Marker combination, method, system and application for diagnosing colorectal cancer

Abstract

The invention discloses a marker combination, method and system for diagnosing colorectal cancer and application, and belongs to the technical field of cancer marker application. The marker combination disclosed by the invention can be used for judging whether a subject suffers from colorectal cancer or has the risk of suffering from colorectal cancer, and the performance is higher.

Description

[0001] Related patents This application is a divisional application of the Chinese invention patent application with application number 2024107041311, application date June 3, 2024, and invention name “Methods, systems, devices and media for screening markers for diagnosing cancer based on methylated cfDNA fragments”. Technical Field

[0002] The present invention relates to the field of cancer markers and application technologies, and in particular, to a marker combination, method, system and application for diagnosing colorectal cancer. Background Art

[0003] With the development of biomedical research, the demand for the detection of DNA methylation level, insert size and terminal sequence in the field of tumor liquid biopsy is becoming more and more urgent. DNA methylation level refers to the addition and removal of methyl groups on DNA molecules, which plays an important role in gene expression regulation, cell differentiation and disease development. Insert size usually refers to the size of DNA fragments obtained by interrupting the DNA molecules in the sample using ultrasound or enzyme cutting technology in the construction of the second-generation sequencing library, and the terminal sequence refers to the sequence of a certain length at both ends of the DNA molecule.

[0004] Currently, existing DNA methylation detection methods include bisulfite conversion and sequencing such as WGBS and RRBS. However, bisulfite conversion of DNA has two serious disadvantages, namely, great DNA loss and destruction of the original DNA fragment structure. As a result, the bisulfite conversion sequencing data of cfDNA cannot detect the size of the inserted fragment and the terminal sequence. Therefore, the current distribution of the size of the inserted fragment and the distribution of the terminal sequence are generally detected using WGS sequencing data, which makes the detection process cumbersome. Summary of the invention

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is as follows: The first aspect of the present invention provides a method for screening markers for diagnosing cancer based on methylated cfDNA fragments, comprising the following steps: S1, obtaining fragment sequence information and methylation information of cfDNA samples of cancer patients and cfDNA samples of normal persons, wherein the methylation information includes methylation position and methylation level; S2, analyzing the fragment sequence information and methylation information obtained in step S1 as follows to obtain a candidate marker level profile: (1) Dividing the whole human genome into different genomic windows according to length, counting the number of methylated cfDNA fragments mapped to each genomic window, and obtaining a methylation level spectrum of the genomic window based on the methylation information; (2) Based on the number of methylated cfDNA fragments of different lengths, obtain the frequencies of fragment features with different step lengths to obtain a fragment feature frequency spectrum; (3) Obtain the distribution frequencies of the terminal sequences of 4mer-6mer at the 5' end of the methylated cfDNA fragments to obtain a terminal sequence frequency spectrum. Thus, the candidate biomarker spectrum includes a genomic window methylation level spectrum, a fragment feature frequency spectrum, and a terminal sequence frequency spectrum. S3. Screen candidate biomarkers with significant differences between cancer patients and normal individuals based on the candidate biomarker level spectrum obtained in step S2, that is, obtain biomarkers for cancer diagnosis, including genomic windows, fragment features, and terminal sequences.

[0006] In the present invention, cfDNA is fully called "cell-free DNA", that is, cell-free DNA, and its sources include but are not limited to peripheral blood, cerebrospinal fluid, saliva, pleural fluid, ascites, urine, and feces.

[0007] In some embodiments of the present invention, the fragment sequence information and methylation information are first obtained by enriching methylated cfDNA fragments in the test sample by means of methylated enrichment protein enrichment, immunoprecipitation, or enzymatic transfer method, and further sequencing. This method can well retain the fragment features of cfDNA while capturing methylated cfDNA, so as to simultaneously detect the methylation level, insert fragment size, and terminal sequence of cfDNA subsequently.

[0008] In some embodiments of the present invention, between step S1 and step S2, a step of preprocessing the fragment information is further included, and the preprocessing includes: (1) Remove the adapter sequences; (2) Filter low-quality sequences with a quality value lower than Q15 for more than 40% of the bases; (3) Filter sequences containing more than 5 Ns; (4) Filter sequences with a length less than 30 (the sequences are too short); (5) Trim 4 bases at the fragment ends with an average quality <Q20.

[0009] In some embodiments of the present invention, a methylated standard and an unmethylated standard are added to both the cancer patient cfDNA sample and the normal individual cfDNA sample. In some specific embodiments of the present invention, the methylated standard is fully methylated positive ginseng λDNA; the unmethylated standard refers to fully unmethylated negative ginseng λDNA.

[0010] Furthermore, bowtie2-2.3.4.2 software was used to align the base sequences of the cleaned sequencing data to the human reference genome hg19 (GRCH37) and the λDNA reference genome to generate bam files, and the bam files were sorted according to the genome coordinates. Picard MarkDuplicates-2.18.25-SNAPSHOT was used to deduplicate the sorted bam files, and finally the paired reads were mapped to the reference genome and the reads with MAPQ>20 were screened.

[0011] According to the comparison results of fully methylated Yang ginseng λDNA and fully unmethylated Yin ginseng λDNA, the reaction specificity is calculated. Preferably, in step S1, the methylation sequencing data whose effective sequencing data amount is greater than the first preset threshold and whose reaction specificity is greater than the second preset threshold is selected, wherein the first preset threshold is 4G~9G, and the second preset threshold is 0.7~0.9. If there is no cfDNA sample of a cancer patient or no methylation sequencing data of a normal person cfDNA sample that meets this requirement, the library is rebuilt and sequenced, or the sample is re-obtained for library construction and sequencing.

[0012] In some embodiments of the present invention, in step S2, the genomic window methylation level profile is performed based on the following steps: The whole human genome is divided into different genomic windows according to length. Based on the methylation sequencing data, for each genomic window, the number of methylated cfDNA fragments in the genomic window is counted and standardized to obtain the methylation level in the genomic window.

[0013] In some specific embodiments of the present invention, the standardized methylation level is obtained using the following formula: .

[0014] There are millions of total methylated cfDNA fragments and the genomic window length is measured in kb.

[0015] In some specific embodiments of the present invention, the window division is performed according to a length of 300 bp, which can be divided into 10318991 genomic windows. Those skilled in the art can select different lengths for window division, or perform exhaustive division within the range of 1 to the genome size.

[0016] In the present invention, the "fragment feature" is also called the "insert fragment size feature", which refers to the division of methylated cfDNA fragments into different fragment intervals according to different lengths (step lengths). In some embodiments of the present invention, the step length is 2bp~10bp. If the step length is 2bp, the divided fragment intervals are 61-62bp, 63-64bp, ..., 399-400bp; if the step length is 3bp, the divided fragment intervals are 61-63bp, 64-66bp, ..., 397-399bp; if the step length is 10bp, the divided fragment intervals are 61-70bp, 71-80bp, ..., 391-400bp), and all methylated cfDNA fragments included in each fragment interval are defined as fragment features. For example, the fragment feature is: 61-65bp, including methylated cfDNA fragments with fragment lengths of 61bp, 62bp, 63bp, 64bp and 65bp. For another example, the fragment feature is: 74-75bp, including methylated cfDNA fragments with fragment lengths of 74bp and 75bp.

[0017] The fragment feature frequency refers to the ratio of the number of methylated cfDNA fragments in a fragment feature to the total number of methylated cfDNA fragments.

[0018] The terminal sequence frequency refers to the ratio of the number of methylated cfDNA fragments with the same terminal sequence to the total number of methylated cfDNA fragments.

[0019] In the present invention, the distribution frequencies of two lengths of terminal sequences at the 5' end of the methylated cfDNA fragments, 4mer (such as CCGT, AGTT, etc.) and 6mer (such as CCGATC, TCGGAT, etc.), are obtained.

[0020] In some embodiments of the present invention, in step S3, candidate markers with significant differences between cancer patients and normal persons are identified by statistical or machine learning methods.

[0021] In some specific embodiments of the present invention, in step S3, the machine learning method includes logistic regression, decision tree, random forest, support vector machine, naive Bayes, K nearest neighbor and neural network.

[0022] The second aspect of the present invention provides a system for screening markers for diagnosing cancer based on methylated cfDNA fragments, comprising: A methylation data input module is used to receive the fragment sequence information and methylation information of cfDNA samples of cancer patients and cfDNA samples of normal persons; A genome data storage module, used for storing human genome data; A data comparison module, which is respectively connected to the methylation data input module and the genomic data storage module, and is used for comparing the fragment information and the methylation information with the human genomic data; An analysis module, which is connected to the data comparison module, and is used for performing the following analysis to obtain a candidate biomarker level spectrum: (1) Divide the human whole genome according to length to obtain different genomic windows, count the number of methylated fragments aligned to each genomic window, and obtain a genomic window methylation level spectrum based on the methylation information; (2) Based on the number of methylated cfDNA fragments of different lengths, obtain the frequencies of fragment characteristics with different step lengths to obtain a fragment characteristic frequency spectrum; (3) Obtain the distribution frequency of the terminal sequences of 4mer-6mer at the 5' end of the methylated cfDNA fragments to obtain a terminal sequence frequency spectrum, A screening module, which is connected to the analysis module, and is used for screening candidate biomarkers with significant differences between cancer patients and normal people based on the candidate biomarker level spectrum.

[0023] In some embodiments of the present invention, the methylation data input module is further used for preprocessing the fragment sequence information, and the preprocessing includes: (1) Removing adapter sequences; (2) Filtering out low-quality sequences with a quality value lower than Q15 for more than 40% of the bases; (3) Filtering out sequences containing more than 5 Ns; (4) Filtering out sequences with a length less than 30 (the sequences are too short); (5) Trimming 4 bases at the fragment ends with an average quality <Q20.

[0024] The third aspect of the present invention provides a biomarker combination for diagnosing cancer, including biomarkers screened by using any method described in the first aspect of the present invention or any system described in the second aspect of the present invention.

[0025] The fourth aspect of the present invention provides another biomarker combination for diagnosing cancer, including: (1) At least one of the genomic window combinations composed of chr15:31775701-31776000, chr7:32467501-32467800, and chr8:24771301-24771600; (2) At least one of the fragment characteristic combinations composed of 165-166bp, 163-165bp, and 167-168bp; and (3) At least one of the terminal sequence combinations composed of ATGGGG, ATAGGC, and ATGAGG, Preferably, the cancer is colorectal cancer; Or include: (1) at least one of the genomic window combinations consisting of chr7:32467501-32467800, chr7:32467801-32468100, and chr20:21376801-21377100; (2) at least one of the fragment feature combinations consisting of 149-150 bp, 151-152 bp, and 153-154 bp; and (3) at least one of the terminal sequence combinations consisting of ATGAGC, ATGG, and ATAGCG, Preferably, the cancer is liver cancer; Or include: (1) at least one of the genomic window combinations consisting of chr20:43726801-43727100, chr6:107955901-107956200, and chr19:19650901-19651200; (2) at least one of the fragment feature combinations consisting of 279-280 bp, 277-279 bp, and 280-282 bp; and (3) at least one of the terminal sequence combinations consisting of CTAGGG, TTCAGC, and ATAGGC, Preferably, the cancer is pancreatic cancer; Or include: (1) at least one of the genomic window combinations consisting of chr7:32467501-32467800, chr6:107955901-107956200, and chr2:39187201-39187500; (2) at least one of the fragment feature combinations consisting of 163-164 bp, 165-166 bp, and 156-160 bp; and (3) at least one of the terminal sequence combinations consisting of AGGGAG, AGGGGA, and TGAAAC, Preferably, the cancer is gastric cancer; Or include: (1) at least one of the genomic window combinations consisting of chr3:51740701-51741000, chr9:140128201-140128500, and chr17:7670401-7670700; (2) at least one of the fragment feature combinations consisting of 267-268 bp, 259-260 bp, and 268-270 bp; and (3) at least one of the terminal sequence combinations consisting of GGGAAC, GGGAGT, and GGGAAG, Preferably, the cancer is lung cancer.

[0026] The fifth aspect of the present invention provides use of the detection reagent and / or device of the marker combination described in the third aspect or the fourth aspect of the present invention in the preparation of a kit for diagnosing cancer.

[0027] A sixth aspect of the present invention provides a computer device, comprising: Memory for storing computer programs; A processor is used to implement the steps of a method for screening markers for diagnosing cancer based on methylated cfDNA fragments as described in the first aspect of the present invention when executing the computer program.

[0028] The seventh aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of a method for screening markers for diagnosing cancer based on methylated cfDNA fragments as described in the first aspect of the present invention are implemented.

[0029] An eighth aspect of the present invention provides a system for diagnosing cancer, comprising the following modules: A marker data input module, used to input the level of each marker in the marker combination described in the third aspect or the fourth aspect of the present invention; The cancer judgment module is connected to the data input module and is used to judge whether the subject has cancer or has a risk of cancer according to the level of each marker in the marker combination.

[0030] In some embodiments of the invention, the marker combination comprises a genomic window, a fragment feature and / or an end sequence.

[0031] When the marker is a genomic window, its level refers to the methylation level, which is obtained based on the number of methylated cfDNA fragments mapped to the genomic window.

[0032] When a marker is a segment feature, its level refers to the segment feature frequency.

[0033] When the marker is a terminal sequence, its level refers to the terminal sequence frequency.

[0034] Furthermore, the inventor provides a computer-readable medium storing computer program instructions, wherein when the computer program instructions are executed by a processor, any of the above methods of the present invention is executed.

[0035] Furthermore, the inventors propose a device comprising: A memory for storing computer program instructions, and a processor for executing the computer program instructions, wherein when the computer program instructions are executed by the processor, the device runs any of the above methods of the present invention.

[0036] In the present invention, the cancer includes solid tumors and blood cancers, such as colorectal cancer, pancreatic cancer, prostate cancer, squamous cell carcinoma, basal cell carcinoma, adenocarcinoma, sweat gland carcinoma, sebaceous gland carcinoma, papillary carcinoma, papillary adenocarcinoma, cystadenocarcinoma, medullary carcinoma, bronchogenic carcinoma, hepatocellular carcinoma, bile duct carcinoma, choriocarcinoma, kidney cancer, cervical cancer, testicular cancer, lung cancer, melanoma; leukemia, such as acute lymphocytic leukemia and acute myeloblastic leukemia (myeloblast, promyeloblast, myelomonocytic, monocytic and erythroblastic leukemia); chronic leukemia (chronic myeloid (granulocytic) leukemia and chronic lymphocytic leukemia), etc.

[0037] Beneficial Effects of the Invention This method combines advanced methylated cfDNA sequencing technology and molecular biology methods, and is characterized by high throughput, high sensitivity and high accuracy.

[0038] In the method of the present invention, the methylated cfDNA fragments of the sample to be tested are first obtained by protein enrichment, immunoprecipitation or enzyme transfer, and the fragment characteristics of cfDNA can be well preserved while capturing the methylated cfDNA fragments, so as to subsequently detect the cfDNA methylation level, insert fragment size and terminal sequence at the same time. The present invention also provides a corresponding system, which includes a methylation data input module, a genome data storage module, a data comparison module, an analysis module and a screening module.

[0039] The method and system of the present invention can simultaneously detect cfDNA methylation levels, fragment characteristics and terminal sequences, provide new biomarkers and targets for disease diagnosis and treatment, and have broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 A schematic flow chart of the method for screening markers for diagnosing cancer based on methylated cfDNA fragment sequencing in Example 1 of the present invention is shown.

[0041] Figure 2 The distribution density of the insert fragment size obtained by the insert fragment size analysis in Example 1 of the present invention is shown.

[0042] Figure 3 A schematic diagram of a system for screening markers for diagnosing cancer based on methylated cfDNA fragment sequencing in Example 1 of the present invention is shown.

[0043] Figure 4 The box plots of the distribution of the most highly enriched methylation level feature chr15:31775701-31776000, fragment feature 165-166bp and terminal sequence feature ATGGGG in colorectal cancer and normal samples, as well as the ROC curves of the three features for distinguishing colorectal cancer and normal samples are shown. The AUC value of the methylation level feature chr15:31775701-31776000 for distinguishing colorectal cancer and normal samples is 0.845, the AUC value of the fragment feature 165-166bp for distinguishing colorectal cancer and normal samples is 0.773, and the AUC value of the terminal sequence feature ATGGGG for distinguishing colorectal cancer and normal samples is 0.778. The AUC value of the three features fused into one feature for distinguishing colorectal cancer and normal samples is 0.87.

[0044] Figure 5 The box plots of the distribution of the most highly enriched methylation level feature chr7:32467501-32467800, fragment feature 163-165bp and terminal sequence feature ATAGGC in colorectal cancer and normal samples, as well as the ROC curves of the three features for distinguishing colorectal cancer and normal samples are shown. The AUC value of the methylation level feature chr7:32467501-32467800 for distinguishing colorectal cancer and normal samples is 0.843, the AUC value of the fragment feature 163-165bp for distinguishing colorectal cancer and normal samples is 0.763, and the AUC value of the terminal sequence feature ATAGGC for distinguishing colorectal cancer and normal samples is 0.776. The AUC value of the three features fused into one feature for distinguishing colorectal cancer and normal samples is 0.862.

[0045] Figure 6 The box plots of the distribution of the most highly enriched methylation level feature chr8:24771301-24771600, fragment feature 167-168bp and terminal sequence feature ATGAGG in colorectal cancer and normal samples, as well as the ROC curves of the three features for distinguishing colorectal cancer and normal samples are shown. The AUC value of the methylation level feature chr8:24771301-24771600 for distinguishing colorectal cancer and normal samples is 0.826, the AUC value of the fragment feature 167-168bp for distinguishing colorectal cancer and normal samples is 0.762, and the AUC value of the terminal sequence feature ATGAGG for distinguishing colorectal cancer and normal samples is 0.771. The AUC value of the three features fused into one feature for distinguishing colorectal cancer and normal samples is 0.870. DETAILED DESCRIPTION

[0046] Unless otherwise indicated, implied from the context, or customary in the prior art, all parts and percentages in this application are based on weight, and the tests and characterization methods used are all synchronized with the filing date of this application. Where applicable, the contents of any patent, patent application or disclosure involved in this application are fully incorporated herein by reference, and their equivalent patent families are also introduced as references, especially the definitions of relevant terms in the art disclosed in these documents. If the definition of a specific term disclosed in the prior art is inconsistent with any definition provided in this application, the definition of the term provided in this application shall prevail.

[0047] In order to make the technical problems, technical solutions and beneficial effects solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments.

[0048] The following examples are used to demonstrate preferred embodiments of the present invention. It will be appreciated by those skilled in the art that the techniques disclosed in the following examples represent techniques discovered by the inventors that can be used to implement the present invention and therefore can be considered as preferred embodiments of the present invention. However, it will be appreciated by those skilled in the art based on this specification that many modifications may be made to the specific embodiments disclosed herein and still achieve the same or similar results without departing from the spirit or scope of the present invention.

[0049] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, and the disclosure and materials cited therein are hereby incorporated by reference.

[0050] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many technical equivalents to the specific embodiments of the invention described herein. Such equivalents are intended to be encompassed by the claims.

[0051] The experimental methods in the following examples are conventional methods unless otherwise specified. The instruments and equipment used in the following examples are conventional laboratory instruments and equipment unless otherwise specified; the experimental materials used in the following examples are purchased from conventional biochemical reagent stores unless otherwise specified.

[0052] Example 1 Screening of markers for cancer diagnosis based on methylated cfDNA fragments Combination Figure 1 This example describes in detail a method for screening markers for diagnosing cancer based on cfDNA methylation sequencing.

[0053] 1. Obtaining methylation sequencing data Mix 10 ng of cfDNA from human peripheral blood, 10 pg of fully methylated positive control λDNA, and 10 pg of fully unmethylated negative control λDNA, and perform methylation protein enrichment, immunoprecipitation, or enzymatic conversion sequencing using the Illumina NovaSeq 6000 sequencing platform.

[0054] Obtain the off-machine data after sequencing.

[0055] 2. Data preprocessing Use the fastp-0.20.0 software to perform the following preprocessing on the off-machine data: (1) Remove adapter sequences; (2) Filter out low-quality sequences with a quality value lower than Q15 for more than 40% of the bases; (3) Filter out sequences containing more than 5 Ns; (4) Filter out sequences with a length less than 30 (sequences are too short); (5) Trim 4 bases at the ends of the fragments with an average quality < Q20.

[0056] The data before and after preprocessing is shown in Table 1: Table 1 Comparison of data before and after preprocessing

[0057] The preprocessing results are shown in Table 2: Table 2 Preprocessing results

[0058] Next, use the command alignment software bowtie2 (v 2.3.4.2) to align the filtered sequence data with the human and λDNA reference genomes respectively to obtain the alignment files (SAM format), and use samtools (v 1.3.1) sort to sort the SAM files according to the genomic coordinates to generate BAM files, and then use the samtools index command to build an index; Then use the command picard (v 2.18.25) MarkDuplicates to remove the duplicate sequences generated by PCR; Further use the command samtools view to screen for reads where both paired reads are aligned to the reference genome and MAPQ > 20; Finally, perform statistics on the final alignment results, as shown in Table 3: Table 3 Comparison results

[0059] 3. Statistically analyze the effective sequencing data volume and reaction specificity rate of the cfDNA in the test samples According to the total number of reads after deduplication and screening compared to the human reference genome, the effective sequencing data volume of the sample cfDNA was calculated (69495052×150 / 1024 / 1024 / 1000=9.941G); according to the comparison results of fully methylated Yang ginseng λDNA and fully unmethylated Yin ginseng λDNA, the reaction specificity was statistically calculated (544928 / (544928+53034)= 0.911), as shown in Table 4: Table 4. Effective cfDNA sequencing data volume

[0060] 4. Insert fragment distribution analysis The insert size analysis was performed based on the valid data of the sample cfDNA. The length of each insert was calculated based on all the inserts after duplication removal and screening compared to the human reference genome. The distribution frequency of the insert sizes of different lengths was statistically analyzed. Some results are shown in Table 5. The distribution density of the insert size is shown in Figure 2 As shown: Table 5 Insert fragment distribution (partial)

[0061] Furthermore, the length of the cfDNA inserted fragment of the sample to be tested is divided into different fragment intervals with a step length of 2bp, 3bp, 4bp, 5bp, ..., 10bp (for example, if the step length is 2bp, the fragment intervals are 61-62bp, 63-64bp, ..., 399-400bp; if the step length is 3bp, the fragment intervals are 61-63bp, 64-66bp, ..., 397-399bp; if the step length is 10bp, the fragment intervals are 61-70bp, 71-80bp, ..., 391-400bp), and all the cfDNA fragments included in each fragment interval are defined as fragment features (inserted fragment size features), and the proportion of the number of cfDNA fragments in each fragment feature to the total number of fragments is calculated to obtain the fragment feature spectrum. The proportions of some fragment features are shown in Table 6: Table 6 Distribution of fragment features (partial)

[0062] 5. Terminal Sequence Analysis The terminal sequence analysis was performed based on the valid data of the cfDNA sample to be tested. The 4-6 base sequences at the 5' end of the inserted fragments were extracted based on all the inserted fragments after duplication removal and screening compared to the human reference genome. The distribution frequencies of the terminal sequences of 4mer and 6mer at the 5' end of the cfDNA were counted to obtain the terminal sequence characteristic spectrum. Some results are shown in Table 7: Table 7 Terminal sequence distribution (partial)

[0063] 6. Methylation Level Analysis According to the valid data of the sample cfDNA to be tested, the methylation level analysis is performed, and the methylation level of any window including but not limited to 300bp in the whole human genome is counted to obtain the methylation level characteristic spectrum. In this embodiment, the methylation level of the 300bp window of the CpG island is counted, as shown in Table 8: Table 8 Methylation levels in the 300bp genomic window of CpG island

[0064] Based on the above method, this embodiment also provides a system for screening markers for diagnosing cancer based on cfDNA methylation sequencing, such as Figure 3 As shown, including: A methylation data input module is used to receive methylation sequencing data of cfDNA samples of cancer patients and cfDNA samples of normal people and perform data preprocessing; A genome data storage module, used for storing human genome data; A data comparison module, connected to the methylation data input module and the genome data storage module, respectively, for comparing the methylation sequencing data of the cancer patient's cfDNA sample and the normal person's cfDNA sample with the human genome data; The analysis module is connected to the data comparison module and is used to perform the following analysis to obtain the candidate marker level spectrum: (1) Obtain the number of methylated cfDNA fragments in a 300 bp genomic window and obtain the methylation level spectrum of the genomic window; (2) Based on the number of methylated cfDNA fragments of different lengths, the frequency of the fragment features with a step length of 2-10 is obtained to obtain the fragment feature frequency spectrum; (3) Obtain the distribution frequency of the terminal sequences of 4mer and 6mer at the 5' end of the methylated cfDNA fragments and obtain the terminal sequence frequency spectrum. The screening module is connected to the analysis module and is used to screen candidate markers with significant differences between cancer patients and normal people based on the candidate marker level spectrum.

[0065] Example 2 Application of markers for diagnosing cancer based on methylated cfDNA fragments In order to verify the effectiveness of the method in Example 1 for identifying tumor diagnostic markers, the inventors collected blood samples from 181 colorectal cancer patients and 147 normal subjects, extracted peripheral blood cfDNA and obtained methylation sequencing data using the method in Example 1; the sequencing data was then preprocessed to obtain valid sequencing data, and the same method was further used to obtain the genomic window methylation level spectrum, fragment characteristic frequency spectrum and terminal sequence frequency spectrum of the samples.

[0066] The Wilcox rank sum test and 5-fold cross validation were used to calculate the AUC value to identify colorectal cancer-specific enriched features from genomic window features, fragment features, and terminal sequence features. The results are shown in Table 9.

[0067] Table 9 Features specifically enriched in colorectal cancer

[0068] As shown in Table 9, the genomic window feature chr15:31775701-31776000, the fragment feature 165-166bp and the terminal sequence feature ATGGGG have the highest AUC for distinguishing colorectal cancer and normal samples (e.g. Figure 4 Furthermore, the inventors used the average value of the minimum-maximum normalization of the three features to fuse the three features into one feature, and used it to improve the AUC value of distinguishing colorectal cancer and normal samples to 0.87, which is higher than the performance of distinguishing colorectal cancer and normal samples using the features of the three dimensions alone (as shown in Figure 4 as shown).

[0069] Similarly, the genomic window feature chr7:32467501-32467800, the fragment feature 163-165bp, and the terminal sequence feature ATAGGC had the second highest AUC for distinguishing colorectal cancer from normal samples (e.g. Figure 5 Furthermore, the inventors used the average value of the three features after minimum-maximum standardization to fuse the three features into one feature, and used it to improve the AUC value of distinguishing colorectal cancer and normal samples to 0.862, which is higher than the performance of distinguishing colorectal cancer and normal samples using the three-dimensional features alone (as shown in Figure 5 as shown).

[0070] Similarly, the genomic window feature chr8:24771301-24771600, the fragment feature 167-168bp, and the terminal sequence feature ATGAGG had the third highest AUC for distinguishing colorectal cancer from normal samples (e.g. Figure 6Furthermore, the inventors used the average value of the minimum-maximum normalization of the three features to fuse the three features into one feature, and used it to improve the AUC value of distinguishing colorectal cancer and normal samples to 0.870, which is higher than the performance of distinguishing colorectal cancer and normal samples using the features of the three dimensions alone (as shown in Figure 6 as shown).

[0071] Based on the above method, this embodiment also provides a system for diagnosing cancer, including the following modules: A marker data input module for inputting the level of a feature; The cancer judgment module is connected to the data input module and is used to judge whether the subject has cancer or has the risk of cancer according to the level of the feature.

[0072] Example 3 Performance of cancer markers based on methylated cfDNA fragments in other cancers In order to further verify the effectiveness of the method in Example 1 for identifying tumor diagnostic markers, the inventors collected blood samples from 50 liver cancer patients, 123 pancreatic cancer patients, 75 gastric cancer patients, 57 lung cancer patients and 142 healthy people, extracted peripheral blood cfDNA and used the method in Example 1 to obtain methylation sequencing data; then the sequencing data was preprocessed to obtain effective sequencing data, and the same method was further used to obtain the genomic window methylation level spectrum, fragment characteristic frequency spectrum and terminal sequence frequency spectrum of the samples.

[0073] The same calculation method as in Example 2 was used to identify the features specifically enriched in liver cancer, pancreatic cancer, gastric cancer, and lung cancer from the genomic window features, fragment features, and terminal sequence features. The results are shown in Tables 10 to 13.

[0074] Table 10 Features of liver cancer-specific enrichment

[0075] Table 11 Features specifically enriched in pancreatic cancer

[0076] Table 12 Gastric cancer-specific enriched features

[0077] Table 13 Features of lung cancer-specific enrichment

[0078] All documents mentioned in the present invention are cited as references in this application, just as each document is cited as reference individually. In addition, it should be understood that after reading the above teachings of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to this application.

Claims

1. A marker combination for diagnosing colorectal cancer, comprising: (1) at least one of the genomic window combinations consisting of chr15:31775701-31776000, chr7:32467501-32467800, and chr8:24771301-24771600; (2) at least one of the fragment feature combinations consisting of 165-166 bp, 163-165 bp, and 167-168 bp; and (3) At least one of the terminal sequence combinations consisting of ATGGGG, ATAGGC, and ATGAGG.

2. A method for diagnosing colorectal cancer, characterized in that: The following steps are involved: Obtaining the level of each marker in the marker combination of claim 1 of the subject; Whether the subject suffers from colorectal cancer or has a risk of suffering from colorectal cancer is determined based on the level of each marker in the marker combination.

3. A system for diagnosing colorectal cancer, characterized in that: Includes the following modules: A marker data input module, used to input the level of each marker in the marker combination of claim 1; The cancer judgment module is connected to the data input module and is used to judge whether the subject has colorectal cancer or has a risk of suffering from colorectal cancer according to the level of each marker in the marker combination.

4. Use of the detection reagent and / or device of the marker combination according to claim 1 in the preparation of a kit for diagnosing colorectal cancer.

Images