Immunophenotyping method and apparatus, storage medium, and program product

Abstract

Provided are an immunophenotyping method and apparatus, a storage medium, and a program product. The method comprises: acquiring sample data, and preprocessing the acquired sample data; on the basis of the preprocessed sample data, calculating enrichment scores of multiple immune cells, and screening for N cells having statistical significance from among the multiple immune cells as alternative cells, wherein N is a natural number; by means of survival analysis, selecting n cells from among the N alternative cells as preferred cells, wherein n is a natural number and less than N, and using the remaining (N-n) cells as second-choice cells; circularly traversing multiple combinations of the second-choice cells, selecting one of the combinations by means of cluster analysis, and jointly using the selected combination and the preferred cells as an immunophenotyping cell combination.

Description

Immunophenotyping method and device, storage medium and program product Technical Field

[0001] The embodiments of the present disclosure relate to, but are not limited to, the field of biotechnology, and in particular to an immunophenotyping method and apparatus, a storage medium, and a program product. Background Art

[0002] Peripheral T cell lymphomas (PTCL) are a heterogeneous group of diseases that originate from postthymic T lymphocytes or mature NK cells. In Europe and the United States, PTCL accounts for approximately 10% to 15% of non-Hodgkin's lymphoma patients, while in China, PTCL accounts for approximately 21.4% of non-Hodgkin's lymphoma patients. The median age of onset is approximately 50-60 years, with a higher prevalence in men. Patients with PTCL often exhibit poor prognostic factors, with only 25% surviving for more than five years after diagnosis.

[0003] Cellular immunophenotyping is an important auxiliary diagnostic tool for lymphoid tumors and has significant value in predicting prognosis. However, the related immunophenotyping process uses an exhaustive method to iterate through all possible cell combinations, resulting in numerous loops, heavy analysis computations, and long runtimes.

[0004] Summary of the Invention

[0005] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0006] The present disclosure provides an immunophenotyping method, comprising:

[0007] Acquire sample data and preprocess the acquired sample data;

[0008] Calculate the enrichment scores of multiple immune cells based on the preprocessed sample data, and select N statistically significant cells from the multiple immune cells as candidate cells, where N is a natural number;

[0009] Selecting n cells from the N candidate cells as preferred cells through survival analysis, where n is a natural number and n<N, and the remaining (Nn) cells as secondary cells;

[0010] The multiple combinations of the secondary selected cells are traversed cyclically, and one of the combinations is selected through cluster analysis, and the selected combination and the primary cells are used together as the cell combination for immunophenotyping.

[0011] An embodiment of the present disclosure also provides an immunophenotyping device, comprising a memory; and a processor connected to the memory, wherein the memory is used to store instructions, and the processor is configured to execute the steps of the immunophenotyping method described in any embodiment of the present disclosure based on the instructions stored in the memory.

[0012] An embodiment of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the immunophenotyping method described in any embodiment of the present disclosure.

[0013] An embodiment of the present disclosure further provides a program product, comprising instructions. When the computer program product is executed by a computer, the instructions execute the immunophenotyping method as described in any embodiment of the present disclosure.

[0014] The present disclosure also provides an immunophenotyping device, comprising: a data acquisition module, a first selection module, a second selection module, and a third selection module, wherein:

[0015] The data acquisition module is configured to acquire sample data and preprocess the acquired sample data;

[0016] The first selection module is configured to calculate the enrichment scores of the plurality of immune cells based on the preprocessed sample data, and screen out N types of cells with statistical significance from the plurality of immune cells as candidate cells, where N is a natural number;

[0017] The second selection module is configured to select n cells from the N candidate cells as preferred cells through survival analysis, where n is a natural number and n<N, and the remaining (Nn) cells as secondary cells;

[0018] The third selection module is configured to traverse the multiple combinations of the secondary selected cells in a loop, select one of the combinations through cluster analysis, and use the selected combination and the primary cells together as the cell combination for immunophenotyping.

[0019] Other aspects will become apparent upon reading and understanding the drawings and detailed description.

[0020] Summary of the Figures

[0021] The accompanying drawings are intended to provide a further understanding of the technical solutions of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the technical solutions of the present disclosure and do not constitute a limitation of the technical solutions of the present disclosure. The shapes and sizes of the components in the drawings do not reflect the actual scale and are intended only to illustrate the contents of the present disclosure.

[0022] FIG1 is a schematic flow chart of an immunophenotyping method provided by an exemplary embodiment of the present disclosure;

[0023] FIG2 is a schematic flow chart of another immunophenotyping method provided by an exemplary embodiment of the present disclosure;

[0024] 3A to 3S are survival curves of 19 candidate cells provided by an exemplary embodiment of the present disclosure;

[0025] FIG4 is a risk ratio diagram of 19 candidate cells provided by an exemplary embodiment of the present disclosure;

[0026] FIG5 is a graph showing p-values ​​and silhouette coefficients corresponding to 32,752 cell combinations obtained by traversing 15 sub-selected cells in an exemplary embodiment of the present disclosure;

[0027] FIG6 is a survival curve diagram of the immunophenotyping cell combination finally selected according to an exemplary embodiment of the present disclosure;

[0028] FIG7 is a silhouette coefficient diagram of the finally selected immunophenotyping cell combination provided by an exemplary embodiment of the present disclosure;

[0029] FIG8 is a graph showing p-values ​​and silhouette coefficients corresponding to 16,369 cell combinations obtained by traversing 14 sub-selected cells in an exemplary embodiment of the present disclosure;

[0030] FIG9 is an immunophenotyping and cell type diagram provided by an exemplary embodiment of the present disclosure;

[0031] FIG10 is a schematic structural diagram of an immunophenotyping device provided by an exemplary embodiment of the present disclosure;

[0032] FIG11 is a schematic structural diagram of another immunophenotyping device provided by an exemplary embodiment of the present disclosure.

[0033] Details

[0034] To make the objectives, technical solutions and advantages of the present disclosure more clearly understood, the embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other in any manner.

[0035] Unless otherwise defined, the technical or scientific terms used in the embodiments of the present disclosure should have the ordinary meaning understood by people with ordinary skills in the field to which the present disclosure belongs. The words "first", "second" and similar words used in the embodiments of the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. The words "include" or "comprising" and similar words mean that the elements or objects preceding the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0036] As shown in FIG1 , the present disclosure provides an immunophenotyping method, comprising:

[0037] Step 101: Acquire sample data and preprocess the acquired sample data;

[0038] Step 102: Calculate the enrichment scores of multiple immune cells based on the preprocessed sample data, and select N cells with statistical significance from the multiple immune cells as candidate cells, where N is a natural number;

[0039] Step 103: select n cells from N candidate cells as first-selected cells through survival analysis, where n is a natural number and n<N, and select the remaining (Nn) cells as second-selected cells;

[0040] Step 104 , looping through multiple combinations of secondary selected cells, and selecting one combination through cluster analysis, and using the selected combination and the primary cells together as the cell combination for immunophenotyping.

[0041] The immunophenotyping method provided in the embodiments of the present disclosure selects n cells as the first choice cells from N candidate cells through survival analysis, and then traverses multiple combinations of the second choice cells in a loop. This can reduce the number of traversals, reduce the amount of analysis calculations, shorten the running time, and achieve better immunophenotyping results.

[0042] In some exemplary embodiments, in step 101 , the sample data obtained may be peripheral T-cell lymphoma patient data.

[0043] Lymphoma is mainly divided into two categories: non-Hodgkin lymphoma and Hodgkin lymphoma. Non-Hodgkin lymphoma includes B-cell-derived non-Hodgkin lymphoma, T-cell-derived non-Hodgkin lymphoma, and NK / T-cell-derived non-Hodgkin lymphoma. There are many subtypes of non-Hodgkin T-cell lymphoma, and patients with different subtypes have different clinical manifestations. The immunophenotyping method of the embodiment of the present disclosure can be used for T-cell-derived non-Hodgkin lymphoma, and can also be used for other lymphomas. The embodiment of the present disclosure does not limit this. The following is an introduction using peripheral T-cell lymphoma as an example.

[0044] In some exemplary embodiments, the acquired sample data includes bulk RNA-Seq expression data and clinical information.

[0045] In the disclosed embodiments, sample data can be downloaded from public databases. For example, bulk RNA-seq expression data and clinical information of patients with peripheral T-cell lymphoma can be obtained from the Gene Expression Omnibus (GEO). The GEO database is a gene expression database created and maintained by the National Center for Biotechnology Information (NCBI). Since its establishment in 2000, GEO has collected gene expression data from research institutions around the world, covering multiple fields such as tumors, non-tumors, chips, next-generation sequencing (NGS), differential analysis and molecular validation.

[0046] In some exemplary embodiments, in step 101, preprocessing the acquired sample data includes:

[0047] The number of datasets detecting transcriptome sequencing expression data and clinical information;

[0048] When the number of datasets of transcriptome sequencing expression data and / or the number of datasets of clinical information is greater than 1, a de-batch operation is performed on the obtained transcriptome sequencing expression data and / or clinical information.

[0049] In multi-omics data, batch effects refer to differences between samples due to variations in experimental conditions, operators, instruments, and other factors. These differences can mask true biological variations, thereby affecting subsequent analytical results. There are various approaches to address batch effects. One approach to de-batch is to use normalization techniques to normalize differences between samples and thereby eliminate batch effects.

[0050] In the disclosed embodiments, when the number of transcriptome sequencing expression data datasets (and / or the number of clinical information datasets) is equal to 1, there is no need to perform a de-batch operation on the transcriptome sequencing expression data (and / or the clinical information). For example, assuming that the number of transcriptome sequencing expression data datasets obtained for a patient with peripheral T-cell lymphoma is greater than 1, and the number of clinical information datasets is equal to 1, then only the transcriptome sequencing expression data needs to be de-batched, and the clinical information does not need to be de-batched.

[0051] Taking peripheral T-cell lymphoma as an example, as shown in FIG2 , an embodiment of the present disclosure provides an immunophenotyping method. First, a cell deconvolution algorithm is used to calculate the enrichment scores of multiple immune cells based on the gene expression data of bulk RNA-seq. Second, a univariate survival analysis is used to screen for clinically significant cell types (i.e., N candidate cell types with statistical significance). The selected candidate cell types are then analyzed by hazard ratio, survival curve, and cell correlation. Combined with the above analysis, a certain proportion of cell types (i.e., n preferred cells) is selected based on the characteristics of peripheral T-cell lymphoma. Finally, an exhaustive method is used to perform unsupervised immunophenotyping on the remaining cell types (i.e., (Nn) secondary cells). The cell combination for immunophenotyping is determined based on the clustering effect, p-value, and survival curve.

[0052] The immunophenotyping method of the disclosed embodiment is based on the characteristics of peripheral T-cell lymphoma. It uses a feature selection method to first select some clinically significant cell types, and then uses an exhaustive method to traverse the uncertain cell types. The effects of all uncertain cell combinations can be compared. The combination of feature selection and exhaustive method can obtain an immunophenotyping cell combination with better effect. Compared with the method of exhaustively traversing all cell types, the number of cycles can be reduced, the analysis calculation amount can be reduced, and the running time can be shortened.

[0053] In some exemplary embodiments, in step 102, the enrichment scores of the plurality of immune cells are calculated based on the preprocessed sample data, and N cells with statistical significance are screened from the plurality of immune cells as candidate cells, including:

[0054] Based on the pre-processed transcriptome sequencing expression data, the enrichment scores of various immune cells are calculated using a deconvolution algorithm;

[0055] Through univariate risk analysis, N statistically significant cells were screened from a variety of immune cells as candidate cells.

[0056] For example, the present disclosure uses the deconvolution algorithm of the xCell package, which is an immune cell infiltration assessment tool based on gene expression data. The xCell package can identify potential immune cell subsets and calculate their relative abundance in tissues. The xCell package can assess the infiltration levels of 64 cell types based on gene expression data. The deconvolution algorithm includes the following steps:

[0057] (1) Extracting expression features of 64 immune cells and stromal cells from bulk gene expression data as cell signatures;

[0058] (2) Single sample gene set enrichment analysis (ssGSEA) was used to calculate the enrichment score of the sample on each cell type signature;

[0059] (3) using the fitting formula to convert the enrichment scores of various cell types into corresponding cell type scores;

[0060] (4) Compensation correction is performed on the scores of closely related cell types to reduce the impact of possible collinearity / correlation between different cell types.

[0061] For example, before screening out N statistically significant cells as candidate cells from a variety of immune cells through univariate risk analysis, some cell types that are impossible to use as immunophenotyping can be removed from the variety of immune cells based on the clinical characteristics of the disease. At this time, the clinical characteristics of the disease are set according to the actual type of disease, and the present disclosure does not impose any restrictions on this.

[0062] Taking peripheral T-cell lymphoma as an example, according to the clinical characteristics of the disease, 24 cell types were removed, including Astrocytes, Basophils, Chondrocytes, CLP, CMP, Eosinophils, Epithelial cells, Erythrocytes, GMP, Hepatocytes, HSC, Keratinocytes, Megakaryocytes, Melanocytes, MEP, Mesangial cells, MPP, MSC, Myocytes, Neurons, Osteoblasts, Platelets, Skeletal muscle, and Smooth muscle. Then, for the remaining 40 cell types, those with statistically significant results (p value ≤ 0.05) were selected as candidate cells based on univariate risk analysis. As shown in Figures 3A to 3S, a total of 19 candidate cells were selected: naive B-cells, Plasma cells, Memory B-cells, Predipocytes, B-cells, Class-switched memory B-cells, CD8+T-cells, CD8+Tcm, Macrophages M1, Mast cells, Th2 cells, Adipocytes, Macrophages M2, iDC, DC, CD4+T-cells, NK cells, Tregs, Th1 cells.

[0063] In statistics, the p-value is a key concept used to test hypotheses. The p-value is defined as the probability of obtaining a result equal to or more extreme than the observed result, given the hypothesis. The p-value threshold is typically set at 0.05 or 0.01; a p-value below this threshold is considered statistically significant. For example, a p-value < 0.05 indicates a less than 5% chance that the difference in the results was due to chance, or a less than 5% chance that the same study, under the same conditions, would yield the opposite conclusion.

[0064] In some exemplary embodiments, in step 103, selecting n cells as preferred cells from N candidate cells through survival analysis includes:

[0065] For each candidate cell, perform the following operations:

[0066] Determine the difference between the high expression curve and the low expression curve in the survival curve graph of the candidate cell, and obtain the p value and HR value corresponding to the candidate cell;

[0067] When the difference between the high expression curve and the low expression curve is greater than or equal to the preset difference threshold, the p value is less than or equal to the preset p value threshold, and the absolute value of the HR value is greater than the preset HR value threshold, the candidate cell is selected as the preferred cell.

[0068] Still taking the aforementioned peripheral T-cell lymphoma as an example, Figures 3A to 3S are survival curves of 19 candidate cells, wherein the horizontal axis is the follow-up time (time, in days) and the vertical axis is the survival rate (Survival probability). Each survival curve includes a high expression curve, a low expression curve and a corresponding p-value, wherein high expression refers to a relatively high expression level in a biological sample, and low expression refers to a relatively low expression level in a biological sample. Generally speaking, the greater the difference between the high expression curve and the low expression curve, the greater the difference in the prognosis of the two groups of patients, and the more likely a statistical difference will occur. The smaller the p-value, the more statistically significant the result, that is, there is a significant difference in the survival time of the two groups of patients.

[0069] HR values, also known as hazard ratios, are primarily used in survival analysis. HR values ​​assess the multiple change in the risk of an outcome event caused by the presence of a factor. An HR value greater than 1 indicates that the exposure factor promotes the occurrence of a positive event, an HR value less than 1 indicates that the exposure factor hinders the occurrence of a positive event, and an HR equal to 1 indicates that the exposure factor has no effect on the occurrence of a positive event. Specifically, extremely large or small HR values ​​indicate a significant inhibitory or promoting effect on the occurrence of the event. Figure 4 shows the HR plot for these 19 candidate cells. The dots represent the model coefficients corresponding to each predictor variable, and the lines represent the 95% confidence intervals. The figure shows the lower bound (lower.95) and upper bound (upper.95) of the 95% confidence intervals for the model coefficients in exponential form, indicating the uncertainty range of the estimate.

[0070] Exemplarily, the difference between the high expression curve and the low expression curve can be determined as follows: the minimum value of the difference in the survival probability of the high expression curve and the low expression curve at multiple time points within a preset time period is determined as the difference between the high expression curve and the low expression curve. When the high expression curve and the low expression curve coincide or nearly coincide at a certain time point, the difference between the high expression curve and the low expression curve is small, and the difference in prognosis is also small; when the distance between the high expression curve and the low expression curve is large within a preset time period, the difference between the high expression curve and the low expression curve is large, and the difference in prognosis is also large. In the embodiment of the present disclosure, the preset difference threshold can be set as needed, and the embodiment of the present disclosure is not limited to this.

[0071] For example, the preset p-value threshold may be between 0.01 and 0.05, and the preset HR value threshold may be greater than 1.25 or less than 0.75. For example, the preset p-value threshold may be 0.05, and the preset HR value threshold may be greater than 1.25 or less than 0.75, however, the embodiments of the present disclosure are not limited thereto.

[0072] In the survival curve diagrams of Figures 3A to 3S, the difference between the two curves in the survival curve diagrams of the four cells, namely, Th1 cells, B-cells, Memory B-cells, and naive B-cells, is greater than the preset difference threshold, and the corresponding p-value is less than the preset p-value threshold, and in the HR diagram of Figure 4, the absolute values ​​of the HR values ​​of these four cells are greater than the preset HR value threshold. Therefore, the embodiment of the present disclosure selects the four cell types, namely, Th1 cells, B-cells, Memory B-cells, and naive B-cells, as the preferred cells.

[0073] In some exemplary embodiments, before looping through the plurality of combinations of sub-selected cells, the method further comprises:

[0074] Calculate the correlation between n preferred cells;

[0075] The number of subtypes to which the n preferred cells belong is determined based on the correlation between the n preferred cells;

[0076] According to the number of subtypes to which the n preferred cells belong and the total number of subtypes, all possible combinations of secondary selected cells in the immunophenotyping cell combination are determined.

[0077] In the disclosed embodiment, the correlation between the n preferred cells can be calculated using the Spearman correlation analysis method. The Spearman correlation coefficient ranges from -1 to 1, with a value of -1 indicating a perfect negative correlation, a value of 1 indicating a perfect positive correlation, and a value of 0 indicating no correlation between the two variables. Still using the aforementioned four preferred cells as an example, the correlation calculation results are shown in Table 1.

[0078] Table 1

[0079] In some exemplary embodiments, determining the number of subtypes to which the n preferred cells belong based on the correlation between the n preferred cells includes:

[0080] When the correlation between two or more preferred cells is greater than or equal to a preset correlation threshold, the two or more preferred cells are judged to belong to the same subtype;

[0081] When the correlation between two preferred cells is less than a preset correlation threshold, it is determined that the two preferred cells do not belong to the same subtype.

[0082] Exemplarily, the preset correlation threshold may be between 0 and 0.3. For example, the preset correlation threshold may be 0.3. However, the embodiment of the present disclosure is not limited thereto.

[0083] Table 1 shows that among these four preferred cells, the correlation between B-cells, memory B-cells and naive B-cells is greater than the preset correlation threshold, and the correlation between Th1 cells, B-cells, memory B-cells and naive B-cells is relatively small. Therefore, it can be judged that the number of subtypes to which these four preferred cells belong is 2.

[0084] In the embodiment of the present disclosure, the total number of subtypes may be 3 or 4. However, the embodiment of the present disclosure is not limited thereto, and the total number of subtypes may also be any other natural number greater than or equal to 2. The following example is exemplified by the total number of subtypes being 3.

[0085] In some exemplary embodiments, determining all possible combinations of secondary selected cells in the immunophenotyping cell combination based on the number of subtypes to which the n primary selected cells belong and the total number of subtypes comprises:

[0086] When the number of subtypes to which n preferred cells belong is equal to the total number of subtypes, the combination of all possible secondary cells is (The first type is that no secondary selected cells are selected in the immunophenotyping cell combination).

[0087] When the number of subtypes to which n preferred cells belong is equal to (total number of subtypes - 1), the combination of all possible secondary cells is kind.

[0088] When the number of subtypes to which n preferred cells belong is equal to (total number of subtypes - 2), the possible combinations of secondary cells are kind.

[0089] In the embodiment of the present disclosure, still taking the aforementioned peripheral T cell lymphoma as an example, excluding the four selected primary cells, 15 secondary cells remain: Plasma cells, Predipocytes, Class-switched memory B-cells, CD8+ T-cells, CD8+ Tcm, Macrophages M1, Mast cells, Th2 cells, Adipocytes, Macrophages M2, iDC, DC, CD4+ T-cells, NK cells, Tregs, that is, Nn=15. Since each subtype contains at least two cell types, all possible combinations of secondary cells include kind.

[0090] In some exemplary embodiments, in step 104 , looping through multiple combinations of sub-selected cells includes looping through all possible combinations of sub-selected cells.

[0091] In the disclosed embodiment, the 15 secondary selected cells can be directly cycled through all cell combinations, with the cycle starting from selecting 2 cells, for a total of 32,752 combinations.

[0092] In some exemplary embodiments, in step 104, selecting one of the combinations through cluster analysis includes:

[0093] According to the p-value, survival curve and clustering effect under each combination, one combination is selected from all the combinations, and the p-value, survival curve and clustering effect of the selected combination meet the following conditions: the p-value is less than or equal to the preset p-value threshold, the difference between the high expression curve and the low expression curve in the survival curve graph is greater than or equal to the preset difference threshold, and the silhouette coefficient is greater than or equal to the preset silhouette coefficient threshold.

[0094] In the disclosed embodiments, a constrained non-negative matrix factorization (CNMF) method can be used for cluster analysis. Non-negative matrix factorization (NMF) is a matrix decomposition method that makes all decomposed components non-negative and simultaneously achieves nonlinear dimensionality reduction. CNMF adds a hard constraint on label information to the original NMF method, so that data with the same class label information remains consistent in the new low-dimensional space.

[0095] Assume that the dataset X contains a total of c samples, of which the label information of the first l sample points is known, and the label information of the remaining nl sample points is unknown. For the first l labeled samples, define the indicator matrix C∈R l×c :

[0096] For all samples, define the sample constraint matrix A∈R n×(n+c-l) :

[0097] Where I∈R (n-l)×(n-l) is the identity matrix. The CNMF method introduces the auxiliary matrix Z∈R (n+c-l)×r The above sample constraint matrix A is embedded in the objective function so that samples belonging to the same class in V are mapped to the same point. Let V = AZ, that is, minimize the following objective function:

[0098] CNMF uses the alternating iteration method to solve the problem. The iteration rules are as follows:

[0099] The Silhouette Coefficient (Sketch Coefficient) is a metric used to evaluate clustering effectiveness. Essentially, it measures the ratio of the distance from each sample point to its own cluster to the distance between the nearest cluster structure. The Sketch Coefficient ranges from -1 to 1. A Sketch Coefficient closer to 1 indicates better clustering performance, while a Sketch Coefficient closer to -1 indicates poorer performance. The resulting Sketch Coefficient is the average of the Sketch Coefficients for each sample point.

[0100] Exemplarily, the preset silhouette coefficient threshold may be between 0.7 and 0.8. For example, the preset silhouette coefficient threshold may be 0.75. However, the embodiment of the present disclosure is not limited thereto.

[0101] Figure 5 shows the p-value and silhouette coefficient graphs corresponding to 32,752 cell combinations obtained by directly looping through the 15 secondary selected cells. As shown in Figure 5, in the embodiment of the present disclosure, the final selected cell combination is a combination of the following 11 cells: Preadipocytes, Class-switched memory B-cells, Mast cells, Adipocytes, Macrophages M2, NK cells, Tregs, Th1 cells, B-cells, Memory B-cells, and naive B-cells. Figure 6 shows the survival curve corresponding to the final selected cell combination. As shown in Figure 6, in the survival curve corresponding to this combination, the differences between the curves of each subtype are large, and the p-values ​​are also small. Figure 7 shows the silhouette graphs of each subtype corresponding to the final selected cell combination. As shown in Figure 7, the total number of samples is 136, the number of samples of subtype 1 is 83, and the average silhouette coefficient is 0.80; the number of samples of subtype 2 is 30, and the average silhouette coefficient is 0.94; the number of samples of subtype 3 is 23, and the average silhouette coefficient is 0.97. The overall average silhouette coefficient is 0.86.

[0102] In some exemplary embodiments, in step 104, looping through multiple combinations of sub-selected cells includes:

[0103] Exclude x types of cells from the secondary selected cells through correlation analysis, determine all possible cell combinations of the remaining (Nnx) types of secondary selected cells, and traverse all possible cell combinations of the remaining (Nnx) types of secondary selected cells in a loop.

[0104] In some exemplary embodiments, excluding x types of cells from the sub-selected cells by correlation analysis comprises:

[0105] When the number of subtypes to which n preferred cells belong is equal to the total number of subtypes, the correlation between each secondary cell and all the preferred cells is calculated, and x cells are excluded from the secondary cells based on the correlation between each secondary cell and all the preferred cells. The correlation between the excluded x cells and all the preferred cells is lower than the preset correlation threshold.

[0106] For example, assuming that the total number of subtypes is 3 and the number of subtypes to which the n preferred cells belong is 3, then the x cells whose correlation is lower than the preset correlation threshold can be directly excluded based on the correlation between each secondary cell and all the preferred cells.

[0107] In some exemplary embodiments, excluding x types of cells from the sub-selected cells by correlation analysis comprises:

[0108] When the number of subtypes to which the n preferred cells belong is less than the total number of subtypes, the correlation between each secondary-selected cell and all the primary cells is calculated, and the third-selected cells are determined from the secondary-selected cells. The third-selected cells are cells that may be included in the first subtype. The first subtype is different from the subtype to which the n preferred cells belong. The correlation between the third-selected cells and all the preferred cells is lower than the preset correlation threshold. The secondary-selected cells except the third-selected cells are used as the fourth-selected cells.

[0109] The correlations between the fourth-selected cells and the first-selected cells and between the fourth-selected cells and the third-selected cells are calculated, and x cells are excluded from the fourth-selected cells, wherein the correlations between the x cells and all the first-selected cells and between the x cells and all the third-selected cells are lower than a preset correlation threshold.

[0110] For example, still taking the aforementioned peripheral T-cell lymphoma as an example, assuming that the total number of subtypes is 3, and the number of subtypes to which the aforementioned four preferred cells belong is 2, the correlation between each secondary cell and all the preferred cells is calculated, and the third-selected cells (i.e., cell types that may be included in the first subtype) that have no correlation (or low correlation) with the aforementioned four preferred cells (Th1 cells, B-cells, Memory B-cells, naive B-cells) are determined: iDC, Adipocytes, DC, Macrophages M2.

[0111] Then, the correlation between the second-selected cells (i.e., fourth-selected cells) other than the third-selected cells and all the first-selected cells and the third-selected cells (i.e., 4+4=8 cell types: Th1 cells, B-cells, Memory B-cells, naive B-cells, iDCs, Adipocytes, DCs, Macrophages M2) was calculated, and cell types that had no correlation (or low correlation) with these 8 cell types were selected for exclusion.

[0112] In some exemplary embodiments, when selecting x types of cells to exclude, exclusion can be performed based on a combination of survival curves and p-values. The difference between the high expression curve and the low expression curve in the survival curve corresponding to the x types of cells to be excluded is lower than a preset difference threshold, and the p-value is greater than a preset p-value threshold. That is, when selecting x types of cell types to exclude, cell types that have no correlation (or low correlation) with any of the eight cell types described above, and for which the difference between the high expression curve and the low expression curve in the survival curve is lower than a preset difference threshold, and the p-value is greater than a preset p-value threshold, can be selected for exclusion.

[0113] In the embodiment of the present disclosure, still taking the aforementioned peripheral T-cell lymphoma as an example, one cell type is selected and removed based on the correlation, survival curve and p-value: Th2 cells. Then, the remaining 14 cell types (Plasma cells, Predipocytes, Class-switched memory B-cells, CD8+T-cells, CD8+Tcm, Macrophages M1, Mast cells, Adipocytes, Macrophages M2, iDC, DC, CD4+T-cells, NK cells, Tregs) are looped through all cell combinations. Since each category contains at least two cell types, the combination starts with selecting two cells. Then, the CNMF method was used for cluster analysis, and the obtained p-value and silhouette coefficient diagram are shown in Figure 8. Finally, a combination of 11 cells with better selection effect was obtained: Preadipocytes, Class-switched memory B-cells, Mast cells, Adipocytes, Macrophages M2, NK cells, Tregs, Th1 cells, B-cells, Memory B-cells, and naive B-cells. By excluding one cell type, the disclosed embodiment reduced the number of cycles by 16,383 times, and the final immunophenotyping cell combination was the same as the immunophenotyping cell combination obtained when Th2 cells were not excluded.

[0114] In the embodiment of the present disclosure, as shown in FIG9 , the following three categories are obtained through cluster analysis: IM1 includes Mast cells, Memory B-cells, naive B-cells, Class-switched memory B-cells, and B-cells; IM2 includes Adipocytes, Tregs, and Preadipocytes; and IM3 includes Macrophages M2, NK cells, and Th1 cells.

[0115] In some exemplary embodiments, the method further comprises:

[0116] Obtain a validation set and use the immunophenotyping cell combination and its differentially expressed genes to validate the validation set.

[0117] In the disclosed embodiment, after obtaining the immunophenotyping cell combination, the cell types and differentially expressed genes determined by immunophenotyping of the training set can be used to verify the validation set. The results show that the validation set results are consistent with the training set.

[0118] As shown in FIG10 , the embodiment of the present disclosure further provides an immunophenotyping device, comprising a data acquisition module 1010 , a first selection module 1020 , a second selection module 1030 , and a third selection module 1040 , wherein:

[0119] The data acquisition module 1010 is configured to acquire sample data and pre-process the acquired sample data;

[0120] The first selection module 1020 is configured to calculate the enrichment scores of multiple immune cells based on the preprocessed sample data, and screen out N types of cells with statistical significance from the multiple immune cells as candidate cells, where N is a natural number;

[0121] The second selection module 1030 is configured to select n cells from N candidate cells as the first choice cells through survival analysis, where n is a natural number and n<N, and the remaining (Nn) cells as the second choice cells;

[0122] The third selection module 1040 is configured to traverse multiple combinations of secondary selected cells in a loop, select one of the combinations through cluster analysis, and use the selected combination and the primary cells together as the cell combination for immunophenotyping.

[0123] In some exemplary embodiments, the acquired sample data includes transcriptome sequencing expression data and clinical information; the data acquisition module 1010 preprocesses the acquired sample data, including:

[0124] The number of datasets detecting transcriptome sequencing expression data and clinical information;

[0125] When the number of datasets of transcriptome sequencing expression data and / or the number of datasets of clinical information is greater than 1, a de-batch operation is performed on the obtained transcriptome sequencing expression data and / or clinical information.

[0126] In some exemplary embodiments, the first selection module 1020 calculates the enrichment scores of multiple immune cells based on the preprocessed sample data, and selects N cells with statistical significance from the multiple immune cells as candidate cells, including:

[0127] Based on the pre-processed transcriptome sequencing expression data, the enrichment scores of various immune cells are calculated using a deconvolution algorithm;

[0128] Through univariate risk analysis, N statistically significant cells were screened from a variety of immune cells as candidate cells.

[0129] In some exemplary embodiments, the second selection module 1030 selects n cells as preferred cells from the N candidate cells through survival analysis, including:

[0130] For each candidate cell, perform the following operations:

[0131] Determine the difference between the high expression curve and the low expression curve in the survival curve graph of the candidate cells, and obtain the p-value and risk ratio corresponding to the candidate cells;

[0132] When the difference between the high expression curve and the low expression curve is greater than or equal to a preset difference threshold, the p value is less than or equal to a preset p value threshold, and the absolute value of the risk ratio is greater than a preset HR value threshold, the candidate cell is selected as the preferred cell.

[0133] In some exemplary embodiments, the third selection module 1040 is further configured to:

[0134] Calculating the correlation between n preferred cells;

[0135] Determining the number of subtypes to which the n preferred cells belong based on the correlations among the n preferred cells;

[0136] According to the number of subtypes to which the n preferred cells belong and the total number of subtypes, all possible combinations of secondary selected cells in the immunophenotyping cell combination are determined.

[0137] In some exemplary embodiments, the third selection module 1040 loops through multiple combinations of sub-selected cells, including looping through all possible determined combinations of sub-selected cells.

[0138] In some exemplary embodiments, the third selection module 1040 loops through multiple combinations of sub-selected cells, including:

[0139] Exclude x types of cells from the secondary selected cells through correlation analysis, determine all possible cell combinations for the remaining (Nnx) types of secondary selected cells, and traverse all possible cell combinations of the remaining (Nnx) types of secondary selected cells in a loop.

[0140] In some exemplary embodiments, the third selection module 1040 excludes x types of cells from the secondary selected cells through correlation analysis, including:

[0141] When the number of subtypes to which the n cells belong is less than the total number of subtypes, the correlation between each secondary-selected cell and all the primary cells is calculated, and a third-selected cell is determined from the secondary-selected cells. The third-selected cell is a cell that may be included in the first subtype, and the first subtype is different from the subtype to which the n primary cells belong. The correlation between the third-selected cell and all the primary cells is lower than a preset correlation threshold, and the secondary-selected cells other than the third-selected cell are used as the fourth-selected cell.

[0142] Calculate the correlation between the fourth-selected cells and the first-selected cells, and between the fourth-selected cells and the third-selected cells, and exclude x types of cells from the fourth-selected cells, wherein the correlation between the x types of cells and all the first-selected cells, and between the x types of cells and all the third-selected cells, is lower than a preset correlation threshold.

[0143] In some exemplary embodiments, the third selection module 1040 excludes x types of cells from the secondary selected cells through correlation analysis, including:

[0144] When the number of subtypes to which n cells belong is equal to the total number of subtypes, the correlation between each secondary selected cell and all the primary selected cells is calculated, and x cells are excluded from the secondary selected cells, and the correlation between the x cells and all the primary selected cells is lower than a preset correlation threshold.

[0145] In some exemplary embodiments, the difference between the high expression curve and the low expression curve in the survival curve graph corresponding to the x types of cells is lower than a preset difference threshold, and the p-value corresponding to the x types of cells is greater than a preset p-value threshold.

[0146] In some exemplary embodiments, the third selection module 1040 selects one of the combinations through cluster analysis, including:

[0147] According to the p-value, survival curve and clustering effect under each combination, one combination is selected from all the combinations, and the p-value, survival curve and clustering effect of the selected combination meet the following conditions: the p-value is less than or equal to the preset p-value threshold, the difference between the high expression curve and the low expression curve in the survival curve graph is greater than or equal to the preset difference threshold, and the silhouette coefficient is greater than or equal to the preset silhouette coefficient threshold.

[0148] In some exemplary embodiments, the immunophenotyping apparatus further comprises: a verification module configured to obtain a verification set and verify the verification set using the immunophenotyped cell combinations and their differential genes.

[0149] An embodiment of the present disclosure also provides an immunophenotyping device, comprising a memory; and a processor connected to the memory, wherein the memory is used to store instructions, and the processor is configured to execute the steps of the immunophenotyping method described in any embodiment of the present disclosure based on the instructions stored in the memory.

[0150] As shown in Figure 11, in one example, the immunophenotyping device may include: a processor 1110, a memory 1120, a bus system 1130 and a transceiver 1140, wherein the processor 1110, the memory 1120 and the transceiver 1140 are connected via the bus system 1130, the memory 1120 is used to store instructions, and the processor 1110 is used to execute the instructions stored in the memory 1120 to control the transceiver 1140 to send and receive signals. Specifically, the transceiver 1140 can obtain sample data under the control of the processor 1110, and the processor 1110 preprocesses the obtained sample data; based on the preprocessed sample data, the enrichment scores of multiple immune cells are calculated, and N cells with statistical significance are screened out from the multiple immune cells as candidate cells, where N is a natural number; n cells are selected from the N candidate cells as preferred cells through survival analysis, where n is a natural number and n<N, and the remaining (Nn) cells are used as secondary cells; multiple combinations of the secondary cells are traversed in a loop, and one of the combinations is selected through cluster analysis, and the selected combination and the preferred cells are used together as the cell combination for immunophenotyping.

[0151] It should be understood that the processor 1110 may be a central processing unit (CPU), or may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0152] The memory 1120 may include a read-only memory and a random access memory, and provides instructions and data to the processor 1110. A portion of the memory 1120 may also include a non-volatile random access memory. For example, the memory 1120 may also store information about the device type.

[0153] In addition to the data bus, the bus system 1130 may also include a power bus, a control bus, a status signal bus, etc. However, for the sake of clarity, various buses are labeled as the bus system 1130 in FIG.

[0154] During the implementation process, the processing performed by the processing device can be completed by the hardware integrated logic circuit in the processor 1110 or the instructions in the form of software. That is, the method steps of the embodiment of the present disclosure can be embodied as being executed by a hardware processor, or being executed by a combination of hardware and software modules in the processor. The software module can be located in a storage medium such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, registers, etc. The storage medium is located in the memory 1120, and the processor 1110 reads the information in the memory 1120 and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0155] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the immunophenotyping method described in any embodiment of the present disclosure. The immunophenotyping method driven by executing the executable instructions is substantially the same as the immunophenotyping method provided in the above embodiments of the present disclosure and is not further described here.

[0156] In some possible embodiments, various aspects of the immunophenotyping method provided by the present disclosure may also be implemented in the form of a program product, which includes a program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of the immunophenotyping method according to various exemplary embodiments of the present disclosure described above in this specification. For example, the computer device can execute the immunophenotyping method described in the embodiments of the present disclosure.

[0157] The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0158] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable, and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media generally embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0159] It should be noted that the above-described embodiments or implementations are merely illustrative and not restrictive. Therefore, the present disclosure is not limited to what is specifically shown and described herein. Various modifications, substitutions, or omissions may be made to the forms and details of the implementations without departing from the scope of the present disclosure.

Claims

1. An immunophenotyping method comprising: Acquire sample data and preprocess the acquired sample data; Calculate the enrichment scores of multiple immune cells based on the preprocessed sample data, and select N statistically significant cells from the multiple immune cells as candidate cells, where N is a natural number; Selecting n cells from the N candidate cells as preferred cells through survival analysis, where n is a natural number and n<N, and the remaining (Nn) cells as secondary cells; The multiple combinations of the secondary selected cells are traversed cyclically, and one of the combinations is selected through cluster analysis, and the selected combination and the primary cells are used together as the cell combination for immunophenotyping.

2. The method according to claim 1, before looping through the plurality of combinations of the sub-selected cells, further comprising: Calculating the correlation between n preferred cells; Determining the number of subtypes to which the n preferred cells belong based on the correlations among the n preferred cells; According to the number of subtypes to which the n preferred cells belong and the total number of subtypes, all possible combinations of secondary selected cells in the immunophenotyping cell combination are determined.

3. The method according to claim 2, wherein: The looping through the multiple combinations of the secondary selected cells includes: looping through all possible determined combinations of the secondary selected cells.

4. The method according to claim 2, wherein: The looping through the multiple combinations of the secondary selected cells includes: Exclude x types of cells from the secondary selected cells through correlation analysis, determine all possible cell combinations for the remaining (Nnx) types of secondary selected cells, and perform a loop traversal on all possible cell combinations of the remaining (Nnx) types of secondary selected cells.

5. The method according to claim 4, wherein The step of excluding x types of cells from the secondary selected cells by correlation analysis comprises: When the number of subtypes to which the n cells belong is less than the total number of subtypes, the correlation between each secondary-selected cell and all the primary cells is calculated, and a third-selected cell is determined from the secondary-selected cells. The third-selected cell is a cell that may be included in the first subtype, and the first subtype is different from the subtype to which the n primary cells belong. The correlation between the third-selected cell and all the primary cells is lower than a preset correlation threshold, and the secondary-selected cells other than the third-selected cell are used as the fourth-selected cell. Calculate the correlation between the fourth-selected cells and the first-selected cells, and between the fourth-selected cells and the third-selected cells, and exclude x types of cells from the fourth-selected cells, wherein the correlation between the x types of cells and all the first-selected cells, and between the x types of cells and all the third-selected cells, is lower than a preset correlation threshold.

6. The method according to claim 4, wherein: The step of excluding x types of cells from the secondary selected cells by correlation analysis comprises: When the number of subtypes to which n cells belong is equal to the total number of subtypes, the correlation between each secondary selected cell and all the primary cells is calculated, and x cells are excluded from the secondary selected cells. The correlation between the x cells and all the primary cells is The correlations are all below the preset threshold.

7. The method according to claim 5 or 6, wherein: The difference between the high expression curve and the low expression curve in the survival curve graph corresponding to the x types of cells is lower than a preset difference threshold, and the p-value corresponding to the x types of cells is greater than a preset p-value threshold.

8. The method according to claim 1, wherein The step of selecting one of the combinations through cluster analysis includes: According to the p-value, survival curve and clustering effect under each combination, one combination is selected from all the combinations, and the p-value, survival curve and clustering effect of the selected combination meet the following conditions: the p-value is less than or equal to the preset p-value threshold, the difference between the high expression curve and the low expression curve in the survival curve graph is greater than or equal to the preset difference threshold, and the silhouette coefficient is greater than or equal to the preset silhouette coefficient threshold.

9. The method according to claim 1, wherein The step of selecting n cells as preferred cells from the N candidate cells through survival analysis comprises: For each candidate cell, perform the following operations: Determine the difference between the high expression curve and the low expression curve in the survival curve graph of the candidate cells, and obtain the p-value and risk ratio corresponding to the candidate cells; When the difference between the high expression curve and the low expression curve is greater than or equal to a preset difference threshold, the p value is less than or equal to a preset p value threshold, and the absolute value of the risk ratio is greater than a preset HR value threshold, the candidate cell is selected as the preferred cell.

10. The method according to claim 1, wherein The acquired sample data includes transcriptome sequencing expression data and clinical information; the preprocessing of the acquired sample data includes: The number of datasets detecting transcriptome sequencing expression data and clinical information; When the number of datasets of transcriptome sequencing expression data and / or the number of datasets of clinical information is greater than 1, a de-batch operation is performed on the obtained transcriptome sequencing expression data and / or clinical information.

11. The method according to claim 10, wherein: The method calculates the enrichment scores of multiple immune cells based on the preprocessed sample data, and screens N cells with statistical significance from the multiple immune cells as candidate cells, including: Based on the pre-processed transcriptome sequencing expression data, the enrichment scores of various immune cells are calculated using a deconvolution algorithm; Through univariate risk analysis, N statistically significant cells were screened from a variety of immune cells as candidate cells.

12. The method according to claim 1, further comprising: Obtain a validation set and use the immunophenotyping cell combination and its differentially expressed genes to validate the validation set.

13. An immunophenotyping device comprising a memory; and a processor connected to the memory, the memory being used to store instructions, the processor being configured to execute the steps of the immunophenotyping method according to any one of claims 1 to 12 based on the instructions stored in the memory.

14. A computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the immunophenotyping method according to any one of claims 1 to 12 is implemented.

15. A computer program product comprising instructions, which, when executed by a computer, execute the immunophenotyping method according to any one of claims 1 to 12.

16. An immunophenotyping device, comprising a data acquisition module, a first selection module, a second selection module, and a third selection module, wherein: The data acquisition module is configured to acquire sample data and preprocess the acquired sample data; The first selection module is configured to calculate the enrichment scores of multiple immune cells based on the preprocessed sample data, and screen out N types of cells with statistical significance from the multiple immune cells as candidate cells, where N is a natural number; The second selection module is configured to select n cells from the N candidate cells as preferred cells through survival analysis, where n is a natural number and n<N, and the remaining (Nn) cells as secondary cells; The third selection module is configured to traverse the multiple combinations of the secondary selected cells in a loop, select one of the combinations through cluster analysis, and use the selected combination and the primary cells together as the cell combination for immunophenotyping.

Images