Biomarker combination for early diagnosis or prediction of colorectal cancer and application

Abstract

The invention belongs to the field of colorectal cancer detection, and particularly relates to a biomarker combination for early diagnosis or prediction of colorectal cancer and application, the biomarker combination comprises one or more of Feline eukemia virus, Sinomonas sp. JC656, Triethyl Pho sphate and Arabic Acid, the biomarker combination is utilized to construct a colorectal cancer early diagnosis model, the model can be used for early diagnosis of colorectal cancer, and the colorectal cancer early diagnosis model can be used for early diagnosis of colorectal cancer. The kit has the advantages that the sensitivity is 88.9%, the specificity is 100%, the accuracy is 95.5%, the kit is economical and practical, the operation is simple and convenient, pain, discomfort and postoperative symptoms caused by enteroscopy invasive examination can be avoided, the false positive rate of colorectal cancer early diagnosis is obviously reduced, and the kit is suitable for clinical application.

Description

Technical Field

[0001] The present invention belongs to the field of colorectal cancer detection, and specifically relates to a biomarker combination and application for early diagnosis of colorectal cancer. Background Art

[0002] Colorectal cancer is one of the malignant tumors with the highest incidence rate in the world. The early symptoms of colorectal cancer are often not obvious or atypical. There may be changes in bowel habits, such as increased or decreased bowel movements, alternating diarrhea and constipation, etc. There may also be changes in stool characteristics, such as thinner stools, mucus or blood. These slight changes are easily ignored by patients. As the disease progresses, abdominal pain may occur, and the degree and nature of pain vary depending on the location and size of the tumor. Patients with colon cancer may experience dull pain or bloating in the abdomen, and patients with rectal cancer may have a feeling of tenesmus, that is, a feeling of incomplete bowel movements. If the tumor causes intestinal obstruction, symptoms of intestinal obstruction such as nausea, vomiting, abdominal distension, and cessation of flatulence and defecation may also occur. However, the above symptoms may be similar to other diseases, resulting in the disease not being discovered and treated in time.

[0003] Colorectal cancer screening is an important means of early detection of colorectal cancer and precancerous lesions. Current screening methods include fecal occult blood test, fecal DNA test, colonoscopy, sigmoidoscopy and CT colonography. If the result of the fecal occult blood test is positive, it indicates that there may be intestinal lesions and further examination is required, such as colonoscopy to clarify the cause of bleeding. The advantage of fecal DNA testing is that it is non-invasive and has high patient acceptance. However, its sensitivity and specificity vary depending on the detection technology and target gene. At present, this detection method is still being improved and the cost is relatively high. It is generally not used as the preferred screening method. It is usually used as a supplementary examination when the fecal occult blood test is positive or other screening methods indicate a risk. However, none of the above schemes can diagnose the disease and further examination is required for diagnosis. The disadvantage of colonoscopy is that it is an invasive examination, which may cause some discomfort to the patient, such as abdominal pain, abdominal distension, etc., and there is a certain risk of complications such as perforation and bleeding, but these risks are relatively low. Sigmoidoscopy can only examine part of the large intestine and will miss lesions located above the sigmoid colon. If lesions are found during sigmoidoscopy, a further colonoscopy is usually required to clarify the condition of the entire large intestine. CT colonography has a high accuracy in detecting larger polyps (diameter greater than 1 cm) and colorectal cancer. Its advantages are that it is non-invasive and the examination speed is relatively fast. However, its detection ability is limited for smaller lesions, and false positive results may occur, which requires further diagnosis in combination with other examinations.

[0004] With the development of science and technology, researchers have found that the occurrence of cancer is accompanied by global changes in metabolic status, which affects tumor tissues and the surrounding microenvironment and macroenvironment. People can use laboratory testing technology to detect proteins, genes and metabolic substances in blood, urine and tissues, and provide important information for disease prediction, diagnosis, post-rainfall evaluation and treatment monitoring.

[0005] The present invention screens specific blood microorganisms and metabolites as a biomarker combination for early diagnosis of colorectal cancer. The biomarker combination includes Feline leukemia virus, Sinomonas sp. JC656, Triethyl Phosphate and Arachidonic Acid, and is used for the diagnosis of colorectal cancer through models such as random forest and linear regression. The present invention utilizes the specificity based on high-throughput sequencing results and combines it with early diagnosis of colorectal cancer. It has high sensitivity and strong specificity, can avoid the pain, discomfort and postoperative complications caused by invasive colonoscopy, significantly reduce the false positive rate of early diagnosis of colorectal cancer, and can be widely used in the early diagnosis of colorectal cancer. Summary of the invention

[0006] In view of the above technical problems, the purpose of the present invention is to provide a combination of blood metabolites and microbial markers for early diagnosis of colorectal cancer and its application. Specifically, it includes the following contents:

[0007] In a first aspect, the present invention provides a biomarker combination for early diagnosis or prediction of colorectal cancer, wherein the biomarker combination includes one or more of Feline leukemia virus, Sinomonas sp. JC656, Triethyl Phosp hate, and Arachidonic Acid.

[0008] In a second aspect, the present invention provides a reagent for detecting the biomarker combination, wherein the reagent is used to detect the expression level of the biomarker combination.

[0009] In a third aspect, the present invention provides the use of the biomarker combination or the reagent in the preparation of a reagent, a kit or a chip for early diagnosis or prediction of colorectal cancer.

[0010] In a fourth aspect, the present invention provides the use of the biomarker combination or the reagent in constructing a colorectal cancer patient screening model.

[0011] In a fifth aspect, the present invention provides a kit for early diagnosis or prediction of colorectal cancer, wherein the kit comprises reagents or instruments for detecting the biomarker combination.

[0012] In a sixth aspect, the present invention provides a method for constructing an early diagnosis model for colorectal cancer, the construction method comprising: inputting the expression abundance data in the blood cfDNA and blood metabolite database into a random forest, performing machine learning with a generalized linear regression model, and constructing the prediction model; the sources of the expression abundance data in the blood cfDNA and blood metabolite database include plasma samples from colorectal cancer patients and plasma samples from normal people; the expression abundance data include the abundance data of the biomarker combination.

[0013] In a seventh aspect, the present invention provides a model for early diagnosis of colorectal cancer, the model comprising:

[0014] (1) Sample processing: extract the blood metabolite marker combination from the blood sample to be tested and perform LC-MS / MS full scan detection to obtain the abundance value of the marker; extract cfDNA from the plasma sample for metagenomic sequencing to obtain the microbiome abundance; (2) Model construction: mark the colorectal cancer samples confirmed by colonoscopy and pathological examination as 1, and mark the healthy population samples as 0. Input the abundance values ​​of the biomarker combination of different samples into the random forest to construct a model and predict the diagnostic results;

[0015] (3) Result determination: When the predicted value is 1, the diagnosis result of colorectal cancer is positive; when the predicted value is 0, the diagnosis result of colorectal cancer is negative.

[0016] In an eighth aspect, the present invention provides a model system for early diagnosis of colorectal cancer, comprising:

[0017] (1) a pre-input module, at least used to input the data to be diagnosed;

[0018] (2) a diagnostic module, at least used to diagnose the data to be diagnosed, which is executed by the model for early diagnosis of colorectal cancer according to claim 7;

[0019] (3) A display module, at least used to display the diagnosis result.

[0020] In a ninth aspect, the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the system of claim 8 implements the following steps, including: (1) collecting and / or inputting colorectal cancer diagnosis data;

[0021] (2) Substituting the diagnostic data into the model for calculation, and obtaining a conclusion on whether the patient has colorectal cancer;

[0022] (3) Output the conclusion of whether the patient has colorectal cancer.

[0023] In a tenth aspect, the present invention provides a computer-readable storage medium on which the computer device is stored.

[0024] The beneficial effects of the present invention are as follows: the present invention provides a colorectal cancer biomarker combination, the biomarker combination comprising one or more of feline leukemia virus, Sinomonas sp. JC656, triethyl phosphate and arachidonic acid; using the biomarker combination, a colorectal cancer early diagnosis model is constructed, the model can be used for early diagnosis of colorectal cancer, with a sensitivity of 88.9%, a specificity of 100%, and an accuracy of 95.5%, being economical and convenient to operate, being able to avoid the pain, discomfort and postoperative complications caused by invasive colonoscopy, significantly reducing the false positive rate of early diagnosis of colorectal cancer, and being suitable for clinical application. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 AUC vs marker curves of random forest model constructed from 325 markers using pROC; the curve shows that the model composed of 52 markers is the best.

[0026] Figure 2 AUC curves of the diagnostic model constructed using four fecal metabolites; RF represents the random forest model, and Glm represents the logistic regression model. DETAILED DESCRIPTION

[0027] The following examples are provided to facilitate a better understanding of the present invention, but are not intended to limit the present invention. The experimental methods in the following examples are conventional methods unless otherwise specified. The experimental materials used in the following examples are commercially available unless otherwise specified.

[0028] In a specific embodiment, sensitivity, specificity, accuracy and other combinations are used to describe the goodness and reliability of the detection method described in the present invention. Several terms used together with the description of sensitivity, specificity and accuracy include: true positive (TP), true negative (TN), false positive (FP), false negative (FN); wherein, if the patient is proved to have a disease and a given screening test also shows the presence of the disease, the test result is considered to be a true positive; if the patient is proved not to have a disease and a given screening test also shows the absence of the disease, the test result is considered to be a true negative; if the result of the screening test shows that a patient who does not actually have the disease has the disease, the test result is a false positive; if the result of the screening test shows that a patient who actually has the disease does not have the disease, the test result is a false negative.

[0029] Sensitivity = TP / (TP+FN) = number of true positive assessments / number of all positive assessments;

[0030] Specificity = TN / (TN+FP) = number of true negative assessments / number of all negative assessments;

[0031] Accuracy = (TN + TP) / (TN + TP + FN + FP) = number of correct evaluations / number of all evaluations.

[0032] 55 patients with colorectal cancer and adenoma and 39 healthy people were selected, and the marker combination provided by the present invention was used to screen the above cases. Based on the test results, the sensitivity, specificity and accuracy of the marker combination provided by the present invention for colorectal cancer detection were determined.

[0033] Example 1: Marker screening and model construction

[0034] 1. Blood sample collection

[0035] Prepare a 5 ml vacuum blood collection tube, collect the patient's blood, and immediately centrifuge it at 3000 rpm / min for 15 minutes. Pipette the supernatant into sterile centrifuge tubes and immediately store it at -80°C.

[0036] 2. Blood biomarker testing

[0037] 2.1 Extraction of cfDNA from blood

[0038] The serum / plasma cfDNA extraction kit is used, and the principle is a purification method based on high-binding magnetic particles. The sample is cleaved and digested step by step under the action of digestion solution and proteinase K to obtain DNA and RNA digestion solution. Magnetic particles and binding solution are added, and DNA or RNA will be adsorbed on the surface of magnetic particles, while impurities such as proteins are not adsorbed and removed. The particles adsorbed with nucleic acids are washed with washing solution to remove proteins and impurities, and washed with ethanol to remove salts; finally, DNA is eluted with elution solution EB.

[0039] (1) Take out the elution plate and add 40-50 μl of elution buffer EB to each well. Take out the required components of the kit and remove the sealing bag and sealing film.

[0040] (2) Take out sample plates A and B, and add 20 μl of proteinase K and 0.2 μg of carrier RNA to each well.

[0041] (3) Take out sample plates A and B, divide the samples into two parts, and add 250-350 μl of serum or plasma sample to each well.

[0042] (4) Place the 96-well magnetic cover into the cleaning plate 1.

[0043] (5) Turn on the machine, start the corresponding program IVD5432-F-96B, and place the 96-well plate into the instrument.

[0044] (6) After 50 minutes, take out the 96-well plate and store the product at -20 to 8°C.

[0045] 2.2 Library construction

[0046] (1) End repair: Prepare the reaction system in a sterile PCR tube.

[0047] (2) Mix the reaction system thoroughly.

[0048] (3) Place the PCR tube on the PCR instrument. The reaction program is shown in Table 5. Set the thermal cover of the PCR instrument to 75°C.

[0049] (4) Take out the required magnetic beads from the 4°C refrigerator in advance and equilibrate them to room temperature. Vortex and oscillate to mix well and equilibrate at room temperature for at least 30 minutes.

[0050] (5) Use Low-EDTA TE Buffer to dilute the Working Adapter.

[0051] (6) Prepare the reaction system on ice. Nuclease-Free Water, Ligation Buffer, and LigaseEnzymes II can be used to prepare a premix, but the Working Adapter needs to be added separately. Mix the prepared ligation reaction system thoroughly and centrifuge.

[0052] (7) Place the PCR tube on a PCR instrument without a heated cover.

[0053] (8) If there are strict requirements on the library size, fragment sorting can be performed after the adapter ligation reaction is completed.

[0054] (9) After ligation, add 88 μL (0.8×) AFTMagNGSDN AClean Beads to each sample in the 110 μL reaction system, vortex to mix, and incubate at room temperature for 5 min.

[0055] (10) Place the PCR tube on a magnetic rack for 2 min. After the solution becomes clear, remove the supernatant, being careful not to touch the magnetic beads.

[0056] (11) Add 200 μL of 80% ethanol to rinse the magnetic beads. Incubate for 30 seconds and remove the supernatant.

[0057] (12) Repeat step (10).

[0058] (13) Keep the PCR tube on the magnetic rack and use a 10 μL pipette to remove the ethanol remaining at the bottom of the tube. Open the tube cap and dry until no ethanol remains.

[0059] (14) Remove the PCR tube from the magnetic rack, add 21 μL Nuclease-Free Water to resuspend the magnetic beads, and let stand at room temperature for 1 min.

[0060] (15) Place the PCR tube on a magnetic rack for 2 min and transfer 20 μL of the supernatant to a new PCR tube.

[0061] (16) Prepare the following PCR reaction system: 20 ml of adaptor-ligated DNA, 25 μL of Gloria Nova HS2X PCR Mix for NGS, 5 μL of 10X PCR Primers, and a total volume of 50 μL.

[0062] (17) Mix thoroughly and centrifuge briefly.

[0063] (18) Place the PCR tube on a thermal cycler and set the thermal cover to 105°C.

[0064] (19) Add 50 μL (1×) AFTMag NGS DNA Clean Beads to the amplified product, vortex to mix, and incubate at room temperature for 5 min.

[0065] (20) Place the PCR tube on the magnetic stand for 2 min. After the solution is clear, remove the supernatant, being careful not to touch the magnetic beads. Add 200 μL 80% ethanol to rinse the magnetic beads. Incubate for 30 s and remove the supernatant.

[0066] (21) Incubate at room temperature for 5 min.

[0067] (22) The lid is dried until no ethanol remains.

[0068] (23) Remove the PCR tube from the magnetic stand, add 31 μL of Nuclease-Free Water to resuspend the magnetic beads, and let stand at room temperature for 1 min.

[0069] (24) The sample was placed on a magnetic rack for 2 min and 30 μL of the supernatant was transferred to a new PCR tube.

[0070] (25) The library was stored at −20°C to facilitate library quality testing and sequencing.

[0071] 2.3 Data processing and species annotation

[0072] Fastp (v0.23.0) was used for quality control to obtain valid data (CleanData). Megahit assembly software was used for assembly analysis. Salmon was used to analyze the abundance of Unigenes and calculate the relative abundance information of each gene in each sample.

[0073] The Unigenes were compared with the NR_meta library using the DIAMOND software, and the comparison results of each Unigene were selected for species classification with Evalue ≤ minimum Evalue * 10. The specific species annotation information of each classification level of the sequence, kingdom, phylum, class, order, family, genus, and species, was obtained using the MEGAN software.

[0074] 2.4 Metabolite extraction

[0075] (1) Take out the sample stored at -80°C, thaw it in an ice-water mixture, and transfer 80 μL of the sample into a 1.5 mL EP tube;

[0076] (2) Add 320 μL of protein precipitant methanol-acetonitrile (V:V = 2:1, containing mixed internal standard, 4 μg / mL) and vortex for 1 min;

[0077] (3) Ultrasonic extraction in an ice-water bath for 10 min and standing at -40°C overnight;

[0078] (4) Centrifuge for 10 min (12000 rpm, 4°C), draw 150 μL of the supernatant with a syringe, filter using a 0.22 μm organic phase pinhole filter, transfer to an LC injection vial, and store at -80°C until LC-MS analysis.

[0079] (5) Quality control samples (QC) were prepared by mixing equal volumes of the extracts from all samples. All extraction reagents were precooled at -20°C before use.

[0080] 2.5 Liquid chromatography-mass spectrometry analysis

[0081] (1) The analytical instrument used in the experiment was a liquid chromatography-mass spectrometry system consisting of a Waters ACQUITY UPLC I-Class plus / Thermo QE ultra-high performance liquid chromatography-tandem high-resolution mass spectrometer.

[0082] (2) Chromatographic column: ACQUITY UPLC HSS T3 (100 mm×2.1 mm, 1.8 μm); column temperature: 45°C; mobile phase: A-water (containing 0.1% formic acid), B-acetonitrile; flow rate: 0.35 mL / min; injection volume: 3 μL.

[0083] (3) Mass spectrometry conditions: spray voltage (3800 V for positive ions, -3000 V for negative ions), capillary temperature (320 °C), auxiliary gas heater temperature (350 °C), sheath gas flow rate (35), auxiliary gas flow rate (8), S-lens RF level (50), mass range (mass-to-charge ratio) (70–1050 Da), full scan mass spectrometry resolution (70,000), tandem mass spectrometry resolution (17,500), normalized collision energy / step normalized collision energy (10, 20, 40).

[0084] 2.5 Data processing

[0085] (1) Data preprocessing Before pattern recognition, the raw data were processed by the metabolomics processing software ProgenesisQI v3.0 for baseline filtering, peak identification, integration, retention time correction, peak alignment and normalization. The main parameters were as follows: precursor ion mass tolerance: 5 ppm (human metabolome database + lipid map) / 10 ppm (LuMet-animal + METLIN database); product ion mass tolerance: 10 ppm (human metabolome database + lipid map) / 20 ppm (LuMet-animal + METLIN database).

[0086] (2) Compound identification was based on multiple dimensions including RT (retention time), accurate mass, secondary fragmentation, and isotope distribution using the Human Metabolome Database (HMDB), Lipidmaps (v2.3), METLIN database, and LuMet-Animal3.0 local database.

[0087] (3) The extracted data were processed by missing value processing, 0 value replacement, score screening, data merging, etc. Missing value processing and 0 value replacement: Delete the ion peaks with missing values ​​(0 value) > 50% in the group, and replace the remaining 0 values ​​with half of the minimum ion intensity of all samples. Score screening: The qualitative compounds were screened according to the qualitative result score (Score) of the compound. The screening standard was 36 points (out of 80 points). The qualitative results below 36 points were considered inaccurate and deleted.

[0088] (4) Multivariate statistical analysis will first use unsupervised principal component analysis (PCA) to observe the overall distribution between samples and the stability of the entire analysis process, and then use supervised partial least squares analysis (PLS-DA) and orthogonal partial least squares analysis (OPLS-DA) to distinguish the overall differences in metabolic profiles between groups and find the differential metabolites between groups.

[0089] 2.6 Marker combination screening

[0090] (1) The markers of colorectal cancer group and healthy group were statistically analyzed for intergroup differences (Wilcoxon test), and species with P < 0.05 were used as potential biomarkers; through the Wilcoxon intergroup difference test, markers with P values ​​less than 0.05 after Bonferroni correction were selected, a total of 325 metabolites.

[0091] (2) The colorectal cancer samples (determined by colonoscopy and pathological examination) were marked as 1, and the healthy samples were marked as 0. The metabolite and microbial abundance matrices of different samples were modeled, analyzed, and predicted. The R software package AUCRF (random forest method) was used to determine the optimal combination of metabolites and microbial biomarkers. The results are shown in Figure 1 As shown, the diagnostic effect is best when the combination contains 52 markers.

[0092] (3) The samples were divided into a training set and a validation set (7:3). Based on the 52 markers screened by the AUC RF package, the pROC package was used to calculate the predictive performance of the markers in the validation set. The AUC values ​​of the remaining marker combinations were determined by the one-by-one elimination method.

[0093] Under the condition that the AUC value remains above 0.95, the markers that do not affect the AUC value are deleted, and finally four markers, Felineleukemia virus (feline leukemia virus), Sinomonas sp. JC656 (Sinomonas sp. JC656), TriethylPhospha te (triethyl phosphate), and Arachidonic Acid (arachidonic acid), are determined as the core markers, which can enable diagnosis.

[0094] (4) The above-mentioned markers were used to simultaneously construct a logistic regression model, and the random forest (RF) and logistic regression models (Glm) were compared. The random forest with better diagnostic performance was selected as the prediction model.

[0095] Through the above analysis, four metabolites of feline leukemia virus, Sinomonas sp. JC656, triethyl phosphate and arachidonic acid were selected as a marker combination for early diagnosis of colorectal cancer, which was used to distinguish between healthy and colorectal cancer groups.

[0096] Example 2: Establishment and verification of an early diagnosis or prediction model for colorectal cancer

[0097] 1. Model construction

[0098] The metabolites and microorganisms obtained from the final screening of Example 1 were selected. Feline leukemia virus, Chinese

[0099] Sinomonas sp. JC656, triethyl phosphate and arachidonic acid were used as a marker combination for early diagnosis of colorectal cancer to construct an early diagnosis or prediction model for colorectal cancer. The specific method is as follows:

[0100] (1) Extracting metabolites and microorganisms Feline leukemia virus, Sinomonas sp., triethyl phosphate, and arachidonic acid from blood samples to obtain the abundance values ​​of the above markers;

[0101] (2) The colorectal cancer samples (determined by colonoscopy and pathological examination) were marked as 1, and the healthy samples were marked as 0. The abundance values ​​of the biomarker combinations of different samples (n*n table) were input into the random forest to build a model and predict the diagnosis results.

[0102] (3) When the abundance matrix of the marker is used and the prediction value of the random forest algorithm is 1, the diagnosis result of the sample is judged to be positive, that is, colorectal cancer; when the prediction value is 0, the diagnosis result of the sample is judged to be negative, that is, healthy.

[0103] 2. Accuracy Verification

[0104] The above-mentioned predicted diagnosis results were compared with the actual diagnosis results (colonoscopy and pathological examination confirmed colorectal cancer) to determine the accuracy.

[0105] 3. Results Analysis

[0106] Four markers, Feline leukemia virus, Sinomonas sp. JC656, Triethyl Phosphate and Arachidonic Acid, were used to distinguish between healthy and colorectal cancer groups. The results are as follows Figure 2 As shown, the sensitivity was 88.9%, the specificity was 100%, and the accuracy was 95.5%.

[0107] The above results show that the biomarker combination described in the invention has high accuracy, specificity and sensitivity for diagnosing colorectal cancer.

[0108] Example 3: Early diagnosis or prediction system for colorectal cancer

[0109] This embodiment provides a system for early diagnosis or prediction of colorectal cancer, including:

[0110] (1) a pre-input module, at least for inputting the biomarker combination abundance data in the stool sample and transmitting it to the evaluation module;

[0111] (2) an evaluation module, at least for analyzing the abundance data of the biomarker combination in the stool sample according to the model described in Example 2 to obtain a result; wherein the abundance data of the biomarker combination can be collected through a pre-input module, or the abundance data of the biomarker combination can be obtained from other sources.

[0112] (3) A display module, at least used to display the evaluation result. When the prediction value is 0, the diagnosis result of colorectal cancer is negative; when the prediction value is 1, the diagnosis result of colorectal cancer is positive.

[0113] Embodiment 4: Computer equipment

[0114] The present embodiment provides an electronic device, which can be expressed in the form of a computing device (for example, a server device), including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the colorectal cancer early diagnosis or prediction model in Embodiment 2 of the present invention when executing the computer program.

[0115] When the processor executes the program, the system implements the following steps, including:

[0116] (1) Collecting and / or inputting biomarker combination abundance data for early diagnosis or prediction of colorectal cancer;

[0117] (2) Substituting the evaluation data into the model for calculation, and obtaining a conclusion on whether the sample is positive or negative for colorectal cancer;

[0118] (3) Output the judgment conclusion whether the sample is positive or negative for colorectal cancer.

[0119] It should be noted that although several units / modules or sub-units / modules of the electronic device are mentioned in the above detailed description, this division is merely exemplary and not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided into multiple units / modules to be embodied.

[0120] Embodiment 5: Computer readable storage medium

[0121] This embodiment provides a computer-readable storage medium on which the computer device is stored. When the program is executed by a processor, the steps of the method for evaluating the geography of the Great Zodiac in the first embodiment of the present invention are implemented.

[0122] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0123] In a possible implementation, the present invention can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the model for early diagnosis or prediction of colorectal cancer in Example 2 of the present invention.

[0124] The program code for executing the present invention may be written in any combination of one or more programming languages, and may be executed entirely on a user device, partially on a user device, as an independent software package, partially on a user device and partially on a remote device, or entirely on a remote device.

[0125] The above description is a detailed description of the preferred feasible embodiments of the present invention, but the embodiments are not intended to limit the scope of the patent application of the present invention. All equivalent changes or modified changes completed under the technical spirit suggested by the present invention should fall within the patent scope covered by the present invention.

Claims

1. A biomarker combination for early diagnosis or prediction of colorectal cancer, characterized in that: The biomarker combination includes one or more of Feline leukemia virus, Sinomonas sp. JC656, TriethylPhosphate, and Arachidonic Acid.

2. A reagent for detecting the biomarker combination as claimed in claim 1, characterized in that: The reagent is used to detect the expression level of the biomarker combination.

3. Use of the biomarker combination according to claim 1 or the reagent according to claim 2 in the preparation of a reagent, a kit or a chip for early diagnosis or prediction of colorectal cancer.

4. Use of the biomarker combination as described in claim 1 or the reagent as described in claim 2 in constructing a colorectal cancer patient screening model.

5. A kit for early diagnosis or prediction of colorectal cancer, characterized in that: The kit comprises reagents or instruments for detecting the biomarker combination according to claim 1.

6. A method for constructing an early diagnosis model for colorectal cancer, characterized in that: The construction method comprises: inputting the expression abundance data in the blood cfDNA and blood metabolite database into a random forest, performing machine learning using a generalized linear regression model, and constructing the prediction model; the sources of the expression abundance data in the blood cfDNA and blood metabolite database comprise plasma samples of colorectal cancer patients and plasma samples of normal people; the expression abundance data comprises the abundance data of the biomarker combination described in claim 1.

7. A model for early diagnosis of colorectal cancer, characterized in that: The model includes: (1) Sample processing: Extract the blood metabolite marker combination from the blood sample to be tested and perform LC-MS / MS full scan detection to obtain the abundance value of the marker; extract cfDNA from the plasma sample for metagenomic sequencing to obtain the microbiome abundance; (2) Model construction: The colorectal cancer samples confirmed by colonoscopy and pathological examination were marked as 1, and the samples of healthy people were marked as 0. The abundance values ​​of the biomarker combinations of different samples were input into the random forest to build a model and predict the diagnosis results; (3) Result determination: When the predicted value is 1, the diagnosis result of colorectal cancer is positive; when the predicted value is 0, the diagnosis result of colorectal cancer is negative.

8. A colorectal cancer early diagnosis model system, characterized in that: include: (1) a pre-input module, at least used to input the data to be diagnosed; (2) a diagnostic module, at least used to diagnose the data to be diagnosed, which is executed by the model for early diagnosis of colorectal cancer according to claim 7; (3) A display module, at least used to display the diagnosis result.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the system of claim 8 implements the following steps, including: (1) Collect and / or input colorectal cancer diagnosis data; (2) Substituting the diagnostic data into the model for calculation, and obtaining a conclusion on whether the patient has colorectal cancer; (3) Output the conclusion of whether the patient has colorectal cancer.

10. A computer-readable storage medium having stored thereon the computer device according to claim 9.

Images