Blood free DNA methylation markers for differentiating lung nodule malignancy and application thereof

Abstract

The present application relates to the field of biomedical technology, and particularly relates to a blood free DNA methylation marker for identifying lung nodule benignity and malignancy and application thereof.The nucleotide sequence of the methylation marker comprises a nucleotide sequence as shown in at least one of SEQ ID NO.1-32 or a complementary sequence thereof, or a variant having 95% homology with the sequence as shown in at least one of SEQ ID NO.1-32 or a complementary sequence thereof and having the same methylation site.The present application is based on methylation sequencing data of lung nodule benign and malignant samples, and a marker with a significant difference in methylation level between benign lung nodules and malignant lung nodules is screened, which can be used for effectively identifying lung nodule benignity and malignancy, has high sensitivity and specificity, and is suitable for early identification and diagnosis of lung nodules.

Description

Technical Field

[0001] This invention relates to the field of biomedical technology, specifically to a blood-free DNA methylation marker for differentiating between benign and malignant pulmonary nodules and its application. Background Technology

[0002] Lung cancer is the leading cause of both incidence and mortality in my country, posing a serious threat to the health and lives of its citizens. According to statistics from the World Health Organization, in 2022, my country saw over 1 million new cases of lung cancer and nearly 740,000 deaths. Clinical data shows that approximately 70% of patients are diagnosed at an advanced stage, completely losing the opportunity for radical treatment. Therefore, early diagnosis is crucial for improving the prognosis and survival rate of lung cancer patients. In particular, accurately distinguishing between benign and malignant nodules during lung nodule screening directly determines the effectiveness of clinical treatment plans and their implementation.

[0003] Currently, low-dose spiral computed tomography (LDCT) has been widely used in lung cancer screening, effectively improving the detection efficiency of pulmonary nodules. However, this method still has significant limitations: relying solely on imaging characteristics for judgment results in a high false-positive rate, easily leading to overdiagnosis and unnecessary invasive procedures, which not only increases the physical and mental burden on patients but also wastes medical resources; at the same time, its accuracy in distinguishing between benign and malignant pulmonary nodules is limited, posing a high risk of misdiagnosis.

[0004] Histopathological examination, as the gold standard for diagnosing benign and malignant pulmonary nodules, is undoubtedly accurate. However, this method is highly dependent on invasive procedures such as bronchoscopy, percutaneous puncture, or surgery, which have drawbacks including significant trauma, high testing costs, and strong operational dependence. This makes it difficult to meet the clinical needs of large-scale screening and long-term follow-up for repeatable testing in high-risk groups of lung cancer. Therefore, developing a highly sensitive, specific, non-invasive, and convenient auxiliary diagnostic technology has become an urgent need to improve the early differentiation of benign and malignant pulmonary nodules and to improve the current state of lung cancer diagnosis and treatment.

[0005] In recent years, the rapid development of liquid biopsy technology, especially the detection methods based on plasma cell-free DNA (cfDNA) methylation biomarkers, has brought new breakthroughs to the field of non-invasive early screening for lung cancer. DNA methylation is a key epigenetic abnormality in the occurrence and development of tumors. Its abnormal signals usually precede lesions visible on imaging, and it has unique advantages such as high sensitivity, strong specificity, and stable tissue origin. cfDNA, derived from genetic material released during tumor cell apoptosis or necrosis, can non-invasively capture early abnormal signals of tumors through peripheral blood samples, enabling dynamic monitoring of tumors. Currently, numerous studies have confirmed that predictive models based on cfDNA methylation biomarkers combined with machine learning algorithms can effectively improve the accuracy and generalization ability of lung cancer screening, providing a feasible technical approach for differentiating benign and malignant lung nodules. Summary of the Invention

[0006] The purpose of this invention is to provide a methylation biomarker for detecting benign and malignant pulmonary nodules and its application. This invention distinguishes between patients with benign and malignant pulmonary nodules by detecting the methylation level of a methylation biomarker in plasma, thereby achieving a more accurate and lower-cost non-invasive and precise diagnosis of pulmonary nodules.

[0007] In order to achieve the above-mentioned objectives of the present invention, the following technical solution is adopted: In a first aspect, the present invention provides a methylation marker for differentiating between benign and malignant pulmonary nodules, said methylation marker comprising at least one of the following: A1. The target genes of the methylation markers include any one or more combinations of the following genes: SLC9C2, TNN, MYOG, HLX, RNF144A, POU3F3, DDX18, HOXD10, HOXD10, HOXD4, CACNA2D2, ADRA2C, COMMD8, ROPN1L, CDH6, SMURF1, GIMAP1, GIMAP1, CALB1, GSDMC, GPR20, NAP1L4, FGF6, SLC10A2, C14orf144, IMP3, SOX8, COG7, HSD3B7, BAIAP2, FAM69C, and LGI4. A2. The methylation sites of the methylation markers are located at chr1:173544806:173545005, chr1:175048818:175049017, chr1:203044797:203045069, chr1:221068017:221068391, chr2:7405011:7405271, chr2:104994933:104995132, chr2:117829233:117829496, chr2:1769767. 42:176976941, chr2:176977014:176977213, chr2:177024578:177024777, chr3:50480507:50480847, chr4:3732379:37 32612, chr4:47427871:47428070, chr5:10520555:10520754, chr5:30864371:30864570, chr7:98668773:98668972, chr7 :150417220:150417419、chr7:150417657:150417856、chr8:91215112:91215311、chr8:130291114:130291313、chr8:14 2313446:142313645, chr11:3028477:3028743, chr12:4554831:4555030, chr13:104117176:104117375, chr14:1048593 27:104859526, chr15:75941416:75941615, chr16:1065279:1065478, chr16:23424782:23425029, chr16:30991774:30992058, chr17:79024681:79024880, chr18:72092947:72093146, chr19:35622037:35622236 and any one or more combinations of the upstream and downstream 500bp regions; A3. The nucleotide sequence of the methylation marker includes at least one nucleotide sequence as shown in SEQ ID NO. 1-32 or a complementary sequence of at least one nucleotide sequence as shown in SEQ ID NO. 1-32, or a variant having 95% homology with the sequence shown in at least one SEQ ID NO. 1-32 or its complementary sequence and having the same methylation site; or the DNA methylation haplotype covered in the region of the sequence shown and the abundance of the DNA methylation haplotype.

[0008] Optionally, the target gene of the methylation marker is selected from any one or more of the following genes: HLX, RNF144A, POU3F3, DDX18, HOXD4, SMURF1, GIMAP1, CALB1, GSDMC, C14orf144, IMP3, COG7, and BAIAP2.

[0009] Optionally, the methylation marker is selected from any one or more combinations of the RNF144A, POU3F3, DDX18, HOXD4, SMURF1, GIMAP1, GSDMC, C14orf144, IMP3, and COG7 genes.

[0010] Optionally, the methylation marker is selected from any one or more combinations of the HLX, POU3F3, DDX18, CALB1, and BAIAP2 genes.

[0011] Optionally, the methylation sites of the methylation markers are located at chr1:173544806:173545005, chr1:175048818:175049017, chr1:203044797:203045069, chr1:221068017:221068391, chr2:7405011:7405271, chr2:104994933:104995132, chr2:117829233:117829496, chr2:1769767. 42:176976941, chr2:176977014:176977213, chr2:177024578:177024777, chr3:50480507:50480847, chr4:3732379:37 32612, chr4:47427871:47428070, chr5:10520555:10520754, chr5:30864371:30864570, chr7:98668773:98668972, chr7 :150417220:150417419、chr7:150417657:150417856、chr8:91215112:91215311、chr8:130291114:130291313、chr8:14 2313446:142313645, chr11:3028477:3028743, chr12:4554831:4555030, chr13:104117176:104117375, chr14:10485932 7:104859526, chr15:75941416:75941615, chr16:1065279:1065478, chr16:23424782:23425029, chr16:30991774:30992058, chr17:79024681:79024880, chr18:72092947:72093146, chr19:35622037:35622236 and any one or more combinations of at least 500bp upstream and downstream regions.

[0012] Optionally, the methylation sites of the methylation markers are located at chr2:7405011:7405271, chr2:104994933:104995132, chr2:117829233:117829496, chr2:177024578:177024777, chr7:98668773:98668972, chr7 :150417657:150417856, chr8:130291114:130291313, chr14:104859327:104859526, chr15:75941416:75941615, chr16:23424782:23425029 and any one or more combinations of the upstream and downstream 500bp regions.

[0013] Optionally, the methylation sites of the methylation markers are located in any one or more combinations of chr1:221068017:221068391, chr2:104994933:104995132, chr2:117829233:117829496, chr8:91215112:91215311, chr17:79024681:79024880 and their upstream and downstream 500bp regions.

[0014] Optionally, the nucleotide sequence of the methylation marker includes at least one nucleotide sequence as shown in SEQ ID NO. 1-32 or a complementary sequence of at least one nucleotide sequence as shown in SEQ ID NO. 1-32, or a variant having 95% homology with the sequence shown in SEQ ID NO. 1-32 or its complementary sequence and having the same methylation site; or the DNA methylation haplotype covered in the region of the shown sequence and the abundance of the DNA methylation haplotype.

[0015] Optionally, the nucleotide sequence of the methylation marker includes at least one of the nucleotide sequences shown in SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:10, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:28, or a complementary sequence to at least one of the nucleotide sequences shown in SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:10, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:28, or a sequence complementary to, or a sequence of, SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:10, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:25, SEQ ID NO:26. SEQ ID NO:28 At least one of the sequences shown or their complementary sequences having 95% homology and having the same methylation sites as variants; or the DNA methylation haplotypes covered in the region of the sequence shown and the abundance of the DNA methylation haplotypes.

[0016] Optionally, the nucleotide sequence of the methylation marker includes at least one of the nucleotide sequences shown in SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:19, SEQ ID NO:30, or a complementary sequence to at least one of the nucleotide sequences shown in SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:19, SEQ ID NO:30, or a variant having 95% homology with the sequence shown in at least one of SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:19, SEQ ID NO:30 or its complementary sequence and having the same methylation site; or the DNA methylation haplotype covered in the region of the shown sequence and the abundance of the DNA methylation haplotype.

[0017] Optionally, a methylation biomarker for detecting benign or malignant pulmonary nodules, the methylation biomarker comprising at least the sequence shown in SEQ ID NO. 6 or its complementary sequence to the nucleotide sequence shown in SEQ ID NO. 6 or a variant having 95% homology with the sequence shown in SEQ ID NO. 6 or its complementary sequence and having the same methylation site; or the DNA methylation haplotype covered in the region of the shown sequence and the abundance of the DNA methylation haplotype.

[0018] In some embodiments, the target gene or the regulatory region of the target gene is selected from any one or more of the sequences shown in SEQ ID No. 1-32 or their complementary sequences.

[0019] In some embodiments, the number of methylation sites included in SEQ ID No. 1-32 is shown in the table below:

[0020] In this invention, the methylation biomarkers differentiate between patients with benign and malignant pulmonary nodules based on the methylation status or level of CpG island-containing regions, fragments, and methylation sites involved in this invention. The methylation level of the methylation biomarkers can be obtained through various detection methods known in the art, such as, but not limited to, obtaining methylation levels through next-generation sequencing.

[0021] The methylation biomarkers described in this invention can be used alone or in combination as lung cancer-related methylation molecular markers for the detection or auxiliary detection or identification of benign or malignant lung nodules and / or lung cancer.

[0022] In a second aspect, the present invention provides primers or probes for detecting the methylation markers, wherein the primers target the nucleotide sequence containing the methylation markers as the target sequence for specific amplification of the target sequence; and the probes specifically capture the nucleotide sequence containing the methylation markers.

[0023] In a third aspect, the present invention provides a kit for detecting benign or malignant pulmonary nodules, the kit containing reagents for detecting the methylation markers. Preferably, the reagents include reagents for detecting the methylation level of the methylation markers.

[0024] Optionally, the reagents include reagents for detecting the methylation level of the methylation markers. Preferably, the reagents include reagents used in any one or a combination of PCR amplification, quantitative real-time PCR, digital PCR, liquid phase microarray, next-generation sequencing, third-generation sequencing, bisulfite sequencing, whole-genome methylation sequencing, and methylation microarray.

[0025] Optionally, for example, the detection reagent may be selected from the following: bisulfite and its derivatives, PCR buffer, polymerase, dNTP, primers, probes, methylation-sensitive or insensitive restriction endonucleases, enzyme digestion buffers, fluorescent dyes, fluorescence quenchers, fluorescent reporter agents, exonucleases, alkaline phosphatase, internal standards, controls, etc.

[0026] Further, the kit includes reagents for detecting the methylation status or level of at least one methylation site of any one or more of the following target genes: SLC9C2, TNN, MYOG, HLX, RNF144A, POU3F3, HOXD10, HOXD4, CACNA2D2, ADRA2C, COMMD8, ROPN1L, CDH6, GIMAP1, CALB1, GSDMC, GPR20, NAP1L4, FGF6, SLC10A2, C14orf144, IMP3, SOX8, COG7, HSD3B7, BAIAP2, FAM69C, and LGI4 genes, in any combination of one or more. The at least one methylation site of the target gene is selected from methylation sites within the target gene or in the regulatory region of the target gene.

[0027] In some embodiments, the kit includes reagents for detecting the methylation status or level of at least one methylation site of any one or more of the following target genes: HLX, RNF144A, POU3F3, DDX18, HOXD4, SMURF1, GIMAP1, CALB1, GSDMC, C14orf144, IMP3, COG7, and a combination of one or more of the BAIAP2 genes.

[0028] In some embodiments, the kit includes reagents for detecting the methylation status or level of at least one methylation site of any one or more of the following target genes: RNF144A, POU3F3, DDX18, HOXD4, SMURF1, GIMAP1, GSDMC, C14orf144, IMP3, and any combination of one or more of the COG7 genes.

[0029] In some embodiments, the kit includes reagents for detecting the methylation status or level of at least one methylation site in the regulatory region of the target gene.

[0030] In some embodiments, the at least one methylation site comprises at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or even at least ten or more methylation sites.

[0031] In some implementations, the regulatory region of the target gene is any continuous region of 100bp-400bp, preferably 150bp-300bp, and more preferably 200bp-250bp.

[0032] In a fourth aspect, the present invention provides a risk assessment system for benign or malignant pulmonary nodules, the risk assessment system comprising: The data acquisition module is used at least to acquire sample datasets; The sequencing module is used, at least, to obtain sequencing data; The data alignment module is at least used to align the sequencing data with the sequence of the methylation marker as described in claim 1 or 2, and to determine the methylation result of the marker in the sequencing data based on the alignment result; The result determination module is used at least to calculate the predicted score threshold through statistical model analysis and determine whether the sample to be tested is a benign or malignant lung nodule.

[0033] Optionally, the sample in the data acquisition module is a liquid sample. Optionally, the liquid sample is mammalian blood or other biologically derived liquid samples, such as peripheral blood, serum, plasma, ascites, urine, cerebrospinal fluid, sputum, saliva, etc. The mammals include rats, mice, and humans; preferably humans. In one embodiment, the sample is a fine-needle aspiration biopsy or plasma. The sample includes genomic DNA or cfDNA. In one embodiment, the sample preferably refers to plasma cfDNA. cfDNA refers to circulating cell-free DNA or cell-free DNA, degraded DNA fragments released into the plasma.

[0034] Optionally, the statistical method for statistical model analysis in the result determination module includes a machine learning model; the machine learning model includes any one of principal component analysis, logistic regression analysis, nearest neighbor analysis, support vector machine, and neural network model; preferably, it is a logistic regression analysis model.

[0035] Optionally, the threshold in the result determination module is 0.524. A value greater than this is predicted as a malignant pulmonary nodule, and a value less than this is predicted as a benign pulmonary nodule.

[0036] In one embodiment, the present invention provides biomarkers for screening benign and malignant pulmonary nodules based on cfDNA methylation.

[0037] In this invention, the terms "benign" and "malignant" refer to the nature of the pulmonary nodules. The pulmonary nodules include solid, partially solid, or ground-glass nodules, preferably partially solid or ground-glass nodules. "Malignant" pulmonary nodules generally refer to pulmonary nodules with the potential to become cancerous.

[0038] In a fifth aspect, this invention provides a method for constructing a predictive assessment model for benign and malignant pulmonary nodules, comprising: The following steps are required: (a) Collect benign and malignant lung nodule samples, and divide them into training and test sets; (b) Extract cfDNA from the sample, construct a library, and sequence it; (c) Perform methylation transformation on the sequences, compare the data, and calculate the AMF value of the samples; (d) Construct a feature matrix from the sample data; build an algorithm model and screen out methylation markers of benign and malignant lung nodules based on the training set samples; (e) Validate the model's performance using test set samples; (f) Identify the methylation markers that will ultimately be used for the assessment of benign and malignant lung nodules.

[0039] Optionally, the algorithm model includes a machine learning model, which employs any one of principal component analysis, logistic regression, nearest neighbor analysis, support vector machine, and neural network models; preferably, it is a logistic regression model.

[0040] This invention uses regression analysis to correlate the methylation levels of selected biomarkers with the benign or malignant nature of nodules, constructing a regression model to obtain a risk assessment or diagnostic model for benign or malignant pulmonary nodules. A logistic regression machine learning model constructed using the methylation levels of 35 methylation biomarkers from this invention can be used to differentiate between benign and malignant pulmonary nodules.

[0041] In one implementation, a new machine learning model is constructed using 10 methylation biomarkers: SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:10, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:25, SEQ ID NO:26, and SEQ ID NO:28.

[0042] In one implementation, five methylation markers, SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:19, and SEQ ID NO:30, are selected to construct a machine learning model.

[0043] In one implementation, 32 methylation markers, SEQ ID NO.1-32, are selected to construct a machine learning model.

[0044] Furthermore, the present invention also provides an information data processing terminal, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: (a) Obtain the methylation level of at least one region of the SEQ ID NO.1-32 sequence in the sample to be tested; (b) A score is calculated by constructing a logistic regression diagnostic model; (c) The benign or malignant nature of the pulmonary nodules is determined based on the score.

[0045] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, performs the following steps: (a) Obtain the methylation level of at least one region of the SEQ ID NO.1-32 sequence in the sample to be tested; (b) The score is calculated by constructing a logistic regression diagnostic model; (c) The benign or malignant nature of the pulmonary nodules is determined based on the score.

[0046] A method for detecting benign or malignant pulmonary nodules for non-disease diagnostic purposes, comprising the following steps: S1: Detect the methylation level of at least one region of the SEQ ID NO.1-32 sequence in the sample; S2: The score is calculated based on the aforementioned risk assessment model for benign and malignant pulmonary nodules. S3: Determine the benign or malignant nature of lung nodules based on the scoring.

[0047] Specifically, a nodule is identified as malignant when the methylation level of the target sequence in the sample meets a certain threshold. For example, for a sample to be tested, if the score is greater than the threshold, the result is positive, indicating a malignant nodule; otherwise, it is negative, indicating a benign nodule. Optionally, the threshold is 0.524.

[0048] The beneficial effects of the present invention include, but are not limited to: This invention identifies 32 methylation biomarkers with high correlation to benign and malignant pulmonary nodules based on high-throughput methylation sequencing of plasma cfDNA. These biomarkers can effectively distinguish between benign and malignant pulmonary nodules with high sensitivity and specificity. Furthermore, the methylation levels of individual biomarkers in this invention have been verified to have good classification effects. These biomarkers can be used to establish a risk prediction and assessment model for benign and malignant pulmonary nodules / lung cancer, for the purpose of differentiating between benign and malignant pulmonary nodules.

[0049] This invention also establishes a risk prediction and assessment system based on the relationship between the methylation level of biomarkers and the benign or malignant nature of lung nodules. The predictive score of this system distinguishes between benign and malignant lung nodules. The model has the advantages of non-invasive detection, safe and convenient detection, high throughput, and high detection accuracy. Based on the methylation biomarkers obtained by this invention, detection costs can be effectively controlled while achieving good detection performance. Attached Figure Description

[0050] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a flowchart of the benign and malignant lung nodule identification technology based on methylation markers in an embodiment of the present invention; Figure 2 This is a box plot of the methylation levels of 32 methylation markers in the test set and training set in this embodiment of the invention; Figure 3 This is a box plot illustrating the methylation level of Seq ID NO:6 in the test and training sets in this embodiment of the invention; Figure 4 This is a scatter plot of the scores predicted by a machine learning model using 32 methylation markers in an embodiment of the present invention. Figure 5 This is the ROC curve of the machine learning prediction model for the methylation level of 32 methylation markers in the embodiments of the present invention; Figure 6 This is a scatter plot of the scores predicted using the random combination 1 machine learning model in this embodiment of the invention; Figure 7 This is the ROC curve of the machine learning model using random combination 1 in an embodiment of the present invention; Figure 8 This is a scatter plot of the scores predicted using the random combination 2 machine learning model in this embodiment of the invention; Figure 9 This is the ROC curve of the effect of the random combination 2 machine learning model in the embodiment of the present invention. Detailed Implementation

[0051] The present invention is described in detail below with reference to the embodiments, but the present invention is not limited to these embodiments. Unless otherwise specified, the raw materials and catalysts in the embodiments of the present invention are all purchased through commercial channels.

[0052] Example 1: Detection Method for Benign and Malignant Biomarkers of Lung Nodules Based on Methylation-Targeted Sequencing The inventors collected a total of 203 patients with pulmonary nodules, and all enrolled patients signed informed consent forms. These samples were divided into training and test sets according to a certain ratio. The training set was used to build the machine learning model described below, and the test set was used to test the model's performance. Sample information is shown in Table 1 below.

[0053] Table 1. Statistics of Sample Information from Training and Test Sets

[0054] This invention obtains methylation sequencing data of 32 biomarkers in plasma cfDNA using a methylation sequencing method. The specific process is as follows (see flowchart). Figure 1 ): 1. Extraction of plasma cfDNA samples A 2ml whole blood sample was collected from the patient using a streck blood collection tube. Plasma was separated by centrifugation within 3 days and then transferred to the laboratory. cfDNA was extracted using the QIAGEN QIAamp Circulating Nucleic Acid Kit according to the instructions.

[0055] 2. Sequencing and Data Preprocessing a) The library was sequenced at 150bp paired ends using an Illumina Nextseq 500 sequencer, with a sequencing volume of no less than 5M.

[0056] b) The Pear (v0.6.0) software merges the paired-end sequencing data of the same fragment (150bp) from the sequencer into a single sequence with a minimum overlap length of 20bp and a minimum length of 30bp after merging.

[0057] c) The merged sequencing data were de-adapted using Trim_galore v 0.6.0 and cutadapt v1.8.1 software. The adapter sequence “AGATCGGAAGAGCAC” was removed from the 5' end of the sequence, and bases with sequencing quality values ​​below 20 at both ends were removed.

[0058] 3. Sequencing data alignment The reference genome data used in this article came from the UCSC database (UCSC: HG19, http: / / hgdownload.soe.ucsc.edu / goldenPath / hg19 / bigZips / hg19.fa.gz).

[0059] a) First, HG19 was transformed into cytosine to thymine (CT) and adenine to guanine (GA) using Bismark software, and the transformed genomes were indexed using Bowtie2 software.

[0060] b) Perform CT and GA conversion on the preprocessed data as well.

[0061] c) Use Bowtie2 software to align the transformed sequences to the transformed HG19 reference genome. The minimum seed sequence length is 20, and mismatches in the seed sequence are not allowed.

[0062] 4. Calculation of the average methylation rate (AMF) for each sample For each HG19 CpG site in the target region, the methylation status of each site was obtained based on the alignment results above. The average methylation rate (AMF) of the target region was calculated. The formula for calculating AMF is as follows:

[0063] Where M is the total number of CpG sites in the target methylation region, i is one of the CpG sites, and N is the number of sites in the target methylation region. C,i N represents the number of reads sequenced as C at this CpG site (i.e., the number of methylated reads). T,i The number of reads sequenced as T for this CpG site (i.e., the number of unmethylated sequencing reads).

[0064] 5. Feature Matrix Construction a) Merge the AMF values ​​of each target region of each sample in the training set and test set into feature matrices for the training set and test set respectively, and handle missing values ​​for sites with a depth of less than 100.

[0065] b) Remove sites with a missing value ratio higher than 10%.

[0066] c) Use the KNN algorithm to learn a transformer for the training set, and use the transformer to impute missing data in the feature matrices of the training and test sets.

[0067] 6. Model Building and Prediction The model was trained using Logistic Regression in the training set, and its performance was validated on the test data (results are shown below). Figure 2 (As shown). The 32 methylation markers, or combinations of multiple methylation markers, can be used as methylation markers for differentiating benign from malignant pulmonary nodules. The genes associated with these 32 methylation markers refer to those located within the gene itself or within 100 kb upstream or downstream. Specific associated genes are shown in Table 2.

[0068] Table 2. Statistical analysis and tests of methylation levels of 32 methylation biomarkers in the training and test sets.

[0069] Taking Seq ID NO:6 as an example, this section details the methylation status of each sample in the training and test sets. Figure 3 This methylation marker showed significant differences in methylation between benign and malignant nodules in both the training and test sets.

[0070] Example 2: Machine learning model for 32 methylation markers In this embodiment, a logistic regression machine learning model is constructed using the methylation levels of 32 methylation markers to distinguish between benign and malignant lung nodules. The model is trained using samples from the training set in Example 1, and then tested using samples from the test set. The specific steps are as follows: a) Use the logistic regression model from the sklearn (V1.0.1) package in Python (V3.9.7): AllModel = LogisticRegression() b) Use the training set samples for training: AllModel.fit(Traindata, TrainPheno), where TrainData is the data in the training set, TrainPheno is the phenotype of the training set samples (1 for malignant nodules, 0 for benign nodules), and the relevant thresholds of the model are determined based on the training set samples.

[0071] c) Use the test set samples for testing: TestPred = AllModel.predict_proba(TestData)[:, 1], where TestData is the test set data and TestPred is the model's predicted score. Use this predicted score and the above threshold to determine the benign or malignant nature of the lung nodules.

[0072] The model prediction scores for the training and test sets are shown below. Figure 4 The figure shows a significant difference in model scores between benign and malignant pulmonary nodule samples. The ROC curve is shown below. Figure 5 The training set AUC was 0.93, and the test set AUC was 0.90. A threshold of 0.524 was set; nodules greater than this value were predicted as malignant, and those less than this value were predicted as benign. At this threshold, the sensitivity of the training set was 0.82, and the specificity was 0.94; the sensitivity of the test set was 0.71, and the specificity was 0.88 (see Table 3 for specific results).

[0073] Example 3: Machine Learning Model for Random Methylation Marker Combination 1 To verify the specific advantages of the combination of 32 methylation biomarkers, this embodiment selects 10 methylation biomarkers (Seq ID NO:5, Seq ID NO:6, Seq ID NO:7, Seq ID NO:10, Seq ID NO:16, Seq ID NO:18, Seq ID NO:20, Seq ID NO:25, Seq ID NO:26, and Seq ID NO:28) from all 32 methylation biomarkers to construct a new machine learning model.

[0074] The method for constructing the machine learning model is consistent with Example 2. The features used only include the methylation levels of the aforementioned 10 methylation markers. The prediction scores of this model on the training and test sets are shown below. Figure 6 ROC curves are shown below. Figure 7 The model showed a significant difference in prediction scores between malignant and benign pulmonary nodules in both the training and test sets. The AUC on the training set was 0.88, and the AUC on the test set was 0.85. A threshold of 0.575 was set, with nodules above this value predicted as malignant and those below as benign. At this threshold, the sensitivity on the training set was 0.80, and the specificity was 0.71; the sensitivity on the test set was 0.71, and the specificity was 0.75 (see Table 3 for detailed results).

[0075] Example 4: Machine learning model for random methylation marker combination 2 This embodiment selects 5 methylation markers from all 32 methylation markers to build a machine learning model. The selected methylation markers are: Seq ID NO:4, Seq ID NO:6, Seq ID NO:7, Seq ID NO:19, and Seq ID NO:30.

[0076] The method for constructing the machine learning model is the same as in Example 2. The features used only include the methylation levels of the five methylation markers mentioned above. The prediction scores of this model on the training and test sets are shown below. Figure 8 ROC curves are shown below. Figure 9 The model showed a significant difference in prediction scores between malignant and benign pulmonary nodules in both the training and test sets. The AUC on the training set was 0.81, and the AUC on the test set was 0.79. A threshold of 0.582 was set; nodules above this value were predicted as malignant, and those below were predicted as benign. At this threshold, the sensitivity on the training set was 0.82, and the specificity was 0.71; the sensitivity on the test set was 0.73, and the specificity was 0.71 (see Table 3 for detailed results).

[0077] Table 3. Performance of logistic regression classification models constructed with different combinations of methylation biomarkers

[0078] This invention contains 32 methylation markers for differentiating between benign and malignant pulmonary nodules. Machine learning models constructed based on different combinations of these methylation markers or the methylation levels of individual methylation markers can effectively distinguish between samples of malignant and benign pulmonary nodules, providing an important reference for the diagnosis of benign and malignant pulmonary nodules.

[0079] The above description is merely an embodiment of the present invention, and the scope of protection of the present invention is not limited to these specific embodiments, but is determined by the claims of the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the technical concept and principle of the present invention should be included within the scope of protection of the present invention.

Claims

1. A methylation marker for differentiating benign from malignant pulmonary nodules, characterized in that, The methylation marker includes at least one of the following: A1. The target genes of the methylation markers include any one or more combinations of the genes SLC9C2, TNN, MYOG, HLX, RNF144A, POU3F3, HOXD10, HOXD4, CACNA2D2, ADRA2C, COMMD8, ROPN1L, CDH6, GIMAP1, CALB1, GSDMC, GPR20, NAP1L4, FGF6, SLC10A2, C14orf144, IMP3, SOX8, COG7, HSD3B7, BAIAP2, FAM69C, and LGI4; A2. The methylation sites of the methylation markers are located at chr1:173544806:173545005, chr1:175048818:175049017, chr1:203044797:203045069, chr1:221068017:221068391, chr2:7405011:7405271, chr2:104994933:104995132, chr2:117829233:117829496, chr2:1769767. 42:176976941, chr2:176977014:176977213, chr2:177024578:177024777, chr3:50480507:50480847, chr4:3732379:37 32612, chr4:47427871:47428070, chr5:10520555:10520754, chr5:30864371:30864570, chr7:98668773:98668972, chr7 :150417220:150417419、chr7:150417657:150417856、chr8:91215112:91215311、chr8:130291114:130291313、chr8:14 2313446:142313645, chr11:3028477:3028743, chr12:4554831:4555030, chr13:104117176:104117375, chr14:1048593 27:104859526, chr15:75941416:75941615, chr16:1065279:1065478, chr16:23424782:23425029, chr16:30991774:30992058, chr17:79024681:79024880, chr18:72092947:72093146, chr19:35622037:35622236 and any one or more combinations of the upstream and downstream 500bp regions; A3. The nucleotide sequence of the methylation marker includes at least one nucleotide sequence as shown in SEQ ID NO. 1-32 or a complementary sequence of at least one nucleotide sequence as shown in SEQ ID NO. 1-32, or a variant having 95% homology with the sequence shown in at least one SEQ ID NO. 1-32 or its complementary sequence and having the same methylation site; or the DNA methylation haplotype covered in the region of the sequence shown and the abundance of the DNA methylation haplotype.

2. A methylation marker for detecting benign or malignant pulmonary nodules, characterized in that, The methylation marker includes at least one of the following: B1. The nucleotide sequence of the methylation marker includes the sequence shown in SEQ ID NO. 6 or its complementary sequence, or a variant having 95% homology with the sequence shown in SEQ ID NO. 6 or its complementary sequence and having the same methylation site; or the DNA methylation haplotype covered in the region of the sequence shown and the abundance of the DNA methylation haplotype; B2. The nucleotide sequence of the methylation marker comprises at least one of the nucleotide sequences shown in SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:19, SEQ ID NO:30, or a complementary sequence to at least one of the nucleotide sequences shown in SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:19, SEQ ID NO:30, or a variant having 95% homology with and having the same methylation site as the sequence shown in SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:19, SEQ ID NO:30 or its complementary sequence; or the DNA methylation haplotype covered in the region of the shown sequence and the abundance of the DNA methylation haplotype; B3. The nucleotide sequence of the methylation marker comprises the nucleotide sequence shown in at least one of SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:10, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:28, or a complementary sequence to, or a sequence of, at least one of SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:10, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:25, SEQ ID NO:26, SEQ ID NO:

28. NO:28 At least one variant of the sequence shown or its complementary sequence having 95% homology and having the same methylation site; or the DNA methylation haplotypes covered in the region of the sequence shown and the abundance of the DNA methylation haplotypes.

3. Primers or probes for detecting the methylation markers according to any one of claims 1 or 2, characterized in that, The primers target the nucleotide sequence containing the methylation marker for specific amplification of the target sequence; the probes specifically capture the nucleotide sequence containing the methylation marker.

4. A reagent kit for detecting benign or malignant pulmonary nodules, characterized in that, The kit contains reagents for detecting the methylation markers as described in any one of claims 1 or 2.

5. The reagent kit according to claim 4, characterized in that, The reagents include those for detecting the methylation level of the methylation markers according to any one of claims 1-2.

6. The reagent kit according to claim 4, characterized in that, The reagents include those used in any one or a combination of PCR amplification, quantitative real-time PCR, digital PCR, liquid-phase microarray, next-generation sequencing, third-generation sequencing, bisulfite sequencing, whole-genome methylation sequencing, and methylation microarray.

7. A system for assessing the risk of benign or malignant pulmonary nodules, characterized in that, The risk assessment system includes: The data acquisition module is used at least to acquire sample datasets; The sequencing module is used, at least, to obtain sequencing data; The data alignment module is at least used to align the sequencing data with the sequence of the methylation marker as described in claim 1 or 2, and to determine the methylation result of the marker in the sequencing data based on the alignment result; The result determination module is used at least to calculate the predicted score threshold through statistical model analysis and determine whether the sample to be tested is a benign or malignant lung nodule.

8. The system according to claim 7, characterized in that, The sample described in the data acquisition module is a liquid sample.

9. The system according to claim 7, characterized in that, The statistical methods used in the statistical model analysis described in the result determination module include machine learning models; the machine learning models include any one of principal component analysis, logistic regression analysis, nearest neighbor analysis, support vector machine, and neural network models; preferably, logistic regression analysis.

10. The use of the methylation marker of claim 1 or 2, or the primer or probe of claim 3, or the kit of any one of claims 4-6, or the system of any one of claims 7-9, in the preparation of a product for detecting benign or malignant pulmonary nodules.

Images