI-FRETM unified forward model is established, which is expressed as: Where X = [X1, X2, X3] T is the target reconstructed FRET sample fluorescence distribution, X1 is the total super-resolution donor fluorescence distribution, X2 is the total super-resolution acceptor fluorescence distribution, X3 is the super-resolution sensitized FRET fluorescence distribution, Unified forward model for SI-FRETM; is the data likelihood term, R(X) is the regularization term, and λ is the weight factor for weighing the data likelihood term and the regularization term; The data likelihood term The likelihood term is characterized in the form of least squares: H FRET is the FRET three-channel point spread function, is the FRET three-channel cosine structured light field, θ and n represent the different direction angles and different phases of the structured light field, respectively, and M is the FRET three-channel spectral mixing parameter; The regularization term R(X) is used to add prior information of the fluorescence image; According to the functional properties of the regularization term, a suitable optimization algorithm is designed to perform forward iterative solution on the regularized reconstruction objective function of the SI-FRETM unified forward model to obtain the reconstructed FRET sample fluorescence distribution.

6. The super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules according to claim 5, characterized in that: The regularization term is a spatial continuity regularization and / or a non-negative regularization.

7. The super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules according to claim 1, characterized in that: The super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution are calculated as follows: Among them, E D is the apparent efficiency centered on the donor, E A is the receptor-centered apparent efficiency, R C is the acceptor-donor concentration ratio distribution, X1 is the reconstructed super-resolution donor fluorescence total distribution, X2 is the reconstructed super-resolution acceptor fluorescence total distribution, and X3 is the reconstructed super-resolution sensitized FRET fluorescence distribution.

8. A super-resolution FRET image reconstruction device based on dipole resonance constraint between donor and acceptor molecules, characterized in that: A super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules as described in any one of claims 1 to 7, comprising an image acquisition module, an image preprocessing module, a model building module, a target solving module and a FRET calculation module; The image acquisition module is used to acquire a FRET three-channel raw image data set of a living cell FRET sample imaging through a structured light illumination super-resolution microscopy imaging system; the FRET three channels include a DD channel for donor excitation and donor emission, a DA channel for donor excitation and acceptor emission, and an AA channel for acceptor excitation and acceptor emission; The image preprocessing module is used to perform image preprocessing on the FRET three-channel original image data set to obtain a preprocessed FRET three-channel image data set; The model building module is used to perform association constraints on super-resolution SIM imaging based on prior information of donor-acceptor molecular dipole resonance, and to construct a SI-FRETM unified forward model; the SI-FRETM unified forward model includes FRET sample fluorescence distribution, FRET three-channel spectral mixing parameters, FRET three-channel cosine structured light field and FRET three-channel point spread function; The target solving module is used to take the FRET sample fluorescence distribution as the target, establish a reconstruction target function according to the SI-FRETM unified forward model and perform forward iterative solution to obtain the reconstructed FRET sample fluorescence distribution; the FRET sample fluorescence distribution includes the total super-resolution donor fluorescence distribution, the total super-resolution acceptor fluorescence distribution and the super-resolution sensitized FRET fluorescence distribution; The FRET calculation module is used to obtain the super-resolution FRET efficiency distribution and the acceptor-donor concentration ratio distribution by pixel-by-pixel calculation according to the reconstructed super-resolution donor fluorescence total distribution, super-resolution acceptor fluorescence total distribution and super-resolution sensitized FRET fluorescence distribution.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory in communication with the at least one processor; wherein, The memory stores computer program instructions that can be executed by the at least one processor, and the computer program instructions are executed by the at least one processor so that the at least one processor can perform the super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a program, characterized in that: When the program is executed by a processor, the super-resolution FRET image reconstruction method based on dipole resonance constraint between donor and acceptor molecules as described in any one of claims 1 to 7 is implemented.

Images