5hmC molecular marker for predicting sensitivity of sudden deafness treatment and application of 5hmC molecular marker

Abstract

The invention belongs to the technical field of biological medicines, and particularly relates to a 5hmC molecular marker for predicting sensitivity of sudden deafness treatment. The invention relates to a 5hmC molecular marker for predicting the sensitivity of sudden deafness treatment. The marker comprises coding genes of proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2. The 5hmC molecular marker provided by the invention can effectively predict the sensitivity of sudden deafness treatment, the AUC reaches up to 0.994, the molecular marker is utilized to predict the treatment result before treatment, patients with poor or invalid curative effects are screened out, comprehensive treatment is given in advance, and the prognosis of the patients is improved.

Description

Technical Field

[0001] The invention belongs to the field of medical biotechnology, and specifically relates to a 5hmC molecular marker for predicting sensitivity to treatment of sudden deafness and an application thereof. Background Art

[0002] Sudden deafness (abbreviated as sudden deafness) is an important cause of sensorineural hearing loss. The pathogenesis is still unclear, and its annual incidence is 5-20 / 100,000. At present, the treatment of sudden deafness is mainly based on drugs. The later the treatment is, the worse the effect is, so it is an acute disease that requires early treatment. The overall efficacy of sudden deafness is about 80%, but about 20% of patients still have no effect. The more severe the hearing loss, the worse the prognosis, and the greater the impact on the patient's quality of life. In recent years, the research on the prognosis of sudden deafness has gradually increased, and many factors that may affect the prognosis have been reported, including clinical characteristics, inflammatory indicators, coagulation function, genetics and other related factors, but these methods have limitations and cannot predict the treatment results. Therefore, there is an urgent need for a biomarker that can predict the efficacy of treatment, so as to intervene in advance for a large number of ineffective patients and improve the prognosis.

[0003] The pathological mechanism of sudden deafness is related to the damage of inner ear hair cells. Previous studies have found that epigenetic regulatory mechanisms play an important role in the development of the inner ear, mainly including the following three aspects: histone modification, DNA methylation and demethylation, and chromosome recombination. Among them, DNA methylation and demethylation can promote the growth and development of cochlear cells. 5-Hydroxymethylcytosine (5hmC) is considered to be a key molecule of epigenetic markers and a relatively stable and representative marker of DNA demethylation. Its enrichment content is significantly different under different body conditions, that is, the number of demethylated cytosines is different, but the DNA sequence has not changed. Based on this feature, researchers have been able to stably capture the changes in the number of 5hmC and widely use it for the diagnosis, treatment and efficacy prediction of various diseases. Studies have found that extracellular vesicle DNA (EVsDNA) is DNA encapsulated in extracellular vesicles in the blood, carrying epigenetic information from primitive cells, with high copy number and sensitivity, and good disease prediction effect. Therefore, the 5hmC marker in EVsDNA was used to study sudden deafness, and 5hmC markers with significantly different enrichment levels were found to predict treatment sensitivity and intervene in advance for ineffective patients. Summary of the invention

[0004] The purpose of the present invention is to provide a new detection marker and a new detection method for predicting the treatment sensitivity of sudden deafness, clarify the treatment effect of patients through 5hmC high-channel sequencing, provide precise and individualized treatment measures, and help improve the treatment effect of patients with sudden deafness.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] The present invention provides a 5hmC molecular marker for predicting the sensitivity of sudden deafness treatment. The 5hmC molecular marker includes coding genes of proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2.

[0007] Preferably, the coding gene sequences of the proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2 are respectively as shown in SEQ ID NO.1, SEQ ID NO.2, SEQ ID NO.3, SEQ ID NO.4, SEQ ID NO.5, SEQ ID NO.6, SEQ ID NO.7, SEQ ID NO.8, SEQ ID NO.9, SEQ ID NO.10 and SEQ ID NO.11; or, the nucleotide sequence of the coding gene of the protein is replaced, deleted or added with nucleotides to have a sequence with a similarity of more than 85% with the sequences shown in SEQ ID NO.1, SEQ ID NO.2, SEQ ID NO.3, SEQ ID NO.4, SEQ ID NO.5, SEQ ID NO.6, SEQ ID NO.7, SEQ ID NO.8, SEQ ID NO.9, SEQ ID NO.10 and SEQ ID NO.11.

[0008] The present invention also provides a kit for predicting the sensitivity of sudden deafness treatment, the kit comprising a reagent for detecting the expression level of the 5hmC molecular marker, the molecular marker comprising the coding genes of the proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2.

[0009] Preferably, the kit is a nucleic acid detection kit, which comprises primer pairs for amplifying the coding genes of proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2, respectively.

[0010] The present invention also provides a process for constructing a model for predicting the sensitivity of sudden deafness treatment, comprising the following steps:

[0011] (1) Samples from multiple patients with sudden deafness who had undergone chemotherapy were tested to construct a 5hmC library and obtain 5hmC sequencing fragments by high-throughput sequencing;

[0012] (2) aligning the 5hmC sequencing fragment obtained in step (1) with the human genome, obtaining the 5hmC-enriched region through a first filtration, and then obtaining the 5hmC molecular marker for predicting sudden deafness through identification and a second filtration;

[0013] (3) Identify effective and ineffective differentially modified 5hmC regions, perform the third filtering, and then use the selected features (parameters used: maxiter = 100, method = "lbfgs") to train a logistic regression CV model to obtain a prediction model; use the trained model to predict the patient's sudden deafness treatment sensitivity;

[0014] The first filtering includes: retaining unique non-repetitive matches to the human genome;

[0015] The identification includes: using MACS software to identify potential 5hmC-enriched regions, using parameters of macs 14-p1e-3-fBAM-g hs;

[0016] The second filtering includes: retaining peak regions that appear in more than 10 samples and are smaller than 1000 bp;

[0017] The identification of effective and ineffective differentially modified 5hmC regions includes: using the EdgeR package for identification, filtering thresholds of pvalue<0.05 and log-2FoldChange>0.5;

[0018] The third filtering includes: using the recursive feature elimination algorithm in Scikit-Learn to filter the DhMRs of the training cohort, with parameters: estimator=Logistic Regression CV(class_weight='balanced', cv=2, maxiter=1000), scoring='accuracy'.

[0019] Preferably, the input variable of the model is a high-throughput sequencing fragment of a 5hmC library.

[0020] More preferably, the 5hmC library is constructed from the patient's peripheral blood through terminal modification, PCR amplification and purification.

[0021] The present invention also provides a reagent for detecting the expression level of the 5hmC molecular marker in the preparation of a product for predicting the sensitivity of sudden deafness treatment, wherein the reagent comprises the coding genes of proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2.

[0022] Beneficial effects of the present invention:

[0023] This invention is the first to apply the efficacy prediction method of the 5hmC molecular marker to sudden deafness. There is no such research at home and abroad. This study is a prospective trial, and sampling before the patient's treatment can understand the treatment effect.

[0024] The detection method of the present invention is minimally invasive and causes very little trauma to the patient. Only a small amount of blood sample is taken when the patient is admitted to the hospital for routine examination. The detection accuracy is high, with an AUC value of up to 0.998, high sensitivity and specificity, and relatively reliable results. The treatment effect of a large number of patients with poor treatment effects can be predicted in advance, and targeted comprehensive treatment can be given in advance to improve the efficacy. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0026] Figure 1 A flowchart of the method for predicting sudden deafness screening and prognosis of the present invention;

[0027] Figure 2 The 5hmC enrichment of sudden deafness patients and normal hearing people in the embodiment of the present invention;

[0028] Figure 3 The heatmap comparison between the sudden deafness patients and the normal hearing people in the embodiment of the present invention;

[0029] Figure 4 This is a cluster analysis comparison between the sudden deafness patients and the normal hearing people according to the embodiment of the present invention.

[0030] Figure 5 The 5hmC enrichment of effective and ineffective people in the sudden deafness prognosis prediction method of the embodiment of the present invention;

[0031] Figure 6 The heatmap comparison of effective and ineffective populations in the sudden deafness prognosis prediction method according to an embodiment of the present invention;

[0032] Figure 7 The cluster analysis comparison of effective and ineffective populations in the sudden deafness prognosis prediction method according to the embodiment of the present invention;

[0033] Figure 8 It is the prediction efficiency of effective and ineffective populations in the sudden deafness prognosis prediction method of the embodiment of the present invention.

[0034] Fig. 9 The heatmap comparison of the effective and ineffective populations in the verification test of the embodiment of the present invention is shown;

[0035] Fig.10 The cluster analysis comparison of effective and ineffective populations in the verification test of the embodiment of the present invention;

[0036] Fig.11 It is the predictive efficacy of the effective and ineffective populations in the validation test of the embodiment of the present invention. DETAILED DESCRIPTION

[0037] The 5hmC molecular markers described in the present invention include genes encoding proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2, and the nucleotide sequences of the genes encoding the proteins are replaced, deleted or added with nucleotides to form sequences having a similarity of more than 85% with the sequences shown in SEQ ID NO.1, SEQ ID NO.2, SEQ ID NO.3, SEQID NO.4, SEQ ID NO.5, SEQ ID NO.6, SEQ ID NO.7, SEQ ID NO.8, SEQ ID NO.9, SEQ ID NO.10 and SEQ ID NO.11, respectively.

[0038] The present invention constructs a process for constructing a model for predicting the sensitivity of sudden deafness treatment, comprising the following steps:

[0039] (1) Samples from multiple patients with sudden deafness who had undergone chemotherapy were tested to construct a 5hmC library and obtain 5hmC sequencing fragments by high-throughput sequencing;

[0040] (2) aligning the 5hmC sequencing fragments obtained in step (1) with the human genome, retaining the unique non-repetitive matches with the human genome after the first filtering, obtaining the 5hmC-enriched regions, and then using MACS software to identify potential 5hmC-enriched regions, using the parameters of macs 14-p 1e-3-fBAM-g hs;

[0041] , the second filtering obtains the 5hmC molecular marker for predicting sudden deafness;

[0042] (3) Use the EdgeR package to identify, filter thresholds p value < 0.05 and log-2FoldChange > 0.5; identify effective and ineffective differentially modified 5hmC regions, and thirdly filter the DhMRs of the training cohort using the recursive feature elimination algorithm in Scikit-Learn, with parameters: estimator = Logistic Regression CV (class_weight = 'balanced', cv = 2, maxiter = 1000), scoring = 'accuracy'. Then use the selected features (parameters used: maxiter = 100, method = "lbfgs") to train the logistic regression CV model to obtain the prediction model; use the trained model to predict the patient's sudden deafness treatment sensitivity;

[0043] The input variables of the model are high-throughput sequencing fragments of the 5hmC library; the 5hmC library is constructed from the patient's peripheral blood through terminal modification, PCR amplification and purification.

[0044] The "patient", "subject" or "individual" described in the embodiments of the present invention may refer to an organism, and in some aspects, the subject may be a human. The subject providing the sample may include a population at potential disease risk or a population diagnosed with a disease. The disease described in the present invention refers to sudden deafness.

[0045] The 5hmC molecular marker of the present invention can provide a method for predicting the sensitivity of sudden deafness treatment. Only a small amount of peripheral blood of the subject needs to be drawn, and the treatment effect of the patient can be clarified through 5hmC high-throughput sequencing, providing accurate and individualized treatment measures. At the same time, gene drugs for the treatment of sudden deafness and detection products for the sensitivity of sudden deafness treatment, such as nucleic acid detection kits, can also be developed based on the 5hmC molecular marker.

[0046] In order to further illustrate the present invention, the technical solution provided by the present invention is described in detail below in conjunction with the accompanying drawings and embodiments, but they should not be construed as limiting the protection scope of the present invention.

[0047] The concentrated kit of the present invention was purchased from ZYMO, AMPure XP beads were purchased from Beckman, vegetarian beads were purchased from Life Technologies, and DBCO-peg4-biotin was purchased from Click Chemistry Tools, with the model being 4.5 mM stock in DMSO.

[0048] The production processes, experimental methods or detection methods involved in the embodiments of the present invention are all conventional methods in the prior art unless otherwise specified, and their names and / or abbreviations are all conventional names in the art. The experimental methods used in the following embodiments are all conventional methods unless otherwise specified, and can be specifically referred to the specific methods listed in the book "Molecular Cloning Experiment Guide (Fourth Edition)" (Author: MR Green, J. Sambrook), or carried out according to the kit and product instructions; the materials, reagents, etc. used in the following embodiments, unless otherwise specified, can be obtained from commercial channels.

[0049] Example 1 Differences in 5hmC enrichment between people with sudden deafness and normal hearing

[0050] The specific steps are as follows:

[0051] The flowchart of the prediction construction method of this embodiment is as follows Figure 1 shown.

[0052] 1. Acquisition of cfDNA and construction of 5hmC library for high-throughput sequencing

[0053] (1) According to the guidelines for the diagnosis and treatment of sudden deafness, 57 patients with sudden deafness and 58 healthy hearing people were selected. 4-8 ml of peripheral blood was collected from the subjects and the blood samples were centrifuged: the first centrifugation was 1350 g for 12 minutes at 4°C, and the supernatant was collected; the second centrifugation was 13500 g for 5 minutes at 4°C, and the supernatant was collected to obtain EVsDNA, which was transferred to cell cryopreservation tubes and stored at -80°C.

[0054] (2) The EVsDNA samples were end-repaired, 3'-adenylated using the KAPAHyper Prep Kit (KAPA Biosystems), and then ligated with an Erie-compatible adapter. The ligated EVsDNA was added to 25 μL of a solution containing 50 mM HEPES buffer (pH 8.0), 25 mM magnesium chloride, 100 μM UDP-6-N3-Glc, and 1 μM β-NEB enzyme (glycosylation) and placed at 37 °C for 2 h.

[0055] (3) Purify EVsDNA using a DNA cleaning and concentration kit. After the purified DNA was incubated with 1 μL of DBCO-peg4-biotin at 37°C for 2 hours, the DNA was purified again using a DNA cleaning and concentration kit. At the same time, 2.5 μL of streptavidin beads were directly added to 1× buffer (5 mM Tris pH 7.5, 0.5 mM EDTA, 1 M sodium chloride, 0.2% Tween20) and reacted at room temperature for 30 minutes. Finally, wash with buffer 8 times, 5 minutes each time. All binding and washing steps were performed at room temperature with gentle rotation.

[0056] (4) Resuspend in RNase-free water and perform 14-16 cycles of PCR amplification. PCR products are purified using AMPure XP beads. The concentration of the library is measured using a Qubit 3.0 fluorometer. Paired-end 39 bp high-throughput sequencing is performed on the NextSeq 500 platform.

[0057] 2. Mapping and Identification of 5hmC-Enriched Regions

[0058] FastQC (version 0.11.5) was used to assess sequence quality. Raw reads were aligned to the human genome using Bowtie2 and further filtered using SAMtools (version 1.3.1) (using parameters: SAMtools view -f2 -F 1548 -q 30 and SAMtoolsrmdup) to retain unique non-duplicate matches to the genome. 5hmC enriched regions were obtained by expansion and conversion to bedgraph format using bedtools, normalized to the total number of aligned reads, and then converted to big format using bedGraphToBigWig of the UCSC genome browser and visualized in the integrated genome viewer.

[0059] Potential 5hmC-enriched regions (hMRs) were identified using MACS (version 1.4.2) using the parameters macs14-p1e-3-fBAM-g hs. Peak calls were merged using bedtools, retaining only peak regions that appeared in more than 10 samples and were smaller than 1000 bp. Genomic regions that tended to show spurious signals were also filtered according to ENCODE. hMRs for each patient were generated by intersecting the individual peak call files with the merged peak files.

[0060] 3.5hmC feature selection

[0061] The train_test_ split in SciPythot-Learn (version 0.22.1) was used. Differentially modified 5hmC regions (DhMRs) of all patients and healthy subjects were identified using the EdgeR package (version 3.24.3) with filtering thresholds (p value < 0.05 and log-2FoldChange > 0.5) as Figure 2 , as shown in 3.

[0062] Figure 3 As shown in the heatmap, the red bar represents that the enrichment content of this marker is significantly different. The greater the difference, the darker the red. The difference includes up-regulation difference or down-regulation difference.

[0063] The recursive feature elimination algorithm (RFECV) in Scikit-Learn was used to further filter the DhMRs of the training cohort (parameters: estimator = Logistic Regression CV (class_weight = 'balanced', cv = 2, maxiter = 1000), scoring = 'accuracy'). Finally, the selected features (parameters used: maxiter = 100, method = "lbfgs") were used to train the logistic regression CV model (LR), and the principal component analysis (PCA) was obtained by cluster analysis. Figure 4 As shown in Figure 2, the trained LR model was used to screen the subjects for sudden deafness.

[0064] 4. Comparison of enriched regions and markers

[0065] The obtained 5hmC enriched regions were compared with 11 5hmC molecular markers, of which 7 markers were down-regulated compared with the enrichment levels of normal hearing people, namely: CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, and the other 4 markers were up-regulated compared with the enrichment levels of normal hearing people, namely: MFNG, ATP9B, EGF, EPHX2 (benchmark: p value <0.05 and log-2FoldChange>0.5). The results of this example show that the enrichment of 5hmC markers in sudden deafness and normal hearing people is significantly different.

[0066] Example 2 Prediction of treatment sensitivity of sudden deafness using 5hmC molecular markers

[0067] 1. Acquisition of cfDNA and construction of 5hmC library for high-throughput sequencing

[0068] (1) According to the diagnosis and treatment guidelines for sudden deafness, 4-8 ml of peripheral blood was drawn from 57 patients diagnosed with sudden deafness before treatment. The blood samples were centrifuged: the first centrifugation was at 1350 g for 12 minutes at 4°C, and the supernatant was collected; the second centrifugation was at 13500 g for 5 minutes at 4°C, and the supernatant was collected to obtain cfDNA, which was transferred to cell cryopreservation tubes and stored at -80°C.

[0069] The treatment plan for sudden deafness is: give the patient a 10-14 day drug treatment, mainly including hormones (dexamethasone), circulation-improving drugs (ginkgo leaf extract), nerve nutrition drugs (methylcobalamin), and fibrinogen-lowering drugs (batomoclazuma). After the treatment, recheck the hearing. An improvement of more than 15dB HL is effective, and less than 15dB HL is ineffective.

[0070] (2) The cfDNA samples were end-repaired, 3'-adenylated using the KAPAHyper Prep Kit (KAPA Biosystems), and then ligated with an Erie-compatible adapter. The ligated cfDNA was added to 25 μL of a solution containing 50 mM HEPES buffer (pH 8.0), 25 mM magnesium chloride, 100 μM UDP-6-N3-Glc, and 1 μM β-NEB enzyme (glycosylation) and placed at 37 °C for 2 hours.

[0071] (3) Purify cfDNA using a DNA cleaning and concentration kit. After the purified DNA was incubated with 1 μL of DBCO-peg4-biotin at 37°C for 2 hours, the DNA was purified again using a DNA cleaning and concentration kit. At the same time, 2.5 μL of streptavidin beads were directly added to 1× buffer (5 mM Tris pH 7.5, 0.5 mM EDTA, 1 M sodium chloride, 0.2% Tween20) and reacted at room temperature for 30 minutes. Finally, wash with buffer 8 times, 5 minutes each time. All binding and washing steps were performed at room temperature with gentle rotation.

[0072] (4) Resuspend in RNase-free water and perform 14-16 cycles of PCR amplification. PCR products are purified using AMPure XP beads. The concentration of the library is measured using a Qubit 3.0 fluorometer. Paired-end 39 bp high-throughput sequencing is performed on the NextSeq 500 platform.

[0073] 2. Mapping and Identification of 5hmC-Enriched Regions

[0074] FastQC (version 0.11.5) was used to assess sequence quality. Raw reads were aligned to the human genome using Bowtie2 and further filtered using SAMtools (version 1.3.1) (using parameters: SAMtools view -f2 -F 1548 -q 30 and SAMtoolsrmdup) to retain unique non-duplicate matches to the genome. 5hmC enriched regions were obtained by expansion and conversion to bedgraph format using bedtools, normalized to the total number of aligned reads, and then converted to big format using bedGraphToBigWig of the UCSC genome browser and visualized in the integrated genome viewer.

[0075] Potential 5hmC-enriched regions (hMRs) were identified using MACS (version 1.4.2) using the parameters macs14-p1e-3-fBAM-g hs. Peak calls were merged using bedtools, retaining only peak regions that appeared in more than 10 samples and were smaller than 1000 bp. Genomic regions that tended to show spurious signals were also filtered according to ENCODE. hMRs for each patient were generated by intersecting the individual peak call files with the merged peak files.

[0076] 3.5hmC feature selection

[0077] The train_test_ split in SciPythot-Learn (version 0.22.1) was used. The effective and ineffective differentially modified 5hmC regions (DhMRs) were identified using the EdgeR package (version 3.24.3) with filtering thresholds (p value < 0.05 and log-2FoldChange > 0.5).

[0078] The filtered raw data were statistically analyzed to show the differences in the expression intensities of the 11 5hmC molecular markers and to create a heat map such as Figure 5 , as shown in 6. Figure 6 Medium blue indicates a small difference in enrichment content, red indicates a large difference in enrichment content, and the darker the red, the greater the difference.

[0079] The recursive feature elimination algorithm (RFECV) in Scikit-Learn was used to further filter the DhMRs of the training cohort (parameters: estimator = Logistic Regression CV (class_weight = 'balanced', cv = 2, maxiter = 1000), scoring = 'accuracy'). Finally, the selected features (parameters used: maxiter = 100, method = "lbfgs") were used to train the logistic regression CV model (LR), and the principal component analysis (PCA) was obtained by cluster analysis. Figure 7 The trained LR model was used to predict the treatment outcomes of patients. Receiver operating characteristic (ROC) analysis was used to evaluate the performance of the model. The sklearn.mectrics module was used to calculate the area under the curve (AUC), the optimal cutoff point, and the Figure 8 shown.

[0080] 4. Comparison of enriched regions and markers

[0081] The obtained 5hmC enriched regions were compared with 11 5hmC molecular markers, and the marker enrichment levels were consistent with the prediction of whether the patient was effective or ineffective. The corresponding down-regulated molecular markers for ineffective patients were CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, and MCAT; the corresponding up-regulated markers for effective patients were MFNG, ATP9B, EGF, and EPHX2 (benchmarks: p value < 0.05 and log-2FoldChange > 0.5).

[0082] The results of this example show that the use of the 5hmC molecular marker of the present invention can provide a method for predicting the sensitivity of treatment of sudden deafness. Only a small amount of peripheral blood of the subject needs to be drawn, and the treatment effect of the patient can be clarified through 5hmC high-throughput sequencing, providing precise and individualized treatment measures. At the same time, gene drugs for treating sudden deafness can also be developed based on the 5hmC molecular marker.

[0083] Example 3 Validation of the 5hmC molecular marker for predicting treatment sensitivity of sudden deafness

[0084] 1. Acquisition of cfDNA and construction of 5hmC library for high-throughput sequencing

[0085] (1) According to the diagnosis and treatment guidelines for sudden deafness, 37 patients with sudden deafness different from those in Example 1-2 were selected. 4-8 ml of peripheral blood was drawn from the patients before treatment, and the blood samples were centrifuged: the first centrifugation was at 1350 g for 12 minutes at 4°C, and the supernatant was taken; the second centrifugation was at 13500 g for 5 minutes at 4°C, and the supernatant was taken to obtain cfDNA, which was transferred to a cell cryopreservation tube for storage at -80°C;

[0086] (2) The cfDNA samples were end-repaired, 3'-adenylated using the KAPAHyper Prep Kit (KAPA Biosystems), and then ligated with an Erie-compatible adapter. The ligated cfDNA was added to 25 μL of a solution containing 50 mM HEPES buffer (pH 8.0), 25 mM magnesium chloride, 100 μM UDP-6-N3-Glc, and 1 μM β-NEB enzyme (glycosylation) and placed at 37 °C for 2 hours.

[0087] (3) Purify cfDNA using a DNA cleaning and concentration kit. After the purified DNA was incubated with 1 μL of DBCO-peg4-biotin at 37°C for 2 hours, the DNA was purified again using a DNA cleaning and concentration kit. At the same time, 2.5 μL of streptavidin beads were directly added to 1× buffer (5 mM Tris pH 7.5, 0.5 mM EDTA, 1 M sodium chloride, 0.2% Tween20) and reacted at room temperature for 30 minutes. Finally, wash with buffer 8 times, 5 minutes each time. All binding and washing steps were performed at room temperature with gentle rotation.

[0088] (4) Resuspend in RNase-free water and perform 14-16 cycles of PCR amplification. PCR products are purified using AMPure XP beads. The concentration of the library is measured using a Qubit 3.0 fluorometer. Paired-end 39 bp high-throughput sequencing is performed on the NextSeq 500 platform.

[0089] 2. Verify 5hmC-enriched regions

[0090] Identify the enrichment of 11 5hmC molecular markers in the library (CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF, EPHX2). Obtain a heat map, such as Fig. 9 As shown, the principal component analysis (PCA) is obtained through cluster analysis, as shown in Fig.10 Receiver operating characteristic (ROC) analysis was used to evaluate the performance of the model. The area under the curve (AUC) and the optimal cutoff point were calculated using the sklearn.mectrics module, as shown in Fig.11 shown.

[0091] The results of this embodiment show that among the 37 samples, 17 were effective and 20 were ineffective after being treated with the treatment regimen of Example 2. The sample set was reselected to verify the results of the 5hmC molecular marker predicting the sensitivity of sudden deafness treatment. It was found that its accuracy was high, with an AUC of 0.994. 5hmC can be used as a molecular marker to predict the sensitivity of sudden deafness treatment.

[0092] Although the above embodiment describes the present invention in detail, it is only a part of the embodiments of the present invention, not all of the embodiments. People can also obtain other embodiments based on this embodiment without creativity, and these embodiments all fall within the protection scope of the present invention.

Claims

1. A 5hmC molecular marker for predicting sensitivity to treatment of sudden deafness, characterized in that: The 5hmC molecular markers include genes encoding proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2.

2. The molecular marker according to claim 1, characterized in that The coding gene sequences of the proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2 are respectively as shown in SEQ ID NO.1, SEQ ID NO.2, SEQ ID NO.3, SEQ ID NO.4, SEQ ID NO.5, SEQ ID NO.6, SEQ ID NO.7, SEQ ID NO.8, SEQ ID NO.9, SEQ ID NO.10 and SEQ ID NO.11; or, the nucleotide sequences of the coding genes of the proteins are replaced, deleted or added with nucleotides to have sequences with more than 85% similarity with the sequences shown in SEQ ID NO.1, SEQ ID NO.2, SEQ ID NO.3, SEQ ID NO.4, SEQ ID NO.5, SEQ ID NO.6, SEQ ID NO.7, SEQ ID NO.8, SEQ ID NO.9, SEQ ID NO.10 and SEQ ID NO.

11.

3. A kit for predicting the sensitivity of sudden deafness treatment, characterized in that: The kit includes reagents for detecting the expression level of 5hmC molecular markers, and the molecular markers include coding genes of proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2.

4. The kit according to claim 3, characterized in that: The kit is a nucleic acid detection kit, which includes primer pairs for respectively amplifying the coding genes of proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2.

5. A process for constructing a model for predicting sensitivity to treatment of sudden deafness, characterized in that: The following steps are involved: (1) Samples from multiple patients with sudden deafness who had undergone chemotherapy were tested to construct a 5hmC library and obtain 5hmC sequencing fragments by high-throughput sequencing; (2) aligning the 5hmC sequencing fragment obtained in step (1) with the human genome, obtaining the 5hmC-enriched region through a first filtration, and then obtaining the 5hmC molecular marker for predicting sudden deafness through identification and a second filtration; (3) Identify effective and ineffective differentially modified 5hmC regions, perform the third filtering, and then use the selected features (parameters used: maxiter = 100, method = "lbfgs") to train a logistic regression CV model to obtain a prediction model; use the trained model to predict the patient's sudden deafness treatment sensitivity; The first filtering includes: retaining unique non-repetitive matches to the human genome; The identification includes: using MACS software to identify potential 5hmC-enriched regions, using parameters of macs 14-p 1e-3-fBAM-g hs; The second filtering includes: retaining peak regions that appear in more than 10 samples and are smaller than 1000 bp; The identification of effective and ineffective differentially modified 5hmC regions includes: using the EdgeR package for identification, filtering thresholds of p value < 0.05 and log-2FoldChange > 0.5; The third filtering includes: using the recursive feature elimination algorithm in Scikit-Learn to filter the DhMRs of the training cohort, with parameters: estimator=Logistic Regression CV(class_weight='balanced', cv=2, maxiter=1000), scoring='accuracy'.

6. The model constructed according to the method of claim 5, characterized in that: The input variables of the model are high-throughput sequencing fragments of the 5hmC library.

7. The model according to claim 6, characterized in that: The 5hmC library is constructed from the patient's peripheral blood through terminal modification, PCR amplification and purification.

8. Use of a reagent for detecting the expression level of 5hmC molecular marker in the preparation of a product for predicting the sensitivity of sudden deafness treatment, characterized in that: The reagents include genes encoding proteins CDK5R1, TAF6L, MFSD9, CDH7, PLCH1, MARCHF7, MCAT, MFNG, ATP9B, EGF and EPHX2.

Images