A combination of seminal plasma metabolic biomarkers for the subtyping diagnosis of male infertile spermatogenic disorders

Abstract

The application discloses a seminal plasma metabolic biomarker combination for male infertility spermatogenesis obstacle typing diagnosis, and relates to the field of biological science and technology.The application records the metabolic fingerprints of 475 seminal plasma samples by means of nanoparticle-enhanced laser desorption / ionization mass spectrometry, and constructs a seminal plasma metabolic fingerprint database related to spermatogenesis obstacle (SPGF). Then, through machine learning and statistical analysis, six metabolic biomarkers for SPGF typing diagnosis are screened and identified in the seminal plasma, and the six metabolic biomarkers can realize accurate typing of SPGF (AUC=0.814). The application has the advantages of simple sample pretreatment, fast analysis speed and low cost by means of recording the metabolic fingerprints of seminal plasma by means of nanoparticle-enhanced laser desorption / ionization mass spectrometry. Through the synthesized iron oxide nanoparticles as a matrix, small molecule metabolites in the seminal plasma can be efficiently captured and ionized, and complex pretreatment steps such as deproteinization and chromatographic separation are not needed, so that super-fast analysis, super-high sensitivity, extremely low sample consumption and excellent repeatability detection are realized.

Description

Technical Field

[0001] This invention relates to the field of bioscience and technology, and in particular to a combination of seminal plasma metabolic biomarkers for the classification and diagnosis of male infertility and spermatogenesis disorders. Background Technology

[0002] Spermatogenesis disorder (SPGF) is the most severe form of male infertility, affecting approximately 7% of men worldwide. Accurate SPGF typing is crucial for guiding individualized treatment strategies, particularly in predicting the success rate of micro-TESE (micro-tracerebrene retrieval for sperm). However, existing methods for SPGF diagnosis and typing have significant limitations. Currently, clinical practice relies primarily on hormonal markers such as follicle-stimulating hormone (FSH), inhibin B, luteinizing hormone (LH), and testosterone for SPGF diagnosis and typing. However, these markers have limited ability to differentiate between F-SPGF and C-SPGF, with sensitivity and specificity maintained at only 50%-70%, exhibiting poor specificity (AUC approximately 0.5-0.7), easily leading to misdiagnosis or missed diagnosis, and failing to provide a reliable basis for precision treatment. While testicular biopsy is considered a diagnostic reference standard, it is an invasive procedure with a risk of iatrogenic injury, and approximately 57% of cases have sampling bias due to tissue heterogeneity. Therefore, over 70% of SPGF patients are accurately typed only after missing the optimal treatment window. Further exploration of non-invasive biomarkers based on seminal plasma for accurate SPGF subtyping diagnosis in a clinical setting is of great significance.

[0003] Metabolic biomarkers derived from biological fluids show great potential in disease diagnosis because they can directly reflect disease phenotypes and real-time functional changes. Compared with systemic circulating serum, seminal plasma, derived directly from the testes, can more specifically reflect the molecular characteristics of local spermatogenesis function. However, existing seminal plasma metabolomics studies on SPGF diagnosis are limited by relatively small-scale cohort studies (sample sizes of approximately 40-144), single diagnostic functions, and a lack of biological function validation. Therefore, identifying seminal plasma metabolic biomarkers and validating their biological functions based on large-scale SPGF cohort studies for potential diagnostic and subtyping purposes is of significant research value.

[0004] Mass spectrometry, as a primary tool for metabolite analysis, is highly favored for its high sensitivity and ability to achieve label-free identification through precise measurement of mass-to-charge ratio (m / z). However, the analysis of metabolites in biological fluids using liquid / gas chromatography-mass spectrometry involves complex sample pretreatment, such as deproteinization and metabolite purification; limited analytical speed and throughput; and high cost, making it difficult to meet the clinical demand for "rapid and efficient detection" and unsuitable for large-scale screening and typing of SPGF patients.

[0005] To address the aforementioned issues, this invention aims to provide a combination of seminal plasma metabolic biomarkers for seminal plasma metabolic fingerprint analysis using nanoparticle-enhanced laser desorption / ionization mass spectrometry (NDEMS), used for the classification and diagnosis of spermatogenesis disorders in male infertility. By combining this biomarker combination with NDEMS, accurate classification of F-SPGF and C-SPGF can be achieved, providing new technical support for the clinical diagnosis and treatment decisions of SPGF. Summary of the Invention

[0006] In view of the above-mentioned deficiencies of the prior art, the technical problem to be solved by the present invention is to develop a combination of seminal plasma metabolic biomarkers for the classification and diagnosis of male infertility spermatogenesis disorders.

[0007] To achieve the above objectives, the present invention provides a combination of seminal plasma metabolic biomarkers for the classification and diagnosis of spermatogenesis disorders in male infertility.

[0008] Furthermore, the combination of seminal plasma metabolic biomarkers includes urea, spermine, proline, leucine, taurine, and aspartic acid; The screening method for the combination of seminal plasma metabolic biomarkers includes the following steps: Step 1: Metabolic detection was performed on seminal plasma samples from the F-SPGF group, C-SPGF group, and healthy control group using nanoparticle-enhanced laser desorption / ionization time-of-flight mass spectrometry to obtain seminal plasma metabolic fingerprint data; The method for obtaining the seminal plasma metabolic fingerprint data is as follows: 1) Prepare a matrix solution of 1 mg / mL using iron oxide nanoparticles with deionized water; 2) Dilute the seminal plasma sample 20 times with deionized water, spot 2 μL of the diluted seminal plasma sample onto the mass spectrometry target plate, allow it to air dry at room temperature, add 2 μL of the matrix solution, and allow it to air dry again at room temperature to obtain the seminal plasma sample for mass spectrometry analysis; 3) Seminal plasma metabolic fingerprint data were acquired using a laser desorption / ionization time-of-flight mass spectrometer with a laser wavelength of 355 nm and a frequency of 1000 Hz. Each sample was acquired for approximately 30 seconds to obtain seminal plasma metabolic fingerprint data. Step 2: Perform data preprocessing on the seminal plasma metabolic fingerprint data collected from the F-SPGF group, C-SPGF group and healthy control group to obtain the target metabolic characteristics; Step 3: Screening biomarkers from target metabolic features: Machine learning screening was performed using the LASSO regression model. The model was trained by optimizing the parameter λ, and seminal plasma metabolic features with a model score > 0.3 were retained. At the same time, combined with statistical analysis, the criteria for significant difference were "p < 0.05 and |log2(foldchange)| > 0.3". Finally, after excluding matrix peaks and exogenous substances, 6 target metabolic biomarkers were obtained.

[0009] Furthermore, in step two, the data preprocessing includes data resampling, spectral smoothing, baseline correction, feature extraction, peak alignment, and missing value imputation.

[0010] Furthermore, the functional regulatory effects of leucine and proline in spermatogenesis of the combination of metabolic biomarkers are to promote GC-1 cell proliferation, reduce GC-1 cell apoptosis, and maintain testicular tissue integrity; while the functional regulatory effects of urea and spermine are to inhibit GC-1 cell proliferation, increase GC-1 cell apoptosis, and damage testicular tissue integrity.

[0011] Furthermore, the method for functional verification of the metabolic biomarker combination includes the following steps:

[0012] a. The GC-1 spermatogonial cell line was cultured in DMEM medium containing 10% fetal bovine serum, and then seeded into 96-well plates at a rate of 2000 cells / well. After overnight equilibration, the target metabolic biomarkers were added, and the cells were treated for 96 hours. Then, 10 μL of LCK-8 reagent was added, and the absorbance was measured at 450 nm to assess the cell proliferation rate.

[0013] b. Obtain testicular tissue with normal spermatogenesis from patients with obstructive azoospermia, and cut it into 8mm sections. 3 Fragments; tissue fragments were placed on agarose gel scaffolds, and the metabolic biomarkers to be tested were added and cultured at 34°C and 5% CO2 for 14 days;

[0014] c. After 14 days of culture, the tissue was fixed, embedded, and sectioned. Ki-67 immunofluorescence staining and cleaved-PARP staining were performed to calculate the positive cell rate. After HE staining, the integrity of seminiferous tubules was assessed according to the 1-4 point standard.

[0015] Furthermore, the application of the aforementioned combination of seminal plasma metabolic biomarkers in the diagnosis of male infertility and spermatogenesis disorders.

[0016] Furthermore, the application includes a rapid detection method for male infertility and spermatogenesis disorders. Specifically, the 317 metabolic features obtained from the preprocessing in step two are grouped into a healthy control group and an SPGF group including the F-SPGF group and the C-SPGF group. A diagnostic model is constructed on the detection queue using a neural network algorithm. The seminal plasma metabolic fingerprint data of the sample to be tested is input into the diagnostic model to output the detection results.

[0017] Furthermore, the diagnostic model demonstrated an AUC of 0.924, a sensitivity of 75.68%, a specificity of 92.98%, and an accuracy of 86.17% on the independent validation set.

[0018] Furthermore, the application of the aforementioned combination of seminal plasma metabolic biomarkers in the classification and diagnosis of male infertility spermatogenesis disorders.

[0019] Furthermore, the application includes a rapid detection method for male infertility spermatogenesis disorder typing, which specifically uses the peak intensity of six metabolic biomarkers as features, employs a neural network algorithm to construct a typing diagnostic model for F-SPGF and C-SPGF, inputs the seminal plasma metabolic fingerprint data of the sample to be tested into the diagnostic model, and outputs the typing detection result of the sample as F-SPGF or C-SPGF.

[0020] Furthermore, the AUC of the genotyping diagnostic model evaluated on the independent validation set was 0.814.

[0021] In a preferred embodiment 1 of the present invention, a method for acquiring seminal plasma metabolic fingerprint data using nanoparticle-enhanced laser desorption / ionization time-of-flight mass spectrometry is described in detail.

[0022] In another preferred embodiment 2 of the present invention, the seminal plasma metabolic fingerprint database and preprocessing method are described in detail.

[0023] In another preferred embodiment 3 of the present invention, a method for constructing and validating a metabolic fingerprint diagnostic model is described in detail.

[0024] In another preferred embodiment 4 of the present invention, the identification method of metabolic biomarkers and the construction method of SPGF typing diagnostic model are described in detail.

[0025] In another preferred embodiment 5 of the present invention, a method for verifying the biological function of metabolic biomarkers is described in detail.

[0026] Technical effects: 1. This invention first records the metabolic fingerprints of 475 seminal plasma samples using nanoparticle-enhanced laser desorption / ionization mass spectrometry, constructing a large-scale (475 cases) seminal plasma metabolic fingerprint database related to SPGF. Then, through machine learning and statistical analysis, six metabolic biomarkers for SPGF typing diagnosis were screened and identified in the seminal plasma. These six metabolic biomarkers can achieve accurate SPGF typing (AUC=0.814), which is significantly better than FSH (AUC=0.527), LH (AUC=0.636), and testosterone (AUC=0.657).

[0027] 2. This invention utilizes nanoparticle-enhanced laser desorption / ionization mass spectrometry to record seminal plasma metabolic fingerprints, enabling rapid detection of these fingerprints. This method offers advantages such as simple sample pretreatment, rapid analysis, and low cost. Using synthesized iron oxide nanoparticles as a matrix, it can efficiently capture and ionize small molecule metabolites in seminal plasma, eliminating the need for complex pretreatment steps such as protein removal and chromatographic separation. This achieves ultra-fast analysis (approximately 30 seconds / sample), ultra-high sensitivity (0.4-9.0 pmol), extremely low sample consumption (approximately 100 nL), and excellent repeatability (CV: 3.1-13.4%).

[0028] 3. This invention, through cell proliferation experiments and an in vitro culture model of human testicular tissue, verified the functional regulatory role of six metabolic biomarkers on germ cell proliferation and tissue integrity, providing mechanistic insights into the pathological mechanism of SPGF at the metabolic level.

[0029] The following will further explain the concept, specific structure, and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Attached Figure Description

[0030] Figure 1 This is a representative seminal plasma metabolic fingerprint of a healthy control group, an F-SPGF group, and a C-SPGFIE group, according to a preferred embodiment 2 of the present invention. Figure 2 This is the Ki-67, cleaved-PARP positive cell rate result of a preferred embodiment 5 of the present invention. Detailed Implementation

[0031] The following description, with reference to the accompanying drawings, illustrates several preferred embodiments of the present invention to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0032] Example 1: Acquisition of seminal plasma metabolic fingerprint data using nanoparticle-enhanced laser desorption / ionization time-of-flight mass spectrometry

[0033] Step 1: Instrument and reagent preparation: matrix-assisted laser desorption / ionization time-of-flight mass spectrometry, seminal plasma sample, deionized water, matrix (iron oxide nanoparticles); The preparation method of the iron oxide nanoparticles is as follows: 1) Dissolve ferric chloride hexahydrate (FeCl3·6H2O, 0.6g) and trisodium citrate dihydrate (0.15g) in ethylene glycol (25mL) and use sonication to aid dissolution; 2) Add anhydrous sodium acetate (0.96g), mix well, and then transfer to a high-pressure reactor lined with polytetrafluoroethylene; 3) Heat treat at 200℃ for 10 hours; 4) The product was washed repeatedly with alternating ethanol and ultrapure water, and dried at 60°C to obtain iron oxide nanoparticles; The key performance parameters of the iron oxide nanoparticles are: Crystal structure: Magnetite (Fe3O4, JCPDS: 99-0073); Particle morphology: Nanoscale rough surface morphology; Hydrodynamic diameter: approximately 200 nm (measured by dynamic light scattering DLS); Surface charge: Zeta potential approximately -25.3 ± 3.1 mV; Characteristic absorption peak: 355nm.

[0034] Step 2: Dilute the seminal plasma sample 20 times with deionized water;

[0035] Step 3: Prepare a matrix solution of 1 mg / mL using deionized water with iron oxide nanoparticles;

[0036] Step 4: Sample preparation was performed on the mass spectrometry target plate. 2 μL of each diluted seminal plasma sample was spotted and dried at room temperature.

[0037] Step 5: Prepare the matrix on the mass spectrometry target plate, spotting 2 μL of each matrix solution and drying at room temperature;

[0038] Step 6: Collect seminal plasma metabolic fingerprint data using a laser desorption / ionization time-of-flight mass spectrometer with a laser wavelength of 355 nm and a frequency of 1000 Hz. Each sample is collected for approximately 30 seconds.

[0039] Example 2: Construction of Seminal Plasma Metabolic Fingerprint Database and Preprocessing of Metabolic Fingerprint Data

[0040] 1. Metabolic fingerprint data were collected from 475 seminal plasma samples (190 healthy controls, 128 F-SPGF group, and 157 C-SPGF group) using the method described in Example 1, and a metabolic fingerprint database was constructed. Figure 1 ); 2. The metabolic fingerprint data of 475 seminal plasma samples were preprocessed, including data resampling, spectral smoothing, baseline correction, feature extraction, peak matching and missing value filling, resulting in 317 metabolic features.

[0041] Example 3: Construction and Validation of an Overall Diagnostic Model for SPGF Based on 317 Metabolic Features

[0042] Using 317 metabolic features obtained from the preprocessing in Example 2 as input variables, and healthy controls (HC) and SPGF groups (including F-SPGF and C-SPGF) as grouping variables, a metabolic fingerprint diagnostic model was constructed on the discovery cohort using a neural network (NN) algorithm. The model parameters were optimized through 5-fold cross-validation. Evaluation on the independent validation set (n=94) showed that the model's AUC for diagnosing SPGF was 0.924, with a sensitivity of 75.68%, a specificity of 92.98%, and an accuracy of 86.17%, providing a more accurate technical means for the overall clinical diagnosis of SPGF.

[0043] Example 4: Identification of metabolic biomarkers and their application in SPGF typing diagnosis

[0044] Step 1: Preparation of instruments and reagents: Fourier transform ion cyclotron resonance mass spectrometry, liquid chromatography-mass spectrometry;

[0045] Step 2: Based on the 317 metabolic features obtained from the preprocessing in Example 2, statistical analysis was performed on the metabolic features of two groups of samples: 128 F-SPGF and 157 C-SPGF. The following criteria were used for screening: p < 0.05 (t-test between F-SPGF and C-SPGF groups, corrected by FDR), |log2(fold change)| > 0.3 (logarithm of the ratio of the mean intensity of the two groups of features), and LASSO score > 0.3 (LASSO regression model based on 317 features, with SPGF classification as the dependent variable). Finally, 13 metabolic features were obtained. After further excluding matrix interference peaks and exogenous substance peaks, 6 metabolic features were determined, with m / z values ​​of 137.071, 138.067, 154.096, 170.07, 210.907, and 240.8831, respectively. The precise molecular weights (<10 ppm, Table 1) of these six metabolic features were measured using Fourier transform ion cyclotron resonance mass spectrometry and liquid chromatography-mass spectrometry.

[0046] Table 1. Information on 6 metabolic biomarkers

[0047] Step 3: Based on the precise molecular weight of these 6 metabolic characteristics, determine the corresponding 6 molecular formulas (CH4N2O, C...). 10 H 26 N4, C5H9NO2, C6H 13NO2, C2H7NO3S, and C4H7NO4 were identified in the Human Metabolome Database (HMDB) as urea, spermine, proline, leucine, taurine, and aspartic acid (Table 1). These six metabolites showed significant differences between the F-SPGF and C-SPGF groups (p < 0.05, Table 2).

[0048] Table 2. Differential expression of 6 metabolic biomarkers in F-SPGF and C-SPGF

[0049] Step 4: From 317 metabolic features, 36 significantly different features were obtained through statistical screening (p<0.05 and |log2(fold change)|>0.3). A subtyping diagnostic model was established using a neural network algorithm. The seminal plasma metabolic fingerprint data of the sample to be tested was input into the diagnostic model, and the output result indicated whether the sample was F-SPGF or C-SPGF. The subtyping diagnostic model was evaluated on an independent validation set (n=60) with an AUC of 0.877 (95% CI: 0.780-0.973), a sensitivity of 84.00%, a specificity of 82.86%, and an accuracy of 83.33%. Step 5: Using the peak intensities of the above 6 metabolites as features, a neural network (NN) algorithm was used to construct a subtyping diagnostic model for F-SPGF and C-SPGF. Evaluation on the validation set (n=60) showed that the model AUC=0.814 (95%CI: 0.742-0.847), which was significantly better than the traditional clinical hormone markers FSH (AUC=0.527), LH (AUC=0.636) and testosterone (AUC=0.657), as shown in Table 3.

[0050] Table 3. Diagnostic performance of SPGF subtyping using a combination of 6 metabolic biomarkers and traditional hormone biomarkers.

[0051] Example 5: Validation of the biological function of metabolic biomarkers

[0052] 1. Cell Culture and CCK-8 Proliferation Assay: GC-1 spermatogonial cells were cultured in DMEM medium containing 10% fetal bovine serum, and then seeded into 96-well plates (2000 cells / well). After overnight equilibration, different concentrations (0.5 mM, 1 mM, 5 mM, 10 mM) of the test metabolic biomarkers (urea, spermine, proline, leucine, taurine, aspartic acid) were added, and the cells were treated for 96 hours. Then, 10 μL of CCK-8 reagent was added, and the absorbance was measured at 450 nm to assess the cell proliferation rate. 2. In vitro culture of human testicular tissue: Testicular tissue with normal spermatogenesis was obtained from patients with obstructive azoospermia (OA) and cut into 8mm sections. 3 Tissue fragments were placed on a (1.5% w / v) agarose gel (solvent: 1×TAE buffer) scaffold and cultured at 34°C and 5% CO2 for 14 days. Metabolic biomarkers were added to the culture medium to a final concentration of 5 mM. 3. Immunofluorescence staining and histological evaluation: After 14 days of culture, the tissue was fixed in 4% PFA, embedded in paraffin, and sectioned (5 μm). Ki-67 immunofluorescence staining (proliferation marker) and cleaved-PARP staining (apoptosis marker) were performed. HE staining was used to assess the integrity of seminiferous tubules (score 1-4). Results analysis: From Figure 2 Leucine and proline significantly promoted GC-1 cell proliferation (p<0.001), increased the rate of Ki-67 positive cells (p<0.001), decreased the rate of cleaved-PARP positive cells (p<0.001), and maintained tissue integrity (p<0.05) in human testicular tissue. Urea and spermine, on the other hand, significantly inhibited GC-1 cell proliferation (p<0.05), increased GC-1 cell apoptosis (p<0.001), and impaired tissue integrity.

[0053] The technical solution of this invention has significant practicality and industrialization prospects, mainly reflected in the following aspects:

[0054] 1. Outstanding clinical diagnostic value

[0055] In terms of overall SPGF diagnosis: the metabolic fingerprint model developed in this invention demonstrates excellent diagnostic performance (AUC=0.924), significantly superior to currently commonly used single biomarkers in clinical practice. In particular, the model's sensitivity is 75.68%, specificity is 92.98%, and accuracy is 86.17%, providing a more accurate technical means for the clinical diagnosis of SPGF.

[0056] In terms of subtyping diagnosis: the combination of six metabolic biomarkers in this invention demonstrated excellent diagnostic ability for subtyping F-SPGF and C-SPGF (AUC=0.814), significantly superior to FSH (AUC=0.527), LH (AUC=0.636), and testosterone (AUC=0.657). This performance is of great significance for improving treatment decisions for SPGF patients, as accurate subtyping can predict the success rate of microsurgical sperm retrieval and guide individualized treatment plans.

[0057] 2. The detection technology has obvious advantages.

[0058] This invention employs nanoparticle-enhanced laser desorption / ionization mass spectrometry for seminal plasma metabolic fingerprint analysis, offering significant advantages such as simple sample pretreatment, rapid analysis speed (approximately 30 seconds per sample), low sample consumption (approximately 100 nL), and low cost. This high-throughput detection method is more suitable for large-scale clinical screening applications, significantly improving detection efficiency and reducing medical costs.

[0059] 3. Solid industrialization foundation

[0060] This invention has been validated based on a large-scale clinical sample of 475 cases, including 190 healthy controls and 285 SPGF patients (128 with F-SPGF and 157 with C-SPGF). The large sample size and precise grouping give the research results strong statistical significance and clinical application value. Furthermore, the experimental methods are easily standardized and the operating procedures are easily translating into clinical testing products.

[0061] 4. Broad market application prospects

[0062] In view of the high incidence of male infertility, the severity of SPGF, and the limitations of current diagnostic methods, the seminal plasma metabolic marker detection scheme of the present invention has great market potential. It can be used as a standalone diagnostic product or integrated with existing hormone detection systems to broaden the application scope of existing products.

[0063] In summary, this invention has significant advantages in clinical application value, technological advancement, and industrialization foundation, possessing promising industrialization prospects and promotional value. The preferred embodiments of this invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of this invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of this invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A combination of seminal plasma metabolic biomarkers for the classification and diagnosis of spermatogenesis disorders in male infertility, characterized in that, The seminal plasma metabolic biomarker combination includes urea, spermine, proline, leucine, taurine, and aspartic acid; The screening method for the combination of seminal plasma metabolic biomarkers includes the following steps: Step 1: Metabolic detection was performed on seminal plasma samples from the F-SPGF group, C-SPGF group, and healthy control group using nanoparticle-enhanced laser desorption / ionization time-of-flight mass spectrometry to obtain seminal plasma metabolic fingerprint data; The method for obtaining the seminal plasma metabolic fingerprint data is as follows: 1) Prepare a matrix solution of 1 mg / mL using iron oxide nanoparticles with deionized water; 2) Dilute the seminal plasma sample 20 times with deionized water, spot 2 μL of the diluted seminal plasma sample onto the mass spectrometry target plate, allow it to air dry at room temperature, add 2 μL of the matrix solution, and allow it to air dry again at room temperature to obtain the seminal plasma sample for mass spectrometry analysis; 3) Seminal plasma metabolic fingerprint data were acquired using a laser desorption / ionization time-of-flight mass spectrometer with a laser wavelength of 355 nm and a frequency of 1000 Hz. Each sample was acquired for approximately 30 seconds to obtain seminal plasma metabolic fingerprint data. Step 2: Perform data preprocessing on the seminal plasma metabolic fingerprint data collected from the F-SPGF group, C-SPGF group, and healthy control group to obtain the target metabolic characteristics; Step 3: Screening biomarkers from target metabolic features: LASSO regression model was used for machine learning screening. The model was trained by optimizing the parameter λ and seminal plasma metabolic features with a model score > 0.3 were retained. At the same time, combined with statistical analysis, p < 0.05 and |log2(fold change)| > 0.3 were used as the screening criteria for significant differences. Finally, after excluding matrix peaks and exogenous substances, 6 target metabolic biomarkers were obtained.

2. The combination of seminal plasma metabolic biomarkers for the classification and diagnosis of male infertility spermatogenesis disorders as described in claim 1, characterized in that, The data preprocessing in step two includes data resampling, spectral smoothing, baseline correction, feature extraction, peak alignment, and missing value imputation.

3. The combination of seminal plasma metabolic biomarkers for the classification and diagnosis of male infertility spermatogenesis disorders as described in claim 1, characterized in that, The metabolic biomarker combination contains leucine and proline, which regulate spermatogenesis by promoting GC-1 cell proliferation, reducing GC-1 cell apoptosis, and maintaining testicular tissue integrity; urea and spermine, on the other hand, regulate GC-1 cell proliferation, increase GC-1 cell apoptosis, and impair testicular tissue integrity.

4. The combination of seminal plasma metabolic biomarkers for the classification and diagnosis of male infertility spermatogenesis disorders as described in claim 1, characterized in that, The method for validating the function of the metabolic biomarker combination includes the following steps: a. The GC-1 spermatogonial cell line was cultured in DMEM medium containing 10% fetal bovine serum, and then seeded into 96-well plates at a rate of 2000 cells / well. After overnight equilibration, the target metabolic biomarkers were added, and the cells were treated for 96 hours. Then, 10 μL of LCK-8 reagent was added, and the absorbance was measured at 450 nm to assess the cell proliferation rate. b. Obtain testicular tissue with normal spermatogenesis from patients with obstructive azoospermia, and cut it into 8mm sections. 3 Fragments; tissue fragments were placed on agarose gel scaffolds, and the metabolic biomarkers to be tested were added and cultured at 34°C and 5% CO2 for 14 days; c. After 14 days of culture, the tissue was fixed, embedded, and sectioned. Ki-67 immunofluorescence staining and cleaved-PARP staining were performed to calculate the positive cell rate. After HE staining, the integrity of seminiferous tubules was assessed according to a 1-4 point scale.

5. The application of the combination of seminal plasma metabolic biomarkers as described in claim 1 in the diagnosis of male infertility and spermatogenesis disorders.

6. The application as described in claim 5, characterized in that, The method includes a rapid detection method for spermatogenesis disorders in male infertility. Specifically, it involves taking the 317 metabolic features obtained from the preprocessing in step two of claim 1, using a healthy control group and an SPGF group including the F-SPGF group and the C-SPGF group as grouping variables, constructing a diagnostic model on the detection queue using a neural network algorithm, and inputting the seminal plasma metabolic fingerprint data of the sample to be tested into the diagnostic model to output the detection results.

7. The rapid detection method for male infertility spermatogenesis disorders as described in claim 6, characterized in that, The diagnostic model achieved an AUC of 0.924, a sensitivity of 75.68%, a specificity of 92.98%, and an accuracy of 86.17% on the independent validation set.

8. The application of the combination of seminal plasma metabolic biomarkers as described in claim 1 in the classification and diagnosis of male infertility spermatogenesis disorders.

9. The application as described in claim 8, characterized in that, The rapid detection method for male infertility spermatogenesis disorder typing specifically uses the peak intensity of the six metabolic biomarkers in claim 1 as features, and uses a neural network algorithm to construct a typing diagnostic model for F-SPGF and C-SPGF. The seminal plasma metabolic fingerprint data of the sample to be tested is input into the diagnostic model, and the typing detection result of the sample to be tested is F-SPGF or C-SPGF.

10. The rapid detection method for male infertility spermatogenesis disorder classification as described in claim 9, characterized in that, The subtyping diagnostic model achieved an AUC of 0.814 on the independent validation set.

Images