New crown growing typing method and device, molecular subtype and clinical variable association method and device and electronic equipment

Abstract

The invention relates to a new crown growing typing method and device, a molecular subtype and clinical variable association method and device and electronic equipment, and relates to the technical field of new crown growing typing. The typing method comprises the following steps: acquiring a gene expression matrix of a to-be-detected sample; and processing the gene expression matrix of a to-be-detected sample by adopting the new crown growing typing model to obtain a typing result of the to-be-detected sample. The association method comprises the following steps: acquiring a gene expression matrix and a typing result of an experimental sample and a gene expression matrix of a control sample; respectively carrying out differential gene analysis on the experimental samples of each molecular subtype and a control sample to obtain significant differential genes; performing functional pathway enrichment analysis, and calculating a gene set expression average value of the functional pathways with significant difference; and carrying out correlation analysis after normalizing the gene set expression average value and clinical variable data to obtain clinical variables associated with each molecular subtype.

Description

Technical Field

[0001] The present disclosure relates to the technical field of long-COVID-19 typing, and specifically, to a method and device for long-COVID-19 typing, a method and device for associating long-COVID-19 molecular subtypes with clinical variables, and electronic equipment. Background Art

[0002] Long COVID-19 refers to the situation that after the patient is infected with the new coronavirus, that is, after recovery and 3 months after the new coronavirus antigen / nucleic acid test is negative, the sick individual still continues to have various physical, psychological and other discomfort symptoms. These symptoms are not limited to fatigue, chest tightness, difficulty breathing, chest pain, headache, olfactory disorder, taste disorder, etc., and last for more than 2 months, which may lead to a decline in the patient's quality of life, increased psychological pressure, and a heavy economic burden brought by long-term care and treatment.

[0003] With the development of high-throughput sequencing technology, transcriptome sequencing (RNA-seq) technology has been widely used in many research fields. The second high-throughput sequencing technology can be used for cDNA detection to comprehensively and quickly detect the existence and quantity of RNA that can be transcribed by the genome at a given moment, providing a new perspective for life science research.

[0004] The patent with application number CN202211309998.4 discloses a probiotic composition and its use method for treating long-term new crown symptoms or reducing the risk of developing these symptoms. The patent with application number CN202180050015.7 discloses a composition that adjusts / regulates ncRNAs, lymphocytes, neutrophils, inflammasomes and p53 genes to prevent, treat and manage immune regulation abnormalities, inflammatory regulation abnormalities and energy regulation abnormalities caused by respiratory viruses (COVID19, influenza, etc.) to improve long-term new crown symptoms. However, the schemes disclosed in the above two patents do not clearly classify long-term new crown patients, so it is impossible to adopt more targeted treatment schemes according to the different pathological mechanisms of long-term new crowns of different types, so as to improve the prognosis of patients with various types of long-term new crowns and improve the quality of life of patients.

[0005] There is a lack of research on subgroup classification of long-COVID patients in related technologies, and it is difficult to achieve precise treatment of long-COVID patients by subdividing long-COVID subgroups. Considering that patients are often repeatedly infected with the new coronavirus, and because the new coronavirus mutates rapidly and spreads widely, patients often experience infection with different epidemic strains, and their vaccination histories vary greatly, it is difficult to clarify the symptoms of long-COVID and its duration. It has not been possible to successfully further subdivide long-COVID patients with large individual heterogeneity, making it difficult to achieve precise treatment of long-COVID patients.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] The purpose of the present disclosure is to overcome the deficiencies of the above-mentioned prior art, and to provide a method and device for typing long COVID-19, a method and device for associating long COVID-19 molecular subtypes with clinical variables, and electronic equipment to achieve typing of long COVID-19, and by associating the molecular subtypes of long COVID-19 with clinical variables, it is beneficial to optimize the treatment plan for patients with long COVID-19 of each molecular typing.

[0008] According to one aspect of the present disclosure, a long-term new crown typing method is provided, the method comprising:

[0009] Obtaining the gene expression matrix of the sample to be tested;

[0010] The long-novel coronavirus typing model is used to process the gene expression matrix of the sample to be tested to obtain the typing results of the sample to be tested.

[0011] In an exemplary embodiment of the present disclosure, the long new crown typing model is established using the following steps:

[0012] S10, obtaining gene expression matrices corresponding to a plurality of experimental samples respectively; the experimental samples are taken from patients with COVID-19;

[0013] S20, using an unsupervised clustering algorithm to analyze the gene expression matrix of the experimental samples to obtain the molecular subtype typing results corresponding to each experimental sample;

[0014] S30. Use the data set to train the random forest model to obtain a long-COVID typing model; wherein the data set includes the gene expression matrix corresponding to each experimental sample, and the molecular subtype typing results corresponding to each experimental sample.

[0015] In an exemplary embodiment of the present disclosure, in step S20, the unsupervised clustering algorithm is a hierarchical clustering algorithm; and the molecular subtyping results include 2 molecular subtypes.

[0016] In an exemplary embodiment of the present disclosure, obtaining a gene expression matrix of a sample to be tested includes:

[0017] Performing transcriptome sequencing on the sample to be tested to obtain transcriptome sequencing results of the sample to be tested;

[0018] Filtering the unexpressed genes in the transcriptome sequencing results of the sample to be tested to obtain the gene expression matrix of the sample to be tested;

[0019] Obtaining the gene expression matrices corresponding to multiple experimental samples includes:

[0020] Performing transcriptome sequencing on the experimental samples to obtain transcriptome sequencing results of the experimental samples;

[0021] The unexpressed genes in the transcriptome sequencing results of the experimental samples were filtered out to obtain the gene expression matrix corresponding to the experimental samples.

[0022] According to a second aspect of the present disclosure, a method for associating long-term novel coronavirus molecular subtypes with clinical variables is provided, the method comprising the following steps:

[0023] S100, obtaining gene expression matrices corresponding to a plurality of experimental samples, molecular subtyping results corresponding to the experimental samples, and gene expression matrices corresponding to a plurality of control samples; the experimental samples are taken from patients with long-term COVID-19, and the control samples are taken from patients without long-term COVID-19;

[0024] S200, according to the gene expression matrix corresponding to the experimental sample in step S100, the gene expression matrix corresponding to the control sample, and the molecular subtype typing result corresponding to the experimental sample, performing differential gene analysis on the experimental samples of different molecular subtypes and the control sample, respectively, to obtain significantly differential genes;

[0025] S300, performing functional pathway enrichment analysis on the significantly different genes to obtain the functional pathways with significant differences corresponding to each molecular subtype; calculating the average expression value of the gene set corresponding to the functional pathways with significant differences in each molecular subtype;

[0026] S400, after normalizing the average expression values ​​of each gene set and the clinical variables included in the association analysis, Spearman correlation analysis was used to determine the clinical variables associated with each molecular subtype.

[0027] In an exemplary embodiment of the present disclosure, in step S400, the clinical variables included in the association analysis include one or more of the white blood cell count, neutrophil count, platelet count, ferritin concentration, procalcitonin concentration, anxiety scale score, depression scale score, post-traumatic stress disorder scale score, and fatigue scale score.

[0028] According to a third aspect of the present disclosure, a long new crown typing device is provided, the device comprising:

[0029] An acquisition module is used to obtain the gene expression matrix of the sample to be tested;

[0030] The processing module is used to receive the gene expression matrix of the sample to be tested transmitted by the acquisition module, and use the long new crown typing model to process the gene expression matrix of the sample to be tested to obtain the typing result of the sample to be tested.

[0031] In an exemplary embodiment of the present disclosure, it also includes a generation module for generating a long-COVID typing model;

[0032] The generation module comprises:

[0033] A data acquisition unit, used to obtain gene expression matrices corresponding to a plurality of experimental samples; the experimental samples are taken from patients with COVID-19;

[0034] The typing unit is used to analyze the gene expression matrix of the experimental samples using an unsupervised clustering algorithm to obtain the molecular subtype typing results corresponding to each experimental sample;

[0035] A training unit is used to train the random forest model using a data set to obtain a long-coronavirus typing model; wherein the data set includes a gene expression matrix corresponding to each experimental sample and a molecular subtyping result corresponding to each experimental sample.

[0036] According to a fourth aspect of the present disclosure, a device for associating long-term novel coronavirus molecular subtypes with clinical variables is provided, the device comprising:

[0037] A data receiving module is used to obtain gene expression matrices corresponding to multiple experimental samples, molecular subtyping results corresponding to each experimental sample, and gene expression matrices corresponding to multiple control samples; wherein the experimental samples are taken from patients with long-term COVID-19, and the control samples are taken from patients without long-term COVID-19;

[0038] The differential gene analysis module is used to perform differential gene analysis on the experimental samples of different molecular subtypes and the control samples according to the gene expression matrix corresponding to the experimental samples, the gene expression matrix corresponding to the control samples, and the molecular subtype typing results corresponding to the experimental samples, to obtain significantly differential genes;

[0039] Functional pathway analysis module, used to perform functional pathway enrichment analysis on significantly different genes, obtain functional pathways with significant differences corresponding to each molecular subtype, and calculate the average expression of gene sets corresponding to functional pathways with significant differences in each molecular subtype;

[0040] The association analysis module is used to normalize the expression mean values ​​of each gene set and the clinical variables included in the association analysis, and then use Spearman correlation analysis to obtain the clinical variables associated with each molecular subtype.

[0041] According to the fifth aspect of the present disclosure, an electronic device is provided, comprising the above-mentioned long new coronavirus typing device and / or the above-mentioned device for associating long new coronavirus molecular subtypes with clinical variables.

[0042] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0044] Figure 1 This is a schematic flow chart of a long-COVID-19 typing method in one embodiment of the present disclosure.

[0045] Figure 2 The present invention is a schematic diagram of the construction process of the long-term new crown typing model in one embodiment of the present invention.

[0046] Figure 3 This is a schematic flow chart of a method for associating long-term novel coronavirus molecular subtypes with clinical variables in one embodiment of the present disclosure.

[0047] Figure 4 This is a schematic diagram of the hierarchical clustering classification results of the experimental group in one embodiment of the present disclosure.

[0048] Figure 5 This is a schematic diagram of the ROC curve of the long-term new crown typing model verification results in one embodiment of the present invention.

[0049] Figure 6 This is a schematic diagram of the enrichment pathway of the first molecular subtype of long-term COVID-19 patients in one embodiment of the present disclosure.

[0050] Figure 7 This is a schematic diagram of the enrichment pathway of the second molecular subtype of long-term COVID-19 patients in one embodiment of the present disclosure.

[0051] Figure 8 This is a schematic diagram of the correlation analysis results between Neu_score and clinical variables of patients with the first molecular subtype of long-term COVID-19 in one embodiment of the present disclosure.

[0052] Fig. 9 This is a schematic diagram of the correlation analysis results between Neu_score and clinical variables of patients with the second molecular subtype of long-term COVID-19 in one embodiment of the present disclosure. DETAILED DESCRIPTION

[0053] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be comprehensive and complete and will fully convey the concept of the example embodiments to those skilled in the art.

[0054] Unless otherwise indicated, the meaning of all terms (including technical and scientific terms) used to disclose the present invention is the same as that commonly understood by those of ordinary skill in the art to which the present invention belongs. By way of further guidance, the following definitions are used to better understand the teachings of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0055] The persistent long-term COVID-19 symptoms after COVID-19 infection will have a huge impact on the patient's physical and mental health and quality of life, and bring heavy medical pressure. However, the infection strains, infection times, and vaccination times of long-term COVID-19 patients are different, which makes long-term COVID-19 patients more heterogeneous, making it difficult to formulate treatment plans and having a poor prognosis.

[0056] To solve the above problems, the present disclosure provides a method for typing long coronavirus, such as Figure 1 As shown, the method includes: first, obtaining the gene expression matrix of the sample to be tested. Then, the gene expression matrix of the sample to be tested is processed using the long new crown typing model to obtain the typing result of the sample to be tested. It can be understood that the sample to be tested is a sample of a long new crown patient, and the sample to be tested can be but not limited to a peripheral blood sample. In this way, the typing result of the long new crown patient can be obtained by inputting the gene expression matrix of the long new crown patient into the long new crown typing model, which shortens the detection cycle, and helps to understand the pathological mechanism of each subtype based on the typing results of the long new crown patient obtained, and more reasonably guide the medication, so as to accurately improve the symptoms of each subtype of long new crown patients and achieve a good prognosis.

[0057] In one example, obtaining the gene expression matrix of the sample to be tested can be to perform transcriptome sequencing on the sample to be tested to obtain the transcriptome sequencing results of the sample to be tested. Then, the non-expressed genes in the transcriptome sequencing results of the sample to be tested are filtered to obtain the gene expression matrix of the sample to be tested. For example, the peripheral blood sample of the patient with long-term COVID-19 to be tested is subjected to transcriptome sequencing, and the non-expressed genes in the transcriptome sequencing results are filtered to obtain the corresponding gene expression matrix.

[0058] In one embodiment of the present disclosure, Figure 2 As shown, the long new crown typing model is established by the following steps: S10, obtaining the gene expression matrices corresponding to multiple experimental samples; the experimental samples are taken from long new crown patients. Exemplarily, the following method can be used to obtain the gene expression matrices corresponding to multiple experimental samples: perform transcriptome sequencing on the experimental samples to obtain the transcriptome sequencing results of the experimental samples; filter out the unexpressed genes in the transcriptome sequencing results of the experimental samples to obtain the gene expression matrix corresponding to the experimental samples.

[0059] S20, using an unsupervised clustering algorithm to analyze the gene expression matrix of the experimental samples to obtain the molecular subtype typing results corresponding to each experimental sample, wherein the data in the gene expression matrix is ​​the gene expression data of the corresponding sample.

[0060] S30. Use the data set to train the random forest model to obtain a long-COVID typing model; wherein the data set includes the gene expression matrix corresponding to each experimental sample, and the molecular subtype typing results corresponding to each experimental sample.

[0061] In this embodiment, the molecular subtypes corresponding to each long new crown patient are obtained by clustering the gene expression data in the gene expression matrix of multiple long new crown patients. In this way, the gene expression matrix of a long new crown patient and its corresponding molecular subtyping results are a data set, and multiple data sets corresponding to multiple long new crown patients constitute a data set. The random forest model is trained using this data set to obtain a long new crown typing model.

[0062] Exemplarily, a portion of the data set in the data set (for example, 70% to 90% of the data set) can be used as a training data set, and another portion of the data set in the data set (for example, 10% to 30% of the data set) can be used as a verification data set. The training data set is used to train the random Senli model to obtain a long new crown typing model. The verification data set can be used to verify the trained long new crown typing model, that is, to verify the accuracy of the long new crown typing model using a verification data set with known results.

[0063] In one embodiment of the present disclosure, the unsupervised clustering algorithm may be, but is not limited to, a hierarchical clustering algorithm. Figure 4 As shown, the molecular subtype typing results obtained by using the hierarchical clustering algorithm are 2 molecular subtypes, including a first molecular subtype and a second molecular subtype.

[0064] The present disclosure also provides a long new crown typing device, which includes an acquisition module and a processing module. Among them, the acquisition module is used to obtain the gene expression matrix of the sample to be tested. The processing module is used to receive the gene expression matrix of the sample to be tested transmitted by the acquisition module, and use the long new crown typing model to process the gene expression matrix of the sample to be tested to obtain the typing result of the sample to be tested.

[0065] In one embodiment of the present disclosure, the long new crown typing device also includes a generation module, which is used to generate a long new crown typing model. The generation module includes a data acquisition unit, a typing unit and a training unit. The data acquisition unit is used to obtain the gene expression matrices corresponding to multiple experimental samples respectively; the experimental samples are taken from long new crown patients. The typing unit is used to analyze the gene expression matrix of the experimental sample using an unsupervised clustering algorithm to obtain the molecular subtyping results corresponding to each experimental sample. The training unit is used to train the random forest model using a data set to obtain a long new crown typing model; wherein the data set includes the gene expression matrix corresponding to each experimental sample, and the molecular subtyping results corresponding to each experimental sample.

[0066] The disclosed embodiments also provide a method for associating long-term new crown molecular subtypes with clinical variables, such as Figure 3 As shown, the following steps are included:

[0067] S100. Obtain gene expression matrices corresponding to multiple experimental samples, molecular subtyping results corresponding to the experimental samples, and gene expression matrices corresponding to multiple control samples; the experimental samples are taken from patients with long-term COVID-19, and the control samples are taken from patients without long-term COVID-19.

[0068] In one example, an unsupervised clustering algorithm can be used to analyze the gene expression data of the gene expression matrix of the experimental samples to obtain the molecular subtyping results corresponding to each experimental sample. In another example, the long new crown typing model constructed by the above embodiment can also be used to analyze the gene expression matrix of the experimental samples to obtain the molecular subtyping results corresponding to each experimental sample.

[0069] S200, according to the gene expression matrix corresponding to the experimental sample in step S100, the gene expression matrix corresponding to the control sample, and the molecular subtype typing results corresponding to the experimental sample, the experimental samples of different molecular subtypes are respectively subjected to differential gene analysis with the control sample to obtain significantly differential genes.

[0070] S300, performing functional pathway enrichment analysis on the significantly different genes to obtain the functional pathways with significant differences corresponding to each molecular subtype, and calculating the average expression of the gene set corresponding to the functional pathways with significant differences in each molecular subtype.

[0071] S400, after normalizing the expression averages of each gene set and the clinical variables included in the association analysis, Spearman correlation analysis is used to obtain the clinical variables associated with each molecular subtype. In this embodiment, the gene expression matrix of long new crown patients and non-long new crown patients with different molecular subtypes is subjected to differential gene analysis to obtain the functional pathways corresponding to the significantly different genes, and the gene set averages of the functional pathways with significant upregulation or significant downregulation are subjected to correlation analysis with the clinical variables included in the association analysis to obtain clinical variables associated with each long new crown molecular subtype. In this way, the pathological mechanism of the long new crown subtype can be known from the gene level, and then the clinical medication can be more accurately guided according to the subtype to which the long new crown patient belongs, and the clinical prognosis of the long new crown patient can be improved. At the same time, it also provides new research directions and research ideas for the long new crown typing.

[0072] In one example, the clinical variables included in the association analysis may include, but are not limited to, one or more of the white blood cell count, neutrophil count, platelet count, ferritin concentration, procalcitonin concentration, anxiety scale score, depression scale score, post-traumatic stress disorder scale score, and fatigue scale score.

[0073] The present disclosure also provides a device for associating long-term novel coronavirus molecular subtypes with clinical variables, which includes a data receiving module, a differential gene analysis module, a functional pathway analysis module and an association analysis module.

[0074] The data receiving module is used to obtain the gene expression matrices corresponding to multiple experimental samples, the molecular subtyping results corresponding to each experimental sample, and the gene expression matrices corresponding to multiple control samples. Among them, the experimental samples are taken from patients with long COVID-19, and the control samples are taken from patients without long COVID-19. The molecular subtyping results of the experimental samples can be obtained by performing unsupervised clustering analysis on the gene expression matrix of the experimental samples using a clustering algorithm. The differential gene analysis module is used to perform differential gene analysis on the experimental samples of different molecular subtypes and the control samples according to the gene expression matrix corresponding to the experimental samples, the gene expression matrix corresponding to the control samples, and the molecular subtyping results corresponding to the experimental samples, to obtain significantly different genes. The functional pathway analysis module is used to perform functional pathway enrichment analysis on the significantly different genes to obtain the functional pathways corresponding to each molecular subtype. The enriched functional pathways include functional pathways with significant differences (functional pathways with significant upregulation and functional pathways with significant downregulation) and non-significant functional pathways; the expression average value of the gene set corresponding to the functional pathway with significant differences in each molecular subtype is solved. The association analysis module is used to normalize the expression mean values ​​of each gene set and the clinical variables included in the association analysis, and then use Spearman correlation analysis to obtain the clinical variables associated with each molecular subtype.

[0075] The solution provided by the present disclosure is further described below in conjunction with specific embodiments.

[0076] Example 1

[0077] This embodiment is a method for establishing a long-term new crown typing model, which specifically includes:

[0078] Step S10, obtaining the gene expression matrix of peripheral blood samples from patients with long-term COVID-19.

[0079] In Example 1, peripheral blood samples were collected from 22 patients with long COVID-19, and each peripheral blood sample was divided into two parts. One part of the peripheral blood samples was used for transcriptome sequencing to obtain the transcriptome sequencing results corresponding to the 22 patients with long COVID-19, and the non-expressed genes in the transcriptome sequencing results were filtered to obtain the corresponding gene expression matrix; the other part of the peripheral blood samples was used for clinical testing, and the test results are shown in Table 1. Among them, the 22 patients with long COVID-19 were the experimental group.

[0080] Table 1 Information of 22 patients with COVID-19

[0081]

[0082]

[0083] Among them, FER: ferritin (μg / L), WBC: white blood cells (*10 9 / L), NEU: Neutrophils (*10 9 / L), PLT: platelets (*10 9 / L), PCT: procalcitonin (ng / mL), ANXU: anxiety scale score, SUPR: depression scale score, PTSD: post-traumatic stress disorder scale score, FATG: fatigue scale score; NA means that relevant data were not collected.

[0084] Step S20: Use a hierarchical clustering algorithm to perform unsupervised cluster analysis on the gene expression data contained in the gene expression matrix of long-COVID patients to determine the number of molecular subtypes of long-COVID.

[0085] Hierarchical clustering is an unsupervised clustering method that can create a hierarchical nested clustering tree to cluster data by calculating the similarity between data points of different categories. Figure 4 As shown, based on the gene expression matrix data of the long-term COVID-19 patients in the experimental group, a hierarchical clustering algorithm was used to remove the discrete values ​​and obtain two clustering results. In this way, the molecular subtypes of the long-term COVID-19 patients include the first molecular subtype and the second molecular subtype.

[0086] Step S30, using the gene expression matrix of the long COVID patients in the experimental group obtained in step S10 and the molecular subtypes of the long COVID patients in the experimental group obtained in step 20 as data sets, training the random forest model to obtain a long COVID typing model.

[0087] The gene expression data and corresponding molecular subtypes of the long-term new crown patients in the experimental group were used as data groups, and the data groups of multiple long-term new crown patients constituted the data set. Among them, 80% of the data groups in the data set were used as training data sets, and the other 20% were used as validation data sets. The training data set was used to train the random forest model with 10,000 iterations; the random forest model was a random forest model package (Random Forest, version 4.7-1.1) based on the R language (version 4.3.1). The validation data set was used to verify the long-term new crown typing model obtained through training. Figure 5 As shown, the accuracy of the model is predicted by the ROC curve, and the area under the ROC curve AUC value represents the accuracy of the long new crown typing model. In Example 1, the accuracy of the long new crown typing model detected by the validation data set was 95.96%, indicating that the typing model has high accuracy and good classification effect, and can effectively distinguish long new crown patients of different molecular subtypes.

[0088] Example 2

[0089] This embodiment is a method for associating long-term new crown molecular subtypes with clinical variables, specifically including:

[0090] Step S100, obtain the gene expression matrix of patients with long-term COVID-19 (the gene expression matrix of the experimental group) and the gene expression matrix of patients without long-term COVID-19 (the gene expression matrix of the control group).

[0091] Among them, the gene expression matrix of the experimental group adopts the gene expression matrix of the long-term new crown patients in Example 1. The gene expression matrix of the control group is to collect peripheral blood samples from 22 non-long-term new crown patients, and each peripheral blood sample is divided into two parts. One part is used for transcriptome sequencing to obtain the corresponding transcriptome sequencing results of 22 non-long-term new crown patients, and the non-expressed genes in the transcriptome sequencing results are filtered to obtain the corresponding gene expression matrix; the other part is used for clinical testing, and the test results are shown in Table 2.

[0092] Table 2 Information of 22 non-COVID patients

[0093]

[0094]

[0095] Among them, FER: ferritin (μg / L), WBC: white blood cells (*10 9 / L), NEU: Neutrophils (*10 9 / L), PLT: platelets (*10 9 / L), PCT: procalcitonin (ng / mL), ANXU: anxiety scale score, SUPR: depression scale score, PTSD: post-traumatic stress disorder scale score, FATG: fatigue scale score.

[0096] Step S200, screen differentially expressed genes and perform functional pathway enrichment to determine the transcriptional expression characteristics corresponding to each molecular subtype of the new coronavirus (the transcriptional expression characteristics represent the corresponding functional pathways).

[0097] Exemplarily, the DESeq2 package is used to perform differentially expressed gene analysis on the gene expression matrices of the two long-new crown molecular subtypes and the gene expression matrices of non-long-new crowns to obtain significantly differentially expressed genes (significantly differentially expressed genes refer to genes with p-values ​​< 0.05 and multiple differences > 2). Based on the Reactome database, the significantly differentially expressed genes are subjected to functional pathway enrichment analysis to obtain the functional pathways corresponding to the two long-new crown molecular subtypes. Based on the functional pathways, the transcriptome characteristics corresponding to the long-new crown molecular subtypes can be obtained. For example, Figure 6 As shown, patients with the first molecular subtype of long-term COVID-19 showed upregulated neutrophil degranulation function; Figure 7 As shown, patients with the second molecular subtype of long COVID-19 showed downregulated neutrophil degranulation function.

[0098] Step S300: Correlate the transcriptional expression characteristics of long coronavirus molecular subtypes with clinical variables through correlation analysis.

[0099] The mean expression values ​​of the gene sets corresponding to the functional pathways with significant changes in the functional pathways corresponding to each molecular subtype were calculated. The mean expression values ​​of each gene set and the clinical variables included in the association analysis were normalized. Among them, the clinical variables include but are not limited to ferritin concentration, white blood cell count, neutrophil count, platelet count, procalcitonin concentration, anxiety scale score, depression scale score, post-traumatic stress disorder scale score, and fatigue scale score.

[0100] Spearman was used to perform correlation analysis on the normalized data to obtain clinical variables related to each long coronavirus subtype.

[0101] For example, Figure 8 and Fig. 9 Taking the neutrophil degranulation functional pathway as an example, a correlation analysis was performed. The Neu score in the figure represents the average expression value of the gene set of the neutrophil degranulation functional pathway corresponding to the molecular subtype. Figure 6 and Figure 8As shown in the figure, the transcriptome characteristics of the first molecular subtype of long-term COVID-19 patients showed upregulated neutrophil degranulation function. In the correlation analysis, Neu score was positively correlated with white blood cells, neutrophils and calcitonin in clinical inflammatory indicators. At the same time, Neu score was positively correlated with clinical variables corresponding to social and psychological problems such as depression and post-traumatic stress disorder. Figure 7 and Fig. 9 As shown in the figure, the transcriptome characteristics of the second molecular subtype of long-term COVID-19 patients are characterized by down-regulation of neutrophil degranulation function. In the correlation analysis, Neu score showed a strong positive correlation with white blood cells and neutrophils in clinical inflammatory indicators, and Neu score was negatively correlated with calcitonin in clinical inflammatory indicators; at the same time, Neu score was negatively correlated with clinical variables corresponding to social and psychological problems of post-traumatic stress disorder.

[0102] Thus, in this example, based on the clinical relevance analysis of the neutrophil degranulation functional pathway, for patients with the first molecular subtype of long COVID-19, the impact of neutrophil-related immune inflammatory response on the clinical manifestations of long COVID-19 should be considered, and psychological intervention for patients of this subtype should also be strengthened. General psychological problems are easily overlooked, and the association analysis between long COVID-19 molecular subtypes and clinical variables also makes up for the possible deficiencies in the detection of psychological problems. For patients with the second molecular subtype of long COVID-19, the clinic should include more of the impact of other factors on the clinical symptoms related to long COVID-19.

[0103] It should be noted that, although the various steps of the long new crown typing method and the method for associating long new crown molecular subtypes with clinical variables in the present disclosure are described in a specific order in the accompanying drawings, this does not require or imply that these steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps, etc.

[0104] The present disclosure also provides an electronic device, the electronic device having the above-mentioned long new crown typing device or the above-mentioned long new crown molecular subtype and clinical variable association device. The electronic device can be expressed in the form of a general computing device; it can also communicate with one or more external devices (such as a keyboard, a pointing device, a Bluetooth device, etc.).

[0105] In an embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above-mentioned long new crown typing method, or a method for associating long new crown molecular subtypes with clinical variables is stored. In some possible embodiments, various aspects of the present disclosure can also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to cause the terminal device to execute the steps of the long new crown typing method described in the above embodiment, or the steps of the method for associating long new crown molecular subtypes with clinical variables.

[0106] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the present disclosure, which follows the general principles of the present disclosure and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A long COVID-19 typing method, characterized in that: include: Obtaining the gene expression matrix of the sample to be tested; The long-novel coronavirus typing model is used to process the gene expression matrix of the sample to be tested to obtain the typing results of the sample to be tested.

2. The long COVID-19 typing method according to claim 1, characterized in that: The long-term new crown typing model is established using the following steps: S10, obtaining gene expression matrices corresponding to a plurality of experimental samples respectively; the experimental samples are taken from patients with COVID-19; S20, using an unsupervised clustering algorithm to analyze the gene expression matrix of the experimental samples to obtain the molecular subtype typing results corresponding to each experimental sample; S30. Use the data set to train the random forest model to obtain a long-COVID typing model; wherein the data set includes the gene expression matrix corresponding to each experimental sample, and the molecular subtype typing results corresponding to each experimental sample.

3. The long COVID-19 typing method according to claim 2, characterized in that: In step S20, the unsupervised clustering algorithm is a hierarchical clustering algorithm; and the molecular subtyping results include 2 molecular subtypes.

4. The long COVID-19 typing method according to claim 2 or 3, characterized in that: Obtaining the gene expression matrix of the sample to be tested includes: Performing transcriptome sequencing on the sample to be tested to obtain transcriptome sequencing results of the sample to be tested; Filtering the unexpressed genes in the transcriptome sequencing results of the sample to be tested to obtain the gene expression matrix of the sample to be tested; Obtaining the gene expression matrices corresponding to multiple experimental samples includes: Performing transcriptome sequencing on the experimental samples to obtain transcriptome sequencing results of the experimental samples; The unexpressed genes in the transcriptome sequencing results of the experimental samples were filtered out to obtain the gene expression matrix corresponding to the experimental samples.

5. A method for associating long coronavirus molecular subtypes with clinical variables, characterized in that: The following steps are involved: S100, obtaining gene expression matrices corresponding to a plurality of experimental samples, molecular subtyping results corresponding to the experimental samples, and gene expression matrices corresponding to a plurality of control samples; the experimental samples are taken from patients with long-term COVID-19, and the control samples are taken from patients without long-term COVID-19; S200, according to the gene expression matrix corresponding to the experimental sample in step S100, the gene expression matrix corresponding to the control sample, and the molecular subtype typing result corresponding to the experimental sample, performing differential gene analysis on the experimental samples of different molecular subtypes and the control sample, respectively, to obtain significantly differential genes; S300, performing functional pathway enrichment analysis on the significantly different genes to obtain the functional pathways with significant differences corresponding to each molecular subtype; calculating the average expression value of the gene set corresponding to the functional pathways with significant differences in each molecular subtype; S400, after normalizing the average expression values ​​of each gene set and the clinical variables included in the association analysis, Spearman correlation analysis was used to obtain the clinical variables associated with each molecular subtype.

6. The method for associating long coronavirus molecular subtypes with clinical variables according to claim 5, characterized in that: In step S400, the clinical variables included in the association analysis include one or more of the white blood cell count, neutrophil count, platelet count, ferritin concentration, procalcitonin concentration, anxiety scale score, depression scale score, post-traumatic stress disorder scale score, and fatigue scale score.

7. A long new crown typing device, characterized in that, include: An acquisition module is used to obtain the gene expression matrix of the sample to be tested; The processing module is used to receive the gene expression matrix of the sample to be tested transmitted by the acquisition module, and use the long new crown typing model to process the gene expression matrix of the sample to be tested to obtain the typing result of the sample to be tested.

8. The long new crown typing device according to claim 7 is characterized in that: It also includes a generation module for generating a long-coronavirus typing model; The generation module comprises: A data acquisition unit, used to obtain gene expression matrices corresponding to a plurality of experimental samples; the experimental samples are taken from patients with COVID-19; The typing unit is used to analyze the gene expression matrix of the experimental samples using an unsupervised clustering algorithm to obtain the molecular subtype typing results corresponding to each experimental sample; A training unit is used to train the random forest model using a data set to obtain a long-coronavirus typing model; wherein the data set includes a gene expression matrix corresponding to each experimental sample and a molecular subtyping result corresponding to each experimental sample.

9. A device for associating long coronavirus molecular subtypes with clinical variables, characterized in that: include: A data receiving module is used to obtain gene expression matrices corresponding to multiple experimental samples, molecular subtyping results corresponding to each experimental sample, and gene expression matrices corresponding to multiple control samples; wherein the experimental samples are taken from patients with long-term COVID-19, and the control samples are taken from patients without long-term COVID-19; The differential gene analysis module is used to perform differential gene analysis on the experimental samples of different molecular subtypes and the control samples according to the gene expression matrix corresponding to the experimental samples, the gene expression matrix corresponding to the control samples, and the molecular subtype typing results corresponding to the experimental samples, to obtain significantly differential genes; Functional pathway analysis module, used to perform functional pathway enrichment analysis on significantly different genes, obtain functional pathways with significant differences corresponding to each molecular subtype, and calculate the average expression of gene sets corresponding to functional pathways with significant differences in each molecular subtype; The association analysis module is used to normalize the expression mean values ​​of each gene set and the clinical variables included in the association analysis, and then use Spearman correlation analysis to obtain the clinical variables associated with each molecular subtype.

10. An electronic device, characterized in that: It includes the long new crown typing device described in claim 7 or 8 and / or the device for associating long new crown molecular subtypes with clinical variables described in claim 9.

Images