Particle screening method, device and equipment in cryo-em imaging and storage medium
By dividing the dataset into two halves and performing 3D reconstruction and iterative screening in cryo-electron microscopy imaging, the problems of reliance on manual screening criteria and long time consumption in existing technologies are solved, achieving efficient and automated particle screening and improving the quality of protein density maps.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TSINGHUA UNIVERSITY
- Filing Date
- 2023-03-22
- Publication Date
- 2026-05-12
AI Technical Summary
Existing cryo-electron microscopy particle screening techniques rely too heavily on manually defined screening criteria, resulting in long screening times and low algorithm accuracy. This makes it difficult to quickly and effectively screen out a large number of particles with low signal-to-noise ratios, thus affecting the quality of protein structure reconstruction.
The final particle dataset of protein reconstruction under cryo-electron microscopy imaging is divided into two half datasets with the same number of particles based on a preset partitioning strategy. Three-dimensional reconstruction is performed on each dataset, and particles are screened using the first and second protein density maps. The process is iterated until a preset number of iterations is reached to determine the optimal screening ratio of the final particle dataset.
It enables rapid and automated particle screening, removing nearly half of the useless particles, improving the clarity and resolution of protein density maps, reducing reliance on human screening criteria, and improving screening efficiency and accuracy.
Smart Images
Figure CN116434827B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of protein particle screening technology, and in particular to a particle screening method, apparatus, equipment and storage medium for cryo-electron microscopy imaging. Background Technology
[0002] Cryo-electron microscopy is a commonly used method for obtaining the three-dimensional structure of proteins. Its basic procedure is as follows:
[0003] 1. Purify and freeze the target protein sample;
[0004] 2. A large number of two-dimensional particle projection images were obtained by using a high-energy electron beam to photograph samples frozen in ice.
[0005] 3. Reconstruct the three-dimensional conformation of proteins from two-dimensional particles using the expectation-maximization algorithm in machine learning.
[0006] However, due to interference from sample preparation, imaging mechanisms, and mathematical modeling, two-dimensional particle datasets contain a large number of particles that affect the accuracy of three-dimensional reconstruction. Therefore, designing effective algorithms for particle screening has become an important issue in the field of cryo-electron microscopy.
[0007] Currently, particle screening in the field of cryo-electron microscopy is accomplished through multiple rounds of clustering, including two-dimensional and three-dimensional classification of particles. In actual operation, after performing multiple rounds of two-dimensional and three-dimensional classification of particles, users roughly select a large number of particles from the micrographs. Then, through multiple rounds of two-dimensional classification, erroneous particles such as ice crystals and carbon films are removed. This process is repeated to reconstruct and classify particles in three dimensions, discarding all particles that are considered "bad particles" based on visual observation. Finally, the particles are classified into categories with homogeneous and uniform structures through three-dimensional classification.
[0008] However, in the process of two-dimensional and three-dimensional classification, the determination of the criteria for good and bad particles in existing technologies relies too heavily on the subjective standards of users. Different users may not have the same standards. Therefore, in order to ensure that low-quality particles are not rejected, users are generally very conservative in their particle screening, tending not to remove particles of ambiguous categories that still have protein shapes. In addition, repeated classification and reconstruction require a lot of time, especially when using three-dimensional classification to select homogeneous particles, which often requires users to repeatedly perform classification, reconstruction, and merging operations for several months. At the same time, the signal-to-noise ratio of particles in cryo-electron microscopy is extremely low, and clustering algorithms are prone to misclassifying particles. If good particles are mistakenly screened out as bad particles, the number of particles used for reconstruction will decrease, reducing the final reconstruction resolution. If bad particles are mistakenly retained as good particles, a certain proportion of bad particles will be mixed in the particles used for reconstruction, reducing the quality of the reconstructed protein structure.
[0009] In summary, existing particle screening technologies rely too heavily on manually defined screening criteria, have long screening times, and low algorithm accuracy, making it difficult to quickly and effectively screen out a large number of particles with low signal-to-noise ratios. These issues urgently need to be addressed. Summary of the Invention
[0010] This application provides a particle screening method, apparatus, device, and storage medium for cryo-electron microscopy imaging, to solve the problems of existing particle screening technologies that rely too much on manually defined screening criteria, have long screening times, low accuracy of screening algorithms, and are difficult to quickly and effectively screen out a large number of low signal-to-noise ratio particles.
[0011] The first aspect of this application provides a particle screening method in cryo-electron microscopy imaging, comprising the following steps: based on a preset partitioning strategy, dividing the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles; performing three-dimensional reconstruction on all particles in each half dataset to obtain a first protein density map, a second protein density map, and a corresponding resolution; performing particle screening on all particles in the corresponding half dataset based on the first and second protein density maps to obtain particles that meet preset conditions, and iterating the three-dimensional reconstruction and particle screening until a preset number of reconstruction-screening iterations is reached to obtain the first protein density map, the second protein density map, and the corresponding resolution for each iteration, and determining the optimal screening ratio for the final particle dataset.
[0012] Optionally, in one embodiment of this application, the step of filtering all particles in the corresponding half-dataset based on the first protein density map and the second protein density map to obtain the particles that meet the preset conditions includes: calculating the features of all particles based on the first protein density map, the second protein density map, and the corresponding half-dataset; sorting the features of all particles in descending order, and removing the M particles with the largest values from each half-dataset, where M is a positive integer.
[0013] Optionally, in one embodiment of this application, the characteristic formula for each particle is:
[0014]
[0015] Among them, T f (x) is the truncation operator in the Fourier domain. For the protein density map, A i For composite observation operators of particles in cryo-electron microscopy imaging, This is a Discrete Fourier Transform, where f is the cutoff frequency and b is the frequency. i For each of the half-datasets, the particle is...
[0016] Optionally, in one embodiment of this application, before dividing the final protein reconstructed particle dataset under cryo-electron microscopy imaging into two half datasets with the same number of particles based on the preset partitioning strategy, the method further includes: estimating the rotation angle, translation amount, and CTF imaging parameters of each particle in the final protein reconstructed particle dataset.
[0017] A second aspect of this application provides a particle screening device for cryo-electron microscopy imaging, comprising: a partitioning module, configured to partition the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles based on a preset partitioning strategy; a reconstruction module, configured to perform three-dimensional reconstruction on all particles in each half dataset to obtain a first protein density map, a second protein density map, and a corresponding resolution; and a screening module, configured to screen all particles in the corresponding half dataset based on the first and second protein density maps to obtain particles that meet preset conditions, and to iterate the three-dimensional reconstruction and particle screening until a preset number of reconstruction-screening iterations is reached to obtain the first protein density map, the second protein density map, and the corresponding resolution for each iteration, and to determine the optimal screening ratio for the final particle dataset.
[0018] Optionally, in one embodiment of this application, the filtering module includes: a calculation unit, configured to calculate the features of all particles based on the first protein density map and the second protein density map and the corresponding half-dataset; and a removal unit, configured to sort the features of all particles in descending order and remove the M particles with the largest values from each half-dataset, wherein M is a positive integer.
[0019] Optionally, in one embodiment of this application, the characteristic formula for each particle is:
[0020]
[0021] Among them, T f (x) is the truncation operator in the Fourier domain. For the protein density map, A i For composite observation operators of particles in cryo-electron microscopy imaging, This is a Discrete Fourier Transform, where f is the cutoff frequency and b is the frequency. i For each of the half-datasets, the particle is...
[0022] Optionally, in one embodiment of this application, it further includes: an estimation module, used to estimate the rotation angle, translation amount, and CTF imaging parameters of each particle in the final particle dataset of protein reconstruction under cryo-electron microscopy imaging before dividing the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles based on the preset partitioning strategy.
[0023] A third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the particle screening method in cryo-electron microscopy imaging as described in the above embodiments.
[0024] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the particle screening method in cryo-electron microscopy imaging as described above.
[0025] Therefore, the embodiments of this application have the following beneficial effects:
[0026] This application embodiment can divide the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two equal-sized datasets based on a preset partitioning strategy. Three-dimensional reconstruction is then performed on all particles in each half-dataset to obtain a first protein density map and a second protein density map. Based on the first and second protein density maps, particle screening is performed on all particles in the corresponding half-dataset to obtain particles that meet preset conditions. The three-dimensional reconstruction and particle screening are iterated until a preset number of reconstruction-screening iterations is reached to obtain the first and second protein density maps and their corresponding resolutions for each iteration, and to determine the optimal screening ratio for the final particle dataset. This application, through an efficient and automated particle screening method, can filter out nearly half of the "useless particles" in the final protein dataset used for reconstruction and reveal clearer side chains. Therefore, it can quickly and effectively screen out a large number of low signal-to-noise ratio particles without relying on manual screening criteria, greatly improving the final reconstructed protein density. This solves the problems of existing particle screening technologies that rely too heavily on manually defined screening criteria, have long screening times, low algorithm accuracy, and difficulty in quickly and effectively screening out a large number of low signal-to-noise ratio particles.
[0027] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0028] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:
[0029] Figure 1 This is a flowchart of a particle screening method in cryo-electron microscopy imaging according to an embodiment of this application;
[0030] Figure 2 A schematic diagram of the logical architecture of a particle screening method in cryo-electron microscopy imaging provided for one embodiment of this application;
[0031] Figure 3 This application provides a quality curve of protein density map obtained by screening the public dataset EMPIAR-10024 using different screening methods based on the Q-score metric.
[0032] Figure 4 An embodiment of this application provides a quality curve of protein density map obtained by performing particle screening on the public dataset EMPIAR-10024 using different screening methods after a model to map resolution metric.
[0033] Figure 5 This application provides a quality curve of protein density map obtained by particle screening of the public dataset EMPIAR-10024 using different screening methods based on a half-map resolution metric.
[0034] Figure 6 A histogram of particle number versus category resolution in two-dimensional classification is provided as an embodiment of this application;
[0035] Figure 7 This application provides a quality curve of protein density map obtained by particle screening of the public dataset EMPIAR-10097 using different screening methods based on a half-density map resolution metric.
[0036] Figure 8 An embodiment of this application provides a model-to-density map resolution metric method that uses different screening methods to perform particle screening on the public dataset EMPIAR-10097 to obtain a quality curve of the protein density map;
[0037] Figure 9 This application provides a quality curve diagram of protein density map obtained after particle screening of the public dataset EMPIAR-10097 using different screening methods based on the Q score metric.
[0038] Figure 10 A protein density map reconstructed from three dimensions using screened and unscreened particles is provided as an embodiment of this application;
[0039] Figure 11 This is an example diagram of a particle sorting device in cryo-electron microscopy imaging according to an embodiment of this application;
[0040] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0041] Among them, 10-particle screening device in cryo-electron microscopy imaging, 100-division module, 200-reconstruction module, 300-screening module, 1201-memory, 1202-processor, and 1203-communication interface. Detailed Implementation
[0042] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0043] The particle screening method, apparatus, device, and storage medium in cryo-electron microscopy imaging according to embodiments of this application are described below with reference to the accompanying drawings. Addressing the problems mentioned in the background art, this application provides a particle screening method in cryo-electron microscopy imaging. In this method, based on a preset partitioning strategy, the final particle dataset of protein reconstruction under cryo-electron microscopy imaging is divided into two halves of the dataset with the same number of particles. Three-dimensional reconstruction is performed on all particles in each half of the dataset to obtain a first protein density map and a second protein density map. Based on the first and second protein density maps, particle screening is performed on all particles in the corresponding half of the dataset to obtain particles that meet preset conditions. The three-dimensional reconstruction and particle screening are iterated until a preset number of reconstruction-screening iterations is reached to obtain the first protein density map, the second protein density map, and the corresponding resolution for each iteration, and to determine the optimal screening ratio for the final particle dataset. This application, through an efficient and automated particle screening method, can filter out nearly half of the "useless particles" in the final dataset used for protein reconstruction and can see clearer side chains. Therefore, without relying on manual screening criteria, it can quickly and effectively screen out a large number of low signal-to-noise ratio particles, greatly improving the final reconstructed protein density. This solves the problems of existing particle screening technologies, such as over-reliance on manually defined screening criteria, long screening time, low accuracy of screening algorithms, and difficulty in quickly and effectively screening out a large number of low signal-to-noise ratio particles.
[0044] Specifically, Figure 1This is a flowchart of a particle screening method in cryo-electron microscopy imaging provided in an embodiment of this application.
[0045] like Figure 1 As shown, the particle screening method in cryo-electron microscopy imaging includes the following steps:
[0046] In step S101, based on a preset partitioning strategy, the final particle dataset of protein reconstruction under cryo-electron microscopy imaging is divided into two half datasets with the same number of particles.
[0047] Those skilled in the art should understand that in the final particles used for reconstruction in the publicly available protein dataset (EMPIAR) commonly used for cryo-electron microscopy, about half of them are "bad particles". When removing the above-mentioned "bad particles" in the embodiments of this application, the final particle dataset of protein reconstruction can be divided into two half datasets with the same number of particles, thereby providing data support for subsequent three-dimensional reconstruction and particle screening.
[0048] Optionally, in one embodiment of this application, before dividing the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles based on a preset partitioning strategy, the method further includes: estimating the rotation angle, translation amount, and CTF imaging parameters of each particle in the final particle dataset of protein reconstruction.
[0049] It should be noted that before reconstructing and filtering the particles, the embodiments of this application further divide the dataset into two halves according to the Gold-Standard criterion for estimating latent parameters. In each half-dataset, the first and second protein density maps are independently reconstructed, and the latent parameters (rotation angle, translation, and CTF imaging parameters of each particle, etc.) in the final protein reconstructed particle dataset are estimated. The same dataset division is performed on the final protein reconstructed particle dataset after the latent parameters have been estimated, thereby ensuring the independence of the two halves and guaranteeing the reliability and effectiveness of the dataset division.
[0050] In step S102, three-dimensional reconstruction is performed on all particles in each half-dataset to obtain a first protein density map, a second protein density map, and the corresponding resolution.
[0051] Furthermore, to ensure the gold standard assumption that the half-datasets commonly used in cryo-electron microscopy remain independent, the embodiments of this application can perform three-dimensional reconstruction and screening operations only within each half-dataset. Specifically, the embodiments of this application can utilize all particles within each half-dataset to perform three-dimensional reconstruction separately to obtain the protein density map and its corresponding resolution for each half-dataset, thereby providing a basis for particle screening.
[0052] In step S103, particle screening is performed on all particles in the corresponding half dataset based on the first protein density map and the second protein density map to obtain particles that meet the preset conditions. The three-dimensional reconstruction and particle screening are iterated until the preset number of reconstruction-screening iterations is reached to obtain the first protein density map, the second protein density map and the corresponding resolution for each iteration, and to determine the optimal screening ratio of the final particle dataset.
[0053] After performing 3D reconstruction on all particles in each half-dataset to obtain the corresponding protein density map, the embodiments of this application can further use the two protein density maps obtained from the 3D reconstruction to screen all particles in the corresponding half-dataset. After removing the screened "bad particles", the remaining "good particles" in each half-dataset are used to continue to perform 3D reconstruction. Then, the 3D reconstruction results are used again to screen the remaining particles in the corresponding half-dataset. The above reconstruction and screening operations are iterated, and the protein density map and resolution corresponding to the two half-datasets are obtained in each iteration. This determines the optimal screening ratio of the final particle dataset to obtain a protein density map with clearer side chains.
[0054] In the embodiments of this application, 10 rounds of reconstruction-screening iterations can be performed, and each round of iteration can filter out 20% of the "bad particles". Those skilled in the art can set different numbers of reconstruction-screening iterations according to the actual situation during the execution process, and no specific limitation is made here.
[0055] Optionally, in one embodiment of this application, particle screening is performed on all particles in the corresponding half-dataset based on the first protein density map and the second protein density map to obtain particles that meet preset conditions, including: calculating the features of all particles based on the first protein density map, the second protein density map and the corresponding half-dataset; sorting the features of all particles in descending order, and removing the M particles with the largest values from each half-dataset, where M is a positive integer.
[0056] It should be noted that, in each iteration of the embodiments of this application, the screening process for each granular part of the half-data set is as follows:
[0057] 1. Given N particles A 3D reconstructed protein density map In the imaging of particle b i The composite observation operator of the translation, convolution CTF, and projection operators is denoted as A. i ,remember For the Discrete Fourier Transform, define the operator T. f(x) is the truncation operator in the Fourier domain, that is, the part of the signal x with frequency f below f is truncated to 0;
[0058] 2. Assuming the number of defective particles filtered out is M, and the cutoff frequency for high frequencies is f (where f and M are hyperparameters), for each particle b i Calculate features;
[0059] 3. Sort all particles by features from high to low, remove the M particles with the largest values in each half-dataset as bad particles from the dataset, and use the remaining NM "good particles" for 3D reconstruction to execute the next iteration.
[0060] Therefore, the embodiments of this application can screen out nearly half of the "useless particles" in the final dataset used for protein reconstruction, thereby achieving rapid and effective screening of a large number of low signal-to-noise ratio particles without relying on human screening criteria, increasing the particle screening ratio and improving protein density.
[0061] Optionally, in one embodiment of this application, the characteristic formula for each particle is:
[0062]
[0063] Among them, T f (x) is the truncation operator in the Fourier domain. For protein density plot, A i For composite observation operators of particles in cryo-electron microscopy imaging, This is a Discrete Fourier Transform, where f is the cutoff frequency and b is the frequency. i For each half-dataset, there are particles.
[0064] In the embodiments of this application, for each particle b in each half-dataset i Calculate the following features:
[0065]
[0066] Among them, T f (x) is the truncation operator in the Fourier domain. For protein density plot, A i For imaging of particle b i The composite observation operator of translation, convolution CTF, and projection operators was created. This is a Discrete Fourier Transform, where f is the cutoff frequency and b is the frequency. i For each half-dataset, there are particles.
[0067] Therefore, in the above-mentioned iterative process, this application calculates the protein density map, Fourier Shell Correlation (FSC), and the above-mentioned particle features of the good particles reconstructed from the two half datasets in each iteration, records the iteration with the highest reconstruction resolution, and uses the good particles left by the screening in that round as the input for the next iteration, thereby ensuring the quality of 3D reconstruction and particle screening.
[0068] The performance of the particle screening method in cryo-electron microscopy imaging of this application will be analyzed below through specific experiments and in conjunction with the accompanying drawings.
[0069] First, this application used the publicly available dataset EMPIAR-10024 to screen particles. The protein in question is the human TRPA1 channel protein, which has C4 symmetry. A total of 43,585 particles were ultimately used to reconstruct the protein structure.
[0070] When using the particle screening method in cryo-electron microscopy imaging proposed in this application, using the remaining 22,228 particles (approximately 51%) for three-dimensional reconstruction yields a protein density map with clearer side chains than when using all the initial particles for three-dimensional reconstruction; using the remaining 11,332 particles (approximately 26%) for three-dimensional reconstruction yields a protein density map with the same resolution as when using all the initial particles for three-dimensional reconstruction.
[0071] Understandably, model-to-density map resolution and Q score are common metrics for comparing protein density maps with their corresponding atomic models. This application uses model-to-density map resolution and Q score to measure the protein density maps (sharp maps) obtained by different screening methods at different particle counts. These maps are then reconstructed using CryoSPARC software commonly used in cryo-electron microscopy and post-processed with FSC Weighting, Auto b-factor sharpening, etc., as shown below. Figure 2 As shown.
[0072] Figures 3-5 The quality curves of protein density maps obtained by the particle screening method, the normalized cross-correlation coefficient (NCC)-based screening method, and the random screening method in cryo-electron microscopy imaging proposed in this application, based on Q-score, model-to-density map resolution, and half-density map resolution, respectively, at different particle numbers are presented. Figures 3-5 It can be seen that, under various screening ratios, the particle screening method in cryo-electron microscopy imaging proposed in this application is superior to screening based on normalized cross-correlation coefficient (NCC) and random screening.
[0073] To confirm that the particles selected through this application are low-resolution defective particles, this application can use two-dimensional classification, commonly used in cryo-electron microscopy, to plot histograms of particle count versus category resolution for both good and defective particles in the two-dimensional classification, as shown below. Figure 6 As shown, a visual comparison is made between the screened bad particles and the retained good particles, such as... Figure 6 As shown, the quality of the bad particles screened out in this application is significantly lower than that of the good particles retained.
[0074] Furthermore, this application can perform particle screening on the publicly available dataset EMPIAR-10097. This protein is an influenza hemagglutinin trimer protein with C3 symmetry. A total of 130,000 particles were ultimately used to reconstruct the protein structure. Using the particle screening method in the cryo-electron microscopy imaging of this application, after screening the remaining 34,078 particles (about 26%) for three-dimensional reconstruction, a density map with clearer side chains can be obtained than that obtained by reconstructing with all the initial particles.
[0075] Figures 7-9 The quality measurement curves of protein density maps obtained from the publicly available dataset EMPIAR-10097 at different particle numbers are presented, respectively, based on Q-score, model-to-density map resolution, and half-density map resolution metrics. These methods include particle selection based on normalized cross-correlation coefficient (NCC), and random selection. Figures 7-9 It can be seen that, under various screening ratios, the particle screening method in cryo-electron microscopy imaging proposed in this application is superior to screening based on normalized cross-correlation coefficients and random screening.
[0076] Figure 10 This is a 3D reconstructed protein density map using both screened and unscreened particles. (Example:) Figure 10 As shown, the protein density map reconstructed using all 130,000 particles (dark) and the protein density map reconstructed using 34,078 particles (light) are of higher quality overall. Locally, it can be observed that the protein density map reconstructed using fewer selected particles has clearer side chains.
[0077] Therefore, experimental analysis shows that removing "bad particles" from the dataset and then performing 3D reconstruction does not worsen the original 3D reconstruction results. On the contrary, under certain metrics, performing 3D reconstruction on the remaining particles after removing nearly half of the "bad particles" can yield a protein density map of higher quality than reconstructing with all particles.
[0078] According to the particle screening method in cryo-electron microscopy imaging proposed in this application, the final particle dataset of protein reconstruction under cryo-electron microscopy imaging is divided into two halves of the dataset with the same number of particles based on a preset partitioning strategy. Three-dimensional reconstruction is performed on all particles in each half of the dataset to obtain a first protein density map and a second protein density map. Based on the first and second protein density maps, particle screening is performed on all particles in the corresponding half of the dataset to obtain particles that meet preset conditions. The three-dimensional reconstruction and particle screening are iterated until a preset number of reconstruction-screening iterations is reached to obtain the first protein density map, the second protein density map, and the corresponding resolution for each iteration, and to determine the optimal screening ratio for the final particle dataset. This application, through an efficient and automated particle screening method, can filter out nearly half of the "useless particles" in the final dataset used for protein reconstruction and can see clearer side chains. Therefore, without relying on manual screening criteria, it can quickly and effectively screen out a large number of low signal-to-noise ratio particles, greatly improving the final protein density.
[0079] Next, with reference to the accompanying drawings, a particle screening device for cryo-electron microscopy imaging according to an embodiment of this application is described.
[0080] Figure 11 This is a block diagram of a particle screening device in cryo-electron microscopy imaging according to an embodiment of this application.
[0081] like Figure 11 As shown, the particle screening device 10 in cryo-electron microscopy imaging includes: a division module 100, a reconstruction module 200, and a screening module 300.
[0082] The partitioning module 100 is used to divide the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles, based on a preset partitioning strategy.
[0083] The reconstruction module 200 is used to perform three-dimensional reconstruction on all particles in each half-dataset to obtain a first protein density map, a second protein density map, and the corresponding resolution.
[0084] The filtering module 300 is used to filter all particles in the corresponding half dataset based on the first protein density map and the second protein density map to obtain particles that meet the preset conditions. The 3D reconstruction and particle filtering are iterated until the preset number of reconstruction-filtering iterations is reached to obtain the first protein density map, the second protein density map and the corresponding resolution for each iteration, and to determine the optimal filtering ratio of the final particle dataset.
[0085] Optionally, in one embodiment of this application, the screening module 300 includes a calculation unit and a rejection unit.
[0086] The computing unit is used to calculate the features of all particles based on the first protein density map, the second protein density map, and the corresponding half dataset.
[0087] The elimination unit is used to sort all particles in descending order of their features and remove the M particles with the largest values from each half dataset, where M is a positive integer.
[0088] Optionally, in one embodiment of this application, the characteristic formula for each particle is:
[0089]
[0090] Among them, T f (x) is the truncation operator in the Fourier domain. For protein density plot, A i For composite observation operators of particles in cryo-electron microscopy imaging, This is a Discrete Fourier Transform, where f is the cutoff frequency and b is the frequency. i For each half-dataset, there are particles.
[0091] Optionally, in one embodiment of this application, the particle screening device 10 in cryo-electron microscopy imaging of this application embodiment further includes: an estimation module, used to estimate the rotation angle, translation amount and CTF imaging parameters of each particle in the final particle dataset of protein reconstruction based on a preset partitioning strategy into two half datasets with the same number of particles.
[0092] It should be noted that the foregoing explanation of the particle screening method embodiment in cryo-electron microscopy imaging also applies to the particle screening device in cryo-electron microscopy imaging of this embodiment, and will not be repeated here.
[0093] According to the particle screening device in cryo-electron microscopy imaging proposed in this application, the final particle dataset of protein reconstruction under cryo-electron microscopy imaging is divided into two halves of the dataset with the same number of particles based on a preset partitioning strategy. Three-dimensional reconstruction is performed on all particles in each half of the dataset to obtain a first protein density map and a second protein density map. Based on the first and second protein density maps, particle screening is performed on all particles in the corresponding half of the dataset to obtain particles that meet preset conditions. The three-dimensional reconstruction and particle screening are iterated until a preset number of reconstruction-screening iterations is reached. The final first and second protein density maps are obtained based on the final particles that meet the preset conditions. This application, through an efficient and automated particle screening method, can filter out nearly half of the "useless particles" in the final dataset used for protein reconstruction and can see clearer side chains. Therefore, without relying on manual screening criteria, it can quickly and effectively screen out a large number of low signal-to-noise ratio particles, greatly improving the final reconstructed protein density.
[0094] Figure 12 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include:
[0095] The memory 1201, the processor 1202, and the computer program stored on the memory 1201 and executable on the processor 1202.
[0096] When the processor 1202 executes the program, it implements the particle screening method in cryo-electron microscopy imaging provided in the above embodiments.
[0097] Furthermore, electronic devices also include:
[0098] Communication interface 1203 is used for communication between memory 1201 and processor 1202.
[0099] The memory 1201 is used to store computer programs that can run on the processor 1202.
[0100] The memory 1201 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.
[0101] If the memory 1201, processor 1202, and communication interface 1203 are implemented independently, then the communication interface 1203, memory 1201, and processor 1202 can be interconnected via a bus to complete communication between them. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 12 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.
[0102] Optionally, in a specific implementation, if the memory 1201, processor 1202, and communication interface 1203 are integrated on a single chip, then the memory 1201, processor 1202, and communication interface 1203 can communicate with each other through an internal interface.
[0103] The processor 1202 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application.
[0104] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the particle screening method in cryo-electron microscopy imaging as described above.
[0105] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0106] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0107] Any process or method described in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0108] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.
[0109] It should be understood that the various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0110] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.
[0111] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.
[0112] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.
Claims
1. A particle screening method in cryo-electron microscopy imaging, characterized in that, Includes the following steps: Based on a pre-defined partitioning strategy, the final particle dataset of protein reconstruction under cryo-electron microscopy imaging is divided into two half datasets with the same number of particles. Three-dimensional reconstruction was performed on all particles in each half-dataset to obtain the first protein density map, the second protein density map, and the corresponding resolution. Based on the first and second protein density maps, particle screening is performed on all particles in the corresponding half dataset to obtain particles that meet the preset conditions. The 3D reconstruction and particle screening are iterated until the preset number of reconstruction-screening iterations is reached to obtain the first protein density map, the second protein density map and the corresponding resolution for each iteration, and to determine the optimal screening ratio of the final particle dataset. The step of filtering all particles in the corresponding half-dataset based on the first and second protein density maps to obtain particles that meet the preset conditions includes: Based on the first protein density map and the second protein density map, as well as the corresponding half dataset, the features of all particles are calculated. Sort all the features of the particles in descending order, and remove the M particles with the largest values from each half dataset, where M is a positive integer; The characteristic formula for each particle is: in, To truncate the Fourier domain, The protein density map, For composite observation operators of particles in cryo-electron microscopy imaging, For Discrete Fourier Transform, To cut off the frequency, For each of the half-datasets, the particle is...
2. The method according to claim 1, characterized in that, Before dividing the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles based on the preset partitioning strategy, the method further includes: Estimate the rotation angle, translation amount, and CTF imaging parameters of each particle in the final particle dataset of the protein reconstruction.
3. A particle sorting device for cryo-electron microscopy imaging, characterized in that, include: The partitioning module is used to divide the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles, based on a preset partitioning strategy. The reconstruction module is used to perform three-dimensional reconstruction of all particles in each half-dataset to obtain the first protein density map, the second protein density map, and the corresponding resolution. The filtering module is used to filter all particles in the corresponding half dataset based on the first protein density map and the second protein density map to obtain particles that meet the preset conditions, and to iterate the three-dimensional reconstruction and particle filtering until the preset number of reconstruction-filtering iterations is reached, so as to obtain the first protein density map, the second protein density map and the corresponding resolution for each iteration, and to determine the optimal filtering ratio of the final particle dataset. The filtering module includes: The computing unit is used to calculate the features of all particles based on the first protein density map, the second protein density map, and the corresponding half dataset; The elimination unit is used to sort the features of all particles in descending order and remove the M particles with the largest values from each half dataset, where M is a positive integer; The characteristic formula for each particle is: in, To truncate the Fourier domain, The protein density map, For composite observation operators of particles in cryo-electron microscopy imaging, For Discrete Fourier Transform, To cut off the frequency, For each of the half-datasets, the particle is...
4. The apparatus according to claim 3, characterized in that, Also includes: The estimation module is used to estimate the rotation angle, translation amount, and CTF imaging parameters of each particle in the final particle dataset of protein reconstruction under cryo-electron microscopy imaging before dividing the final particle dataset of protein reconstruction under cryo-electron microscopy imaging into two half datasets with the same number of particles based on the preset partitioning strategy.
5. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the particle screening method in cryo-electron microscopy imaging as described in any one of claims 1-2.
6. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the particle screening method in cryo-electron microscopy imaging as described in any one of claims 1-2.