29 results about "Cell clustering" patented technology

The application relates to the technical field of artificial intelligence, in particular to a cell clustering method, device and electronic equipment. The method comprises the following steps: acquiring MDT data coverage boundary maps of a plurality of cells within a first preset time and network index change data of the plurality of cells within a second preset time; performing first clustering on the cells with similar coverage form features of the MDT data coverage boundary maps and similar time distribution features of the network index change data, to obtain cells with different cluster labels; acquiring user number change data of the plurality of cells within a third preset time; and performing second clustering on the cells with similar spatiotemporal distribution features of the user number change data in the cells with different cluster labels. The cell clustering method, device and electronic equipment provided in the application can solve the technical problem that the existing cell clustering / classification method is not good in effect.

The invention discloses a structure-driven single cell transcriptomics immune cell annotation system and method, and the system comprises a data loading and preprocessing module which is used for loading a marker table and data after single cell clustering, and extracting the expression characteristic statistic of each cluster from the marker table and data to form a characteristic matrix; the basic structure annotation module is used for carrying out preliminary structure annotation on peripheral blood mononuclear cells and tissue samples; the tissue extension annotation module is used for completing structure extension and refinement under tissue expression drift on the basis of preliminary structure annotation according to a sample structure annotation requirement from peripheral tissues; and the structure dimension auxiliary annotation module is configured to integrate the dimension reduction coordinates and the adjacency graph information after being triggered, and perform spatial consistency verification and auxiliary correction on the low-confidence annotation result. According to the method, an automatic annotation function with high accuracy, strong controllability and good interpretability can be realized, and the method has high interpretability and robustness.

The invention provides a determination method of signal channels of a cell subset and a cell model, a prescription efficacy prediction method and a related device, and relates to the technical field of traditional Chinese medicines. The method comprises the following steps: performing low-depth single cell transcriptome sequencing and cell grouping and annotation analysis on a disease sample and a normal control sample, screening cell subgroups of which the quantity proportion is remarkably changed in the disease sample, and performing gene expression data aggregation analysis and signal path enrichment analysis, thereby obtaining the disorder signal path of the disease-related cell subgroups. Based on the corresponding relation between the disease-related cell subpopulation and the cell model of the traditional Chinese medicine effect, the cell model and the imbalance signal channel for predicting the efficacy of the prescription are determined, and applications such as quantitative prediction of the efficacy of the prescription with the reversal of the disease signal channel as the core are further developed.

The application discloses a battery cell fault diagnosis method and device, and relates to the technical field of batteries. According to the charging voltage of a battery cell, the sum of the absolute values of MN of a to-be-tested cell and the Euclidean distance are calculated, the sample entropy-scale factor curve slope of the to-be-tested cell is calculated according to the sampling time of the charging voltage, the first Hamming closeness of the sum of the absolute values of MN, the Euclidean distance and the sample entropy-scale factor curve slope of the to-be-tested cell to the normal cell clustering center is calculated, the second Hamming closeness of the sum of the absolute values of MN, the Euclidean distance and the sample entropy-scale factor curve slope of the to-be-tested cell to the fault cell clustering center is calculated, and if the second Hamming closeness is greater than the first Hamming closeness, the to-be-tested cell is judged to be faulty. The battery cell fault is comprehensively diagnosed from three dimensions.

The invention provides a spatial transcriptome signal enhancement method and device, electronic equipment and a storage medium. The spatial transcriptome signal enhancement method comprises the following steps: clustering cells in a target slice based on spatial transcription data of the target slice to obtain a cell cluster set; determining candidate cells of the target cells according to a preset distance threshold; determining first gene expression data of the candidate cells and second gene expression data of the target cells from the spatial transcription data; and combining the first gene expression data and the second gene expression data to enhance the gene expression of the target cell. The embodiment of the invention can provide a simple spatial transcriptome signal enhancement method.

The application relates to a single-cell Hi-C clustering method and system based on contact number weight smoothing and feature fusion, which comprises the following steps: (1) data preprocessing; (2) contact number weight-based smoothing; using a weight matrix to quantify the influence of different neighbor fragments on a target chromosome fragment; (3) restart random walk smoothing; using a restart random walk algorithm to perform smoothing processing on the chromosome contact matrix after the contact number weight-based smoothing processing; (4) contact binarization; (5) principal component extraction; using KPCA to perform dimension reduction and generate cell embedding; (6) feature fusion; (7) spectral clustering; using a spectral clustering algorithm to realize cell clustering. The single-cell Hi-C clustering method provided by the application is obviously superior to existing clustering methods, greatly improves the clustering effect when a large-scale single-cell Hi-C data set is processed, and can identify cell types with a small number of cells in the data set.

The invention relates to the technical field of data analysis, in particular to an integrated analysis method and system for multi-source single cell transcriptome data. The method comprises the following steps: converting all gene names of single cell transcriptome data from different sources into a unified version, and filtering source specific genes; a specific background gene pool is constructed, pollution genes are identified, and background RNA pollution is eliminated; calculating the average sequencing depth of each sample cell, and performing sequencing depth correction on each sample cell; based on transcriptome data of single cells from different sources, cell types are identified through cell grouping and characteristic gene expression quantity analysis, and each cell type is obtained; and constructing anchor point cells for each cell type, and correcting the deep equilibrium cell expression profile of the same cell type to obtain an integrated cell expression profile. The accuracy of data integration analysis of the single cell transcriptome is improved, and the influence of batch effect caused by multiple sources on other subsequent analysis is reduced.

The application belongs to the field of single cell clustering methods, and provides a single cell clustering method and system based on consistency matrix scoring, multiple consistency matrices are obtained after combined dimension reduction based on gene expression data, each consistency matrix is clustered to obtain a corresponding clustering result; the f-value of each consistency matrix is calculated by using a scoring method in combination with the consistency matrix and the corresponding clustering result, the consistency matrix corresponding to the highest f-value score is the optimal consistency matrix; based on the obtained optimal consistency matrix, a distance matrix between cells is constructed, and the final clustering result is obtained by using hierarchical clustering on the distance matrix between cells. The indirect distance between cells is fully utilized, and the clustering effect is improved.

The invention discloses a single cell clustering analysis method and tool, and relates to the technical field of single cell sequencing data analysis. The method comprises the following steps: step 1, subdivision clustering: extracting a specific cell cluster from a Seurt object, carrying out secondary clustering under different resolutions and PCA dimension settings, and carrying out reclassification on cells by utilizing dimension reduction of Seurt and a neighborhood detection algorithm so as to identify a hidden subgroup structure; and step 2, label mapping and integration: a clustering result after subdivision is mapped back to an original Seurt object, a new refinedcluster column is created, an original clustering label and a new label coexist, and different clustering schemes are allowed to be flexibly switched in analysis. According to the method, the self-defined clustering resolution and the PCA dimension are supported, and parameters can be flexibly adjusted according to different data sets and research targets; the subdivided subgroup labels are seamlessly integrated with the original Seurt objects, and original clustering information is reserved at the same time, so that subsequent analysis is facilitated; a visualization function which is perfectly compatible with Seurt is provided, and visual display and comparison of subdivision results are supported.

This invention discloses a cross-modal cell alignment method for pathological images in non-registerable scenarios. The method acquires H&E and mIF stained images and their data-augmented views corresponding to spatial anatomical locations; constructs H&E modality processing branches, mIF modality processing branches, and a reverse self-attention module; and sequentially performs three-stage progressive training: intra-modal contrastive pre-training based on the data-augmented view; using mIF modality-level semantic aggregation tokens from the same spatial anatomical location as positive samples and mIF modality-level semantic aggregation tokens from different spatial anatomical locations as negative samples, achieving H&E to mIF modality-level semantic alignment through contrastive learning; inputting the aligned H&E modality-level semantic aggregation tokens and the projected H&E cell feature embedding set into the reverse self-attention module, and achieving unsupervised cell-level semantic alignment by minimizing the Hungarian matching loss; finally, the output H&E cell features semantically aligned with mIF are embedded into a refined set, which can be directly used for downstream pathological analysis tasks such as high-precision cell clustering.

The embodiment of the invention provides a cell subset division optimization method and device based on a single cell clustering result. The method is applied to the technical field of medical data analysis, and comprises the following steps: performing clustering analysis on single-cell RNA sequencing data according to a preset initial clustering resolution to obtain a plurality of cell subgroups after preliminary clustering; performing differential expression analysis on each cell subset obtained by preliminary clustering to obtain a differential expression gene quantity corresponding to each cell subset; and under the condition that the cell subgroups with the differential expression gene quantity lower than a preset threshold value exist in all the cell subgroups obtained by the preliminary clustering, performing automatic iteration merging on the cell subgroups obtained by the preliminary clustering until the differential expression gene quantity of all the merged cell subgroups is not lower than the preset threshold value. And outputting an optimized cell subset division result. According to the method, the accuracy of cell subset division and the efficiency of single cell data analysis are improved, and the biological rationality and interpretability of an analysis result are improved.

The invention belongs to the technical field of single cell clustering, and particularly relates to a single cell clustering method and system based on multi-granularity Transform. In order to solve the problems of insufficient cell genome relation extraction, limited clustering accuracy and the like in scRNA-seq data, the method provides a multi-granularity feature learning framework, hierarchical Token division is carried out on gene expression data, multi-level association from local gene collaboration to a global function module is captured by utilizing an attention mechanism, and the clustering accuracy is improved. And thus, cell global representation with higher discrimination is constructed. Furthermore, by introducing a cluster contrast learning mechanism, feature collision possibly caused by negative sample pairs in traditional contrast learning is avoided, and the clustering friendliness of a feature space is enhanced.

The application discloses a single-cell RNA sequence data clustering method based on a robust residual graph convolutional network, and comprises the following steps: determining a data set: selecting several public single-cell RNA sequence data sets; single-cell RNA sequence data preprocessing: performing cell and gene screening operations on the single-cell RNA sequence data set; network construction: constructing a single-cell RNA sequence data clustering network based on a robust residual graph convolutional network; network training: inputting the single-cell RNA sequence data into the constructed network for network training, and evaluating the current clustering performance by four clustering evaluation indexes after the training is completed; and cell clustering: inputting the single-cell RNA sequence data into the trained clustering network to obtain a clustering result. The application can make high-accuracy clustering on single-cell RNA sequence data in view of the noise problem existing in the single-cell RNA sequence data, and can be used for single-cell type annotation and recognition.

The invention relates to the technical field of infrared focal plane detector performance testing, and discloses a blind pixel clustering analysis-based infrared focal plane detector failure mechanism determination method, which comprises the steps of test preparation, response data acquisition, blind pixel identification, clustering analysis and failure mechanism inference. According to the method, on one hand, blind pixel recognition is completed by collecting the multi-dimensional original response data and combining multi-threshold judgment, and the accuracy of a blind pixel sample is ensured; and on the other hand, a multi-feature parameter set is formed by extracting the output level average value, the noise root-mean-square value and the detection response rate of the blind element, feature parameters are classified and combined by means of a clustering model to lock a core feature combination, and a failure mechanism type is inferred by combining a preset clustering cluster and failure mechanism mapping relation. The accuracy and the standardization degree of failure mechanism diagnosis are improved, meanwhile, reliability evaluation of the infrared focal plane detector is completed synchronously, and the actual requirements of detector failure root accurate positioning and reliability evaluation are met.

This invention proposes a cancer single-cell type identification method based on optimal transport subspace clustering. By employing a local graph-guided learning strategy, this invention effectively addresses a common problem in deep subspace clustering: the tendency to converge to suboptimal solutions during self-expression learning, leading to poor clustering results. This strategy significantly improves the accuracy and robustness of the clustering process. Furthermore, this invention introduces the Wasserstein regularized self-expression learning method, combined with the optimal transport algorithm, greatly enhancing the model's ability to learn subspace structures in cell clustering tasks. This enables the model to obtain more robust and reliable subspace representations, thereby improving clustering quality.

The invention discloses a single-cell multi-omics dimension reduction method based on a Gaussian process hidden variable model, and aims to solve the problems that single-omics and multi-omics data analysis requirements are difficult to consider and high sparsity and technical noise of single-cell data cannot be effectively handled in the prior art. According to the method, a shared potential space probability generation framework is constructed, variational inference and a sparse Gaussian regression strategy are combined, data of different modalities are non-linearly mapped to a unified low-dimensional space, and meanwhile, an adjustable modal weight is introduced, so that the model can perform efficient dimensionality reduction on single-cell transcriptome and multi-omics data. When the method is used for carrying out dimension reduction on single cell sequencing data, low-dimensional representation learned by the model can be used for completing downstream tasks such as cell clustering, visualization and differential expression analysis, cell heterogeneity and potential biological regulation signals are accurately revealed, and the method has a good application prospect.

The invention provides a rare type cell detection method and system, and belongs to the technical field of data processing, and the method comprises the following steps: inputting a single cell gene expression or apparent characteristic matrix including N cells and M characteristics; calculating a neighborhood internal connectivity Q value of each cell according to the number of connecting edges between each cell and k neighbor cells (including itself) and the total number of edges; initializing each cell into an independent class group, performing network diffusion of class group labels based on a gradient of a neighborhood internal connectivity Q value, and determining a cell class group; and merging and optimizing a part of low-quality class groups through the internal connectivity Qc value of the cell class groups so as to determine a final cell clustering result. According to the technology, the rare cell population can be quickly found from single-omics, multi-omics and high-resolution spatial omics data, and the prediction efficiency, accuracy and applicability are improved.

The application discloses a kind of based on multi-scale contrast learning's spatial transcriptomics cell clustering method, comprising the following steps: S1, data preprocessing is carried out;S2, using the data of processing is carried out graph construction;S3, data enhancement is carried out;S4, using GCN extracts cell gene expression information;S5, using ContraNorm layer rich node information;S6, using multi-scale graph contrast learning further learns information, and using the final information learned carries out biological analysis.The application utilizes computer-aided cell type analysis, without a large number of high-quality data, just using gene expression data and cell spatial position information data, can predict which cells belong to the same category, provide more effective information to help researchers identify the origin of cancer and disease development.

The invention relates to bioinformatics technology and specifically to an analysis method of a spatial transcriptome-based 4D-printed stem cell scaffold for enhancing diabetic skin injury healing. Spatial transcriptome sequencing is performed on tissues from both the PBS group and the combined treatment group. Cell clustering and annotation identify main cell types, followed by visualization of their spatial distributions in both groups. The proportions of various cell types are statistically quantified and plotted. Cell subsets exhibiting significant differential gene expression are enriched and verified. Using a 4D-printed stem cell scaffold combined with a piRNA inhibitor and stem cell therapy, the method reveals spatial distribution and dynamic cellular changes during wound healing. It identifies precise cellular locations and gene expression patterns within tissues, providing critical insights into the biological processes by which biomaterials promote diabetic injury repair. This approach offers new directions for future drug development.

The invention discloses a self-supervised cross-modal alignment and joint representation learning method oriented to single-cell multi-modal data, and relates to a self-supervised cross-modal alignment and joint representation learning method. The invention aims to solve the problems that in the prior art, single-modal analysis cannot correlate T cell antigen specificity and functional phenotype, and a multi-modal tool is insufficient in characterization granularity and weak in generalization ability. The method comprises the following steps: collecting and preprocessing single cell transcriptome and single cell TCR sequencing pairing data, and constructing a training set and a test set; constructing a gene expression single-mode module and a TCR single-mode module; alignment and fusion of two modes are realized through a multi-mode fusion module. Meanwhile, a multi-mode and single-mode loss function is designed for self-supervised training, and GEXunimodal, TCRunimodal and multi-mode combined representation are finally output, and downstream tasks such as cell clustering annotation, antigen specificity prediction and TCR-to-gene expression generation are adapted. The invention belongs to the technical field of bioinformatics and single-cell multi-omics analysis.

The application relates to a self-optimization single-cell clustering method, and belongs to the technical field of single-cell RNA sequencing data analysis in bioinformatics. A ZINB model and a graph attention autoencoder are combined to denoise single-cell RNA sequencing data, the influence of data noise on a clustering result is reduced, data characteristic distribution is better fitted, the denoising performance of the autoencoder is improved, and the data is subjected to learning of intercellular potential features, so that an initial clustering result is obtained. Meanwhile, a self-optimization model is used to iteratively optimize the initial clustering result, so that a final clustering result is obtained, and the accuracy of the cell clustering result is improved.

The invention discloses a chronic kidney disease glycolysis-immunoregulation analysis method based on multiple omics, which comprises the following steps: acquiring GSE209781 single cell and GSE37171 transcriptome data, and analyzing glycolysis pathway activity and differential genes after preprocessing and cell clustering annotation; molecular subtypes are determined through consensus clustering and immune infiltration assessment; key diagnostic genes such as ETFB and the like are screened in combination with three types of machine learning; molecular docking verifies the binding potential of ETFB and therapeutic drugs. According to the invention, near-end renal tubular cells are positioned as a glycolysis core, an MIF-mediated regulatory network is disclosed, the diagnosis accuracy is improved, a clear treatment target is provided, a standardized analysis system is constructed, and accurate CKD diagnosis and treatment are assisted.

The present application relates to the technical field of data analysis, in particular to a multi-source single-cell transcriptome data integration analysis method and system. The method comprises the following steps: converting all gene names of different source single-cell transcriptome data into a unified version, filtering source-specific genes; constructing a specific background gene pool and identifying contaminant genes to eliminate background RNA contamination; calculating the average sequencing depth of each sample cell and correcting the sequencing depth of each sample cell; based on different source single-cell transcriptome data, cell type identification is carried out through cell clustering and characteristic gene expression analysis, and each cell type is obtained; for each cell type, an anchor cell is constructed, and the expression profile of the depth-balanced cell of the same cell type is corrected to obtain the integrated cell expression profile. The accuracy of single-cell transcriptome data integration analysis is improved, and the influence of batch effect caused by multi-source on subsequent other analysis is reduced.

1. The name of the design product: display screen panel of bioinformatics cloud platform graphical user interface with single cell data analysis. 2. The use of the design product: the design product is used to display the graphical user interface and interact with the user to run the program, and the display screen panel can be used for mobile phones, tablets, notebooks and desktop computers. 3. The design points of the design product: the graphical user interface content in the screen and the interface layout, the display screen is the existing design. 4. The picture or photo that best shows the design points: front view. 5. The design product is a flat product, and the rear view, left view, right view, top view and bottom view are omitted. 6. The use of the graphical user interface: for single cell data analysis, parameter adjustment and real-time interactive analysis of charts in the space-time tracing bioinformatics cloud platform. The front view is the login page of the cloud platform, which shows the user login account, password and verification code. At the same time, the registration function is provided on the page. After login, the interface change state diagram 1 appears, which shows the personal information and project information of the current user; click "file upload" in the second module "cell analysis" from "home page" to the bottom in the interface change state diagram 1, and the interface change state diagram 2 appears, a new project is created and data is uploaded; click the "start execution" button in the interface change state diagram 2 to appear the interface change state diagram 3, and perform data quality control; click "next step" in the interface change state diagram 3 to appear the interface change state diagram 4, and perform sample high variable gene screening; click "next step" in the interface change state diagram 4 to appear the interface change state diagram 5, and perform sample dimension reduction processing; click "next step" in the interface change state diagram 5 to appear the interface change state diagram 6, and perform sample cell clustering; click "next step" in the interface change state diagram 6 to appear the interface change state diagram 7, and perform differential gene analysis; click "next step" in the interface change state diagram 7 to appear the interface change state diagram 8, and perform cell annotation; click "next step" in the interface change state diagram 8 to appear the interface change state diagram 9, and perform enrichment analysis; click "next step" in the interface change state diagram 9 to appear the interface change state diagram 10, and perform pseudo-time sequence analysis.

The embodiment of the invention discloses a cell functionalization stage classification method, a cell functionalization stage determination method, a cell functionalization stage classification device, a cell functionalization stage determination device and electronic equipment. The classification method comprises the steps that a quasi-time sequence value of cells in a target cell cluster is acquired, the quasi-time sequence value is determined based on quasi-time sequence analysis, and the target cell cluster is a cell cluster obtained after cell clustering; determining continuous density distribution of the cells in the target cell cluster according to the quasi-time sequence value of the cells in the target cell cluster; according to the continuous density distribution and a fitting model corresponding to the shape of the continuous density distribution, fitting the continuous density distribution to obtain a fitting function; and determining a cell functionalization stage contained in the target cell cluster and a quasi-time sequence value range corresponding to the cell functionalization stage based on the fitting function. By adopting the embodiment of the invention, the cell functionalization stage can be determined without depending on a cell annotation result.

This invention discloses a clustering analysis method and application for high-dimensional single-cell data based on multi-view canonical correlation analysis, belonging to the interdisciplinary field of bioinformatics and computer science. Addressing the issue of insufficient accuracy in cell clustering analysis of high-dimensional single-cell RNA-seq data, this invention generates multi-view feature representations from single-cell sequencing data using various dimensionality reduction methods. A mapping method combining autoencoders and generalized canonical correlation analysis is used to map these multi-view feature representations to a common subspace. Then, the projection matrices of each view in the common subspace are weighted and fused to obtain a fused representation matrix. Finally, the fused representation matrix is ​​clustered using the KMeans method, and the clustering effect is evaluated by adjusting the Land index (ARI) and normalized mutual information (NMI). This invention can fully exploit the high-dimensional nonlinear structure of single-cell data, effectively compensate for the information loss in single-view methods, and significantly improve the accuracy and robustness of high-dimensional single-cell data clustering.

A spatial transcriptome cell clustering method, comprising the steps of: preprocessing gene expression of each cell in a spatial transcriptome; generating an adjacency matrix A according to the cell coordinates of the spatial transcriptome, obtaining a graph structure representation of the spatial transcriptome cells, representing the cell gene expression by a cell feature matrix X, inputting the adjacency matrix A and the cell feature matrix X into a trained graph convolutional neural network model DGI; the graph convolutional neural network model DGI outputs a node feature representation with spatial information; after dimension reduction and clustering algorithm processing on the node feature representation, the spatial transcriptome cell type is identified and obtained.

The application provides a single-cell rare cell type identification method and system based on a divide-and-conquer strategy, comprising the following steps: S1, using a variational autoencoder to encode single-cell transcriptome data into Gaussian distribution hidden variables; S2, performing spectral clustering on the Gaussian distribution hidden variables to obtain a single-cell coarse-grained clustering result; and S3, optimizing the single-cell coarse-grained clustering result based on intra-class cluster loss and inter-class cluster loss until the intra-class cluster loss and the inter-class cluster loss are minimized to obtain a single-cell clustering result containing rare cell types. The application can overcome the deficiency that previous single-cell clustering and identification methods do not optimize rare cell types, leading to misclassification or difficulty in identifying rare cell types.

This application discloses a method for determining developmental trajectories based on single-cell multi-omics clustering, belonging to the field of biomedical data mining technology. The method includes: acquiring multi-omics data of single cells from the same tissue; determining the potential representation of each single cell in different omics based on feature encoding technology; constructing a K-nearest neighbor graph for each omics based on the distance between single cells; and determining the corresponding multi-order similarity matrix; using the multi-order similarity matrix to complete the missing potential representations of single cells, obtaining the complete potential representation of each omics; performing cluster analysis on each omics to obtain single-cell clustering results; weighted fusion of the potential representations of each single cell in different omics to obtain a comprehensive potential representation; and analyzing the developmental trajectory of single cells based on the comprehensive potential representation. This application can stably and accurately cluster single cells under conditions of missing single-cell omics or significant differences in omics quality, thereby accurately determining the developmental trajectory of single cells.