Screening method of respiratory tract infection related biomarkers based on metatranscriptomics

By screening respiratory infection-related biomarkers through macrotranscriptomics and combining it with machine learning algorithms, the problems of slow detection speed and low sensitivity in traditional diagnostic methods have been solved, and rapid and accurate diagnosis of respiratory infections has been achieved, especially with high discrimination ability in severe conditions.

CN120738336AActive Publication Date: 2025-10-03中国人民解放军总医院第八医学中心

Patent Information

Application Number
CN202510655121.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-21
Publication Date
2025-10-03
Estimated Expiration
2045-05-21

AI Technical Summary

Technical Problem

Traditional diagnostic methods have slow detection speed and low sensitivity in respiratory tract infections and cannot accurately identify pathogens, especially in lower respiratory tract infections, making diagnosis difficult.

Method used

A metatranscriptomics approach was used to sequence, quality control, and compare respiratory infection samples. Combined with differential expression analysis and co-expression analysis, the LASSO algorithm, random forest model, and SVM model were used to screen out intersection genes as biomarkers, and their diagnostic value was verified through machine learning.

Benefits of technology

The rapid and accurate diagnosis capability of respiratory infections has been improved. The biomarker combination has shown good clinical diagnostic performance and accuracy in the diagnosis of respiratory infections, especially in identifying severe conditions, with a diagnostic accuracy of 81.07%.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120738336A_ABST
    Figure CN120738336A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of biological detection, and discloses a screening method of respiratory tract infection related biomarkers based on metatranscriptomics. Respiratory tract infection samples with different clinical phenotypes are subjected to metatranscriptome sequencing, data quality control, comparison and transcript quantification, pathogen and host information is reserved, then genes which are stably expressed or remarkably changed in different groups are identified by combining differential expression analysis and co-expression analysis, and potential biomarkers are obtained. And taking an intersection gene of three machine learning algorithms including an LASSO algorithm, a random forest model and an SVM model to obtain the biomarker. The invention provides a biomarker screening method for rapid and accurate identification of respiratory tract infection.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of biological detection technology and relates to a method for screening respiratory tract infection-related biomarkers based on macrotranscriptomics. Background Art

[0002] Respiratory infections, in which pathogens invade the respiratory tract, leading to an inflammatory response, are the most common cause of human illness and can easily lead to hospitalization and death. Individuals with compromised immunity (weakened heart, lungs, and immune systems), the elderly, and infants are also at risk of developing serious complications. Lower respiratory tract infections (LRTIs), in particular, cause over 3 million deaths annually and present numerous diagnostic challenges. Traditional diagnostic methods can only isolate a small number of pathogens, leaving the vast majority of microorganisms that cannot be isolated and cultured unavailable for further study. These methods also suffer from slow detection speeds and low sensitivity, resulting in inaccurate pathogen identification in many cases. Conventional microbiological methods identify the pathogen in only 30-40% of LRTI cases.

[0003] Metatranscriptomics is a research method that directly extracts the entire RNA of microbial communities in a specific environment and analyzes their gene expression dynamics and regulatory mechanisms. This method can comprehensively reveal the composition and function of the microbial community in that environment, as well as the host immune response. Metatranscriptomics not only detects the transcriptional activity of pathogens but also analyzes the host response to infection, providing more comprehensive information for diagnosis and treatment. Compared with traditional diagnostic methods, metatranscriptomics can provide more comprehensive pathogen detection and host response analysis, helping to improve diagnostic accuracy and efficiency.

[0004] Biomarkers are characteristics that can be objectively measured and evaluated, reflecting physiological or pathological processes, as well as the biological effects of exposure or therapeutic interventions. They are used to indicate normal biological processes, pathological processes, or responses to exposure or intervention. Exploring biomarkers for respiratory infections based on metatranscriptomics is crucial for the rapid and accurate identification of respiratory infections. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a method for screening respiratory tract infection-related biomarkers based on macrotranscriptomics. Respiratory tract infection samples with different clinical phenotypes are subjected to macrotranscriptome sequencing, data quality control, alignment, transcript quantification, and host transcriptome level information. Subsequently, differential expression analysis and co-expression analysis are combined to identify genes that are stably expressed or significantly changed in different groups to obtain potential biomarkers. The intersection genes of three machine learning algorithms, LASSO algorithm, random forest model and SVM model, are taken to obtain the biomarkers. The present invention provides a method for screening biomarkers for the rapid and accurate identification of respiratory tract infections.

[0006] To achieve the technical objectives of the present invention, on the one hand, the present invention provides a method for screening respiratory tract infection-related biomarkers based on macrotranscriptomics, which specifically comprises the following steps: Obtain samples from different groups and extract total RNA samples; Metatranscriptome sequencing to obtain metatranscriptome data; The macrotranscriptome data are quality controlled and aligned, transcripts are quantified, and differential expression analysis and co-expression analysis are combined to screen differentially expressed genes and co-expressed genes in different groups to obtain potential biomarkers; The differentially expressed genes and co-expressed genes are selected by a machine learning algorithm and then the intersection genes are obtained to obtain biomarkers; The diagnostic value of the biomarkers was verified.

[0007] Furthermore, the types of samples of the present invention include bronchoalveolar lavage fluid and blood.

[0008] Furthermore, the present invention divides patients with respiratory tract infection into an infection group and a non-infection group, and the patients in the infection group meet any one of the following conditions: inflammatory marker CRP ≥ 10 g / L, inflammatory marker PCT ≥ 0.5 ng / mL, positive serotype test (Streptococcus pneumoniae or Haemophilus influenzae) result, positive clinical culture result (fungal or bacterial pathogens are cultured from the analysis sample), G test result > 80 pg / mL, and positive mNGS result; the patients in the non-infection group have CRP < 10 g / L and PCT < 0.5 ng / mL, negative serotype test result, negative clinical culture result, G test result ≤ 80 pg / mL, and negative mNGS result.

[0009] Furthermore, the present invention also divides the infection group into a severe pneumonia group and a mild pneumonia group. The patients in the severe pneumonia group meet any one of the inflammatory markers CRP ≥ 30g / L, the inflammatory marker PCT ≥ 2ng / mL, and the G test result ≥ 200pg / mL. The patients in the mild pneumonia group meet the inflammatory markers 10g / L ≤ CRP < 30g / L and the inflammatory markers 0.5ng / mL ≤ PCT < 2ng / mL, 80pg / mL < G test result < 200pg / mL.

[0010] Furthermore, the present invention collects bronchoalveolar lavage fluid and blood from patients with respiratory tract infection in different groups as analysis samples, and extracts total RNA samples from the analysis samples.

[0011] Furthermore, the adapter sequences and low-quality bases in the total RNA sample are removed, rRNA is removed, and then the RNA is reverse transcribed into cDNA. The ends are repaired, tailed, and the sequencing adapters are connected. The cDNA is then PCR amplified to construct a cDNA library and perform macrotranscriptome sequencing to obtain macrotranscriptome data.

[0012] Furthermore, the metatranscriptome data were quality controlled and filtered by FastQC and Fastp, with default parameters set. RNA quality was assessed by the RIN value, which was required to be ≥ 7.0. Bowtie2 was used to map the quality-controlled reads to the human reference genome GRCh38 to generate a BAM file containing position information.

[0013] Furthermore, based on the BAM file, Cufflink‌ was used to calculate the raw read counts of each gene for transcript quantification.

[0014] Furthermore, the differential expression analysis compares genes between the infection group and the non-infection group, and between the severe pneumonia group and the mild pneumonia group, sets a threshold, and screens to obtain the differentially expressed genes; The thresholds were set as |log2FC|≥1 and FDR≤0.05.

[0015] Furthermore, the co-expression analysis includes constructing a weighted gene co-expression network, setting a soft threshold, calculating the R value, where R is the Person correlation coefficient, and when R 2 When the connectivity is greater than 0.8 and the average connectivity is high, the soft threshold is set to 7. The co-expression analysis also includes calculating the correlation between the module characteristic genes and the symptom phenotype, screening the significantly correlated modules, and extracting the genes with the top 10% connectivity within the module from the significantly correlated modules to obtain the co-expressed genes; the symptom phenotype is an infection state or a severe state.

[0016] Specifically, the infection status is divided into infected and non-infected, which is determined by clinical culture results and mNGS results. If the clinical culture result or mNGS result is positive, it is judged as infected; if the clinical culture result and mNGS result are both negative, it is judged as non-infected.

[0017] Specifically, severe conditions are divided into severe and mild conditions, which are distinguished based on clinical indicators. The clinical indicators are SOFA score and APACHE II score. 2 points < SOFA score ≤ 5 points and 15 points < APACHE II score ≤ 24 points are judged as mild conditions, and SOFA score > 5 points and APACHE II score > 24 points are judged as severe conditions.

[0018] Furthermore, the machine learning algorithm is LASSO algorithm, random forest model and SVM model.

[0019] Furthermore, the validation includes establishing a ROC curve of the biomarker, and an AUC>0.8 is considered to have diagnostic value.

[0020] Compared with the prior art, the technical solution provided by the present invention has at least the following beneficial effects or advantages: (1) The present invention performs metatranscriptome sequencing on respiratory infection samples with different clinical phenotypes, removes non-informative sequences such as host RNA through quality control, and retains pathogen and host information. Subsequently, differential expression analysis and co-expression analysis are combined to identify genes that are stably expressed or significantly changed in different groups to obtain potential biomarkers. The biomarkers are obtained by taking the intersection genes of three machine learning algorithms: the LASSO algorithm, the random forest model, and the SVM model. The present invention provides a biomarker screening method for the rapid and accurate identification of respiratory infections.

[0021] (2) Based on the screening method for respiratory tract infection-related biomarkers provided by the present invention, the biomarkers screened out as IL6, CXCL8, TNF-α, HSP90AA1, and S100A9 were related to respiratory tract infection. The diagnostic performance of the biomarkers was further verified, and the ROC curve was analyzed. The results showed that the area under the curve of the biomarker combination was 0.847, which had good clinical diagnostic performance, and the diagnostic accuracy of the above biomarkers reached 80.14%, respectively. The biomarkers had a high discriminant ability in the diagnosis of respiratory tract infection. The present invention used the above screening method to screen biomarkers related to whether the disease was severe, and the biomarkers obtained were IL6, CXCL8, and TNF-α. The diagnostic performance of the biomarkers was further verified, and the ROC curve was analyzed. The results showed that the area under the curve of the biomarker combination was 0.862, which had good clinical diagnostic performance, and the diagnostic accuracy of the above biomarkers reached 81.07%, and the biomarkers had a high discriminant ability in the diagnosis of whether the respiratory tract infection was severe. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention.

[0023] Figure 1 Flowchart of the screening method for respiratory tract infection-related biomarkers based on metatranscriptomics. DETAILED DESCRIPTION

[0024] The technical solutions of the present invention are now described with reference to the following examples. However, the present invention is not limited to the following examples. The experimental methods and detection methods described in the following examples are conventional methods unless otherwise specified; the reagents and materials described are commercially available unless otherwise specified.

[0025] Example 1 This example provides a method for screening biomarkers related to respiratory tract infection based on macrotranscriptomics, the flow chart is as follows: Figure 1 As shown, screening for biomarkers for identifying infection status specifically includes the following steps: 1. Sample Data Acquisition: Biological samples, including bronchoalveolar lavage fluid and blood, were collected from patients with respiratory tract infections as analytical samples. Total RNA samples were extracted using a tissue cell RNA extraction kit (Trizol extraction method) (purchased from Shanghai Shangbao Biotechnology Co., Ltd.) and stored at -80°C. 2. Library Construction: Cutadapt was used to remove adapter sequences and low-quality bases, and the Ribo-Zero™ rRNA Removal Kit (Gram-Negative Bacteria) 1 Kit (purchased from Shanghai Cent Biotechnology) was used to remove rRNA. Magnetic beads with Oligo (dT) were used to enrich eukaryotic mRNA through AT complementary pairing and binding to the ployA tail of mRNA. Fragmentation buffer was added to break the mRNA into short fragments. Subsequently, the first-strand cDNA was synthesized using RNA as a template with six-base random primers. The second-strand cDNA was synthesized by adding buffer, dNTPs, and DNA polymerase I. The double-stranded cDNA was then purified using AMPure XP beads, end-repaired, tailed, and ligated with sequencing adapters. Finally, the cDNA was amplified by PCR to construct a cDNA library. The enriched cDNA was subjected to metatranscriptome sequencing using NovaSeq on the Illumina platform at a sequencing depth of approximately 10 Gb to obtain metatranscriptome data. 3. Data quality control and alignment: Metatranscriptome data were quality controlled and filtered using FastQC (v 0.11.9) and Fastp (v 0.20.1), with default parameters. RNA quality was assessed using the RNA integrity number (RIN ≥ 7.0). Bowtie2 (v 2.3.5.1) was used to map the quality-controlled reads to the human reference genome GRCh38, generating a BAM file containing position information. 4. Transcript quantification: Based on the alignment results (BAM files), the raw read counts (Raw Counts) of each gene are calculated using Cufflink. 5. Standardized entry of patient clinical information: A structured clinical database will be established, covering basic patient information (age, immune status) and symptom characteristics. Patients will be grouped according to their symptoms into an infection group (satisfying any one of the following: inflammatory marker CRP ≥ 10 g / L, inflammatory marker PCT ≥ 0.5 ng / mL, positive serotype test (Streptococcus pneumoniae or Haemophilus influenzae) result, positive clinical culture result (analysis sample cultured for fungal or bacterial pathogens), G test result > 80 pg / mL, or positive mNGS result) and a non-infection group (CRP < 10 g / L and PCT < 0.5 ng / mL, negative serotype test result, negative clinical culture result, G test result ≤ 80 pg / mL, and negative mNGS result). 6. Biomarker screening: Combine differential expression analysis, co-expression analysis, and machine learning algorithms to conduct in-depth mining of gene expression data from different groups to identify genes that are stably expressed or significantly changed in the infected group as biomarkers; (1) Differential expression analysis: The edgeR package was used for pairwise comparison (infected group vs. non-infected group), and the threshold (|log2FC| ≥ 1 and FDR ≤ 0.05) was set to screen genes that were significantly changed only in a certain symptom group; (2) Co-expression analysis: Construct a weighted gene co-expression network, set a soft threshold (power value) to ensure a scale-free topological structure, and calculate the Person correlation coefficient (R). 2 When the mean connectivity is higher than 0.8, the soft threshold is set to 7; the Spearman correlation coefficient between the module characteristic genes (MEs) and the symptom phenotype (infection status) is calculated. The infection status is divided into infected and uninfected, which is determined by clinical culture results and mNGS results. If the clinical culture results or mNGS results are positive, it is determined to be infected. If the clinical culture results and mNGS results are both negative, it is determined to be uninfected. The significantly correlated modules ( P<0.05); extract hub genes from significant modules (top 10% connectivity within the module) and screen candidate genes based on differential expression results; (3) Functional annotation: clusterProfiler was used to perform functional annotation and enrichment analysis on differentially expressed genes and co-expressed genes; (4) Machine learning algorithm: Based on differentially expressed genes and co-expressed genes, redundant genes were compressed through cross-validation based on LASSO regression, and variables with high contribution to symptom classification (the top 10 contributing genes) were retained; the random forest model in R language was used to predict respiratory infection-related genes, and the R software package "randomForest" was used to construct the random forest model. The 10% genes with the highest scores were selected based on the gene importance score; the "rfe" of the R software package "caret" was used to gradually remove the features that contributed the least to the model to evaluate the importance of genes. In each iteration, the SVM model was used to sort the features, and 20% of the least important features were removed each time. The performance of the feature subset was evaluated by combining cross-validation, and the feature combination with the largest AUC was selected as the final result; the results of the LASSO, random forest, and SVM model algorithms were integrated, and the intersection genes were taken as biomarkers; 7. Verification of screening results: The robustness of the biomarker was verified in an independent data set, and the ROC curve of the biomarker was established. If AUC>0.8, it was considered to have clinical value.

[0026] Based on the above screening method, the biomarkers associated with respiratory tract infection were IL6, CXCL8, TNF-α, HSP90AA1, and S100A9. The diagnostic performance of the biomarkers was further verified and the ROC curve was analyzed. The results showed that the area under the curve of the biomarker combination was 0.847, which has good clinical diagnostic performance. The diagnostic accuracy of the above biomarkers reached 80.14%, and the biomarkers have a high discriminatory ability in the diagnosis of respiratory tract infections.

[0027] Example 2 This example provides a method for screening biomarkers related to respiratory tract infection based on macrotranscriptomics, the flow chart is as follows: Figure 1 As shown, screening for biomarkers for identifying severe conditions specifically includes the following steps: 1. Sample Data Acquisition: Biological samples, including bronchoalveolar lavage fluid and blood, were collected from patients with respiratory tract infections as analytical samples. Total RNA samples were extracted using a tissue cell RNA extraction kit (Trizol extraction method) (purchased from Shanghai Shangbao Biotechnology Co., Ltd.) and stored at -80°C. 2. Library Construction: Cutadapt was used to remove adapter sequences and low-quality bases, and the Ribo-Zero™ rRNA Removal Kit (Gram-Negative Bacteria) 1 Kit (purchased from Shanghai Cent Biotechnology) was used to remove rRNA. Magnetic beads with Oligo (dT) were used to enrich eukaryotic mRNA through AT complementary pairing and binding to the ployA tail of mRNA. Fragmentation buffer was added to break the mRNA into short fragments. Subsequently, the first-strand cDNA was synthesized using RNA as a template with six-base random primers. The second-strand cDNA was synthesized by adding buffer, dNTPs, and DNA polymerase I. The double-stranded cDNA was then purified using AMPure XP beads, end-repaired, tailed, and ligated with sequencing adapters. Finally, the cDNA was amplified by PCR to construct a cDNA library. The enriched cDNA was subjected to metatranscriptome sequencing using NovaSeq on the Illumina platform at a sequencing depth of approximately 10 Gb to obtain metatranscriptome data. 3. Data quality control and alignment: Metatranscriptome data were quality controlled and filtered using FastQC (v 0.11.9) and Fastp (v 0.20.1), with default parameters. RNA quality was assessed using the RNA integrity number (RIN ≥ 7.0). Bowtie2 (v 2.3.5.1) was used to map the quality-controlled reads to the human reference genome GRCh38, generating a BAM file containing position information. 4. Transcript quantification: Based on the alignment results (BAM files), the raw read counts (Raw Counts) of each gene are calculated using Cufflink. 5. Standardized entry of patient clinical information: A structured clinical database will be established, covering basic patient information (age, immune status) and symptom characteristics. Patients will be grouped according to different symptoms. The infection group (meeting any one of the following: inflammatory marker CRP ≥ 10 g / L, inflammatory marker PCT ≥ 0.5 ng / mL, positive serotype test (Streptococcus pneumoniae or Haemophilus influenzae) result, positive clinical culture result (analysis sample cultured for fungal or bacterial pathogens), G test result > 80 pg / mL, or positive mNGS result) will be divided into a severe pneumonia group (meeting any one of the following: inflammatory marker CRP ≥ 30 g / L, inflammatory marker PCT ≥ 2 ng / mL, or G test result ≥ 200 pg / mL) and a mild pneumonia group (meeting any one of the following: inflammatory marker 10 g / L ≤ CRP < 30 g / L, inflammatory marker 0.5 ng / mL ≤ PCT < 2 ng / mL, inflammatory marker 80 pg / mL < G test result < 200 pg / mL). 6. Biomarker screening: In-depth mining of gene expression data from different groups was conducted using differential expression analysis, co-expression analysis, and machine learning algorithms to identify genes that were stably expressed or significantly changed in the severe pneumonia group as biomarkers; (1) Differential expression analysis: The edgeR package was used for pairwise comparison (severe pneumonia group vs mild pneumonia group), and the threshold (|log2FC| ≥ 1 and FDR ≤ 0.05) was set to screen genes that were significantly changed only in a certain symptom group; (2) Co-expression analysis: Construct a weighted gene co-expression network, set a soft threshold (power value) to ensure a scale-free topological structure, and calculate the Person correlation coefficient (R). 2 When the mean connectivity is higher than 0.8, the soft threshold is set to 7; the Spearman correlation coefficient between the module characteristic genes (MEs) and the symptom phenotype (severe state) is calculated, and the severe state is scored based on clinical indicators (SOFA score, APACHE II score). The scoring table is shown in Table 1. The significantly correlated modules ( P <0.05); extract hub genes from significant modules (top 10% connectivity within the module) and screen candidate genes based on differential expression results; (3) Functional annotation: clusterProfiler was used to perform functional annotation and enrichment analysis on differentially expressed genes and co-expressed genes; (4) Machine learning algorithm: Based on differentially expressed genes and co-expressed genes, redundant genes were compressed through cross-validation based on LASSO regression, and variables with high contribution to symptom classification (the top 10 contributing genes) were retained; the random forest model in R language was used to predict respiratory infection-related genes, and the R software package "randomForest" was used to construct the random forest model. The 10% genes with the highest scores were selected based on the gene importance score; the "rfe" of the R software package "caret" was used to gradually remove the features that contributed the least to the model to evaluate the importance of genes. In each iteration, the SVM model was used to sort the features, and 20% of the least important features were removed each time. The performance of the feature subset was evaluated by combining cross-validation, and the feature combination with the largest AUC was selected as the final result; the results of the LASSO, random forest, and SVM model algorithms were integrated, and the intersection genes were taken as biomarkers; 7. Verification of screening results: The robustness of the biomarker was verified in an independent data set, and the ROC curve of the biomarker was established. If AUC>0.8, it was considered to have clinical value.

[0028] Table 1 Severe illness score table

[0029] Based on the above screening method, the biomarkers associated with severity were IL6, CXCL8, and TNF-α. The diagnostic performance of the biomarkers was further verified, and the ROC curve was analyzed. The results showed that the area under the curve of the biomarker combination was 0.862, which has good clinical diagnostic performance. The diagnostic accuracy of the above biomarkers reached 81.07%. The biomarkers have a high discriminant ability in diagnosing whether respiratory infections are severe.

[0030] The specification only describes the preferred embodiments of the present invention. The present invention is not limited to the above embodiments. Without departing from the spirit and scope of the present invention, various changes and improvements made to the technical solutions of the present invention by ordinary technicians in this field should fall within the protection scope determined by the present invention.

Claims

1. A method for screening respiratory tract infection-related biomarkers based on macrotranscriptomics, characterized in that: include: Obtain samples from different groups and extract total RNA samples; Metatranscriptome sequencing to obtain metatranscriptome data; The macrotranscriptome data are quality controlled and compared, and differential expression analysis and co-expression analysis are combined to screen out differentially expressed genes and co-expressed genes in different groups to obtain potential biomarkers; The differentially expressed genes and co-expressed genes are selected by a machine learning algorithm and then the intersection genes are obtained to obtain biomarkers; The different groups include infected and non-infected groups; The infection group consisted of a severe pneumonia group and a mild pneumonia group; The differential expression analysis compares genes in the infection group with those in the non-infection group, and compares genes in the severe pneumonia group with those in the mild pneumonia group, sets a threshold, and screens to obtain the differentially expressed genes; The thresholds were set as |log2FC|≥1 and FDR≤0.

05.

2. The screening method according to claim 1, wherein Patients in the infection group meet any one of the following conditions: inflammatory marker CRP ≥ 10 g / L, inflammatory marker PCT ≥ 0.5 ng / mL, positive serotype test result, positive clinical culture result, G test result > 80 pg / mL, or positive mNGS result; The patients in the non-infection group met the conditions of CRP < 10 g / L and PCT < 0.5 ng / mL, the serotype test result was negative, the clinical culture result was negative, the G test result was ≤ 80 pg / mL, and the mNGS result was negative.

3. The screening method according to claim 1, wherein The co-expression analysis includes constructing a weighted gene co-expression network and setting a soft threshold; R 2 When ≥0.8 and the average connectivity is high, the soft threshold is set to 7; The R 2 The R in it is the Person correlation coefficient.

4. The screening method according to claim 1, wherein The co-expression analysis further includes calculating the correlation between module characteristic genes and symptom phenotypes, screening significantly correlated modules, and extracting genes with top 10% connectivity within the modules from the significantly correlated modules to obtain the co-expressed genes; The symptom phenotype is manifested by an infection state and a severe state.

5. The screening method according to claim 4, characterized in that The infection status is divided into infected and uninfected, and the infection status is determined by clinical culture results and mNGS results; The severe condition is divided into severe and mild conditions, which are distinguished according to clinical indicators, and the clinical indicators are SOFA score and APACHE II score.

6. The screening method according to claim 1, wherein The machine learning algorithms are LASSO algorithm, random forest model and SVM model.

7. The screening method according to claim 1, wherein Validating the diagnostic value of the biomarkers; The validation includes establishing a ROC curve of the biomarker, and an AUC > 0.8 is considered to have diagnostic value.

8. The screening method according to claim 1, wherein The sample types included bronchoalveolar lavage fluid and blood.

9. The screening method according to claim 1, wherein The quality control was to filter the macrotranscriptome data through FastQC and Fastp, both of which were set to default parameters.

10. The screening method according to claim 1, wherein The alignment is to map the quality-controlled data to the human reference genome GRCh38 using Bowtie2 to generate a BAM file containing position information; After the alignment, transcript quantification was performed using Cufflink.

Citation Information

Patent Citations

  • Treatment of inflammation, respiratory tract infections and cystic fibrosis

    CN107206020A

  • Respiratory tract infection pathogen detection method based on metagene / metatranscriptome sequencing

    CN114107454A

  • Biomarkers for respiratory tract infection diagnosis

    CN115136009A

  • Biomarker for meningitis diagnosis and application thereof

    CN115505637A

  • Lung fungal infection auxiliary diagnosis method and device based on ensemble learning

    CN117831754A

Cited By

  • Spectrum-based method for detecting infection markers in respiratory tract secretions

    CN121384916A