Combined marker for detecting tumor radiotherapy sensitivity and application thereof

Abstract

The invention provides a combined marker for detecting tumor radiotherapy sensitivity and application thereof, and belongs to the technical field of tumor radiotherapy sensitivity diagnosis. The combined markers comprise a first combined marker and a second combined marker. The combined marker can be used for effectively distinguishing the tumor radiotherapy sensitivity of a cancer patient and can also be used for grading identification of the tumor radiotherapy sensitivity of the cancer patient, the accuracy of judging the radiotherapy sensitivity degree of the cancer patient is improved, and an important theoretical basis is provided for efficiently treating the cancer patient through radiotherapy.

Description

Technical Field

[0001] The present invention belongs to the technical field of tumor radiotherapy sensitivity diagnosis, and in particular relates to a combined marker for detecting tumor radiotherapy sensitivity and an application thereof. Background Art

[0002] Colorectal cancer accounts for approximately 10% of all cancers and cancer-related deaths diagnosed worldwide each year. It is the second most common cancer in women and the third most common cancer in men. Radiotherapy is an important part of the comprehensive treatment of colorectal cancer, and the preoperative chemoradiotherapy treatment strategy remains the standard treatment strategy for locally advanced rectal cancer (stage II and III). However, while radiotherapy brings therapeutic benefits to patients, it also increases its toxic side effects. The proportion of people who benefit is relatively small, and there are still many problems that need to be solved.

[0003] In the field of radiotherapy, clinical efficacy varies greatly between individuals. Even if the same regimen and dose of radiotherapy are received, about half of the tumors still do not achieve obvious regression, and even progress. The resistance of colorectal cancer (CRC) cells to radiotherapy largely leads to poor clinical prognosis of colorectal cancer patients. However, there are still obstacles in distinguishing colorectal cancer patients with high sensitivity (HS), moderate sensitivity (MS) and low sensitivity (LS) to radiotherapy. In the evaluation of radiotherapy response, tumor regression grading (TRG) is widely accepted. Previous studies have shown that AJCC-TRG grading accurately predicts the prognosis of rectal cancer after chemoradiotherapy. However, TRG grading fails to grade the sensitivity of radiation response of colorectal cancer patients before treatment. Therefore, it is of great significance to obtain a method for screening tumor radiotherapy sensitivity markers for the prediction of radiotherapy sensitivity. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a combined marker for detecting tumor radiotherapy sensitivity and its application. The present invention uses the combined marker to diagnose cancer patients' sensitivity to tumor radiotherapy and its degree with high accuracy.

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

[0006] The present invention provides a combined marker for detecting tumor radiotherapy sensitivity, wherein the combined marker is a first combined marker and a second combined marker;

[0007] The first combination marker is dihydrofolic acid, (R)-3-Hydroxy-tetradecanoic acid, 2-Methylglutaric acid and s__Roseburia_inulinivorans;

[0008] The second combination marker is Cholic acid, (R)-3-Hydroxy-tetradecanoic acid, Dihydrofolic acid, D-Glutamine, Deoxycholic acid, Thyroxine, Phenylglyoxylicacid, s__Roseburia_inulinivorans and s__Prevotella_nigrescens.

[0009] The present invention provides an application of a combined marker in the preparation of a product for diagnosing tumor radiotherapy sensitivity.

[0010] The present invention provides an application of a combined marker in the preparation of a product for diagnosing tumor radiotherapy sensitivity grading.

[0011] Preferably, the tumor radiotherapy sensitivity is graded into highly sensitive tumor radiotherapy and moderately sensitive tumor radiotherapy.

[0012] The present invention provides a system for predicting tumor radiotherapy sensitivity using the above-mentioned combined markers, comprising a nucleic acid sample separation unit for separating a fecal flora nucleic acid sample from a test sample;

[0013] A data processing unit, using the above-mentioned combined markers to establish a logistic regression model to obtain a critical value;

[0014] The result determination unit is used to compare the critical value obtained by the data processing unit with the set diagnosis value.

[0015] Preferably, the logistic regression model is a first logistic regression model and a second logistic regression model;

[0016] The first logistic regression model is a logistic regression model for constructing stool samples to be tested vs stool samples of colorectal cancer patients who are insensitive to radiotherapy using the first combination of markers. When the critical value is <0.95, it indicates that the tumor patient is insensitive to radiotherapy. When the critical value is ≥0.95, it indicates that the tumor patient is moderately sensitive to radiotherapy or highly sensitive to radiotherapy.

[0017] The second logistic regression model is a logistic regression model constructed by using the above-mentioned second combination marker to construct a critical value ≥0.95 for the stool sample to be tested vs the stool sample of the colorectal cancer patient with moderate sensitivity to radiotherapy. When the critical value is <0.52, it indicates that the tumor patient is moderately sensitive to radiotherapy, and when the critical value is ≥0.52, it indicates that the tumor patient is highly sensitive to radiotherapy.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] The present invention provides a combined marker for detecting tumor radiotherapy sensitivity and its application. The present invention can effectively distinguish the tumor radiotherapy sensitivity of cancer patients by using the combined marker, and can also be used for grading and identifying the tumor radiotherapy sensitivity of cancer patients, thereby improving the accuracy of judging the radiotherapy sensitivity of cancer patients and providing an important theoretical basis for efficiently using radiotherapy to treat cancer patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is the 16S rDNA region of bacteria;

[0021] Figure 2 The principle of PacBio sequencing;

[0022] Figure 3 It is the PacBio CCS model;

[0023] Figure 4 This is the full-length 16S experimental process;

[0024] Figure 5 To draw the ROC curve according to the method of fitting the logistic regression model, A is the ROC curve of the sample to be tested vs. the sample of the colorectal cancer patient who is not sensitive to radiotherapy; B is the ROC curve of the sample to be tested with P ≥ 0.95 vs. the sample of the colorectal cancer patient who is moderately sensitive to radiotherapy;

[0025] Figure 6 To draw a ROC curve based on the combination of markers involved in the logistic regression model equation, A is a ROC curve of the markers involved in the ROC curve of the sample to be tested vs the sample of the colorectal cancer patient who is not sensitive to radiotherapy using the combination of markers in Example 3; B is a ROC curve of the markers involved in the ROC curve of the sample to be tested with P≥0.95 vs the sample of the colorectal cancer patient who is moderately sensitive to radiotherapy using the combination of markers in Example 3. DETAILED DESCRIPTION

[0026] The present invention provides a combined marker for detecting tumor radiotherapy sensitivity, wherein the combined marker is a first combined marker and a second combined marker;

[0027] The first combination marker is dihydrofolic acid, (R)-3-Hydroxy-tetradecanoic acid, 2-Methylglutaric acid and s__Roseburia_inulinivorans;

[0028] The second combination marker is Cholic acid, (R)-3-Hydroxy-tetradecanoic acid, Dihydrofolic acid, D-Glutamine, Deoxycholic acid, Thyroxine, Phenylglyoxylicacid, s__Roseburia_inulinivorans and s__Prevotella_nigrescens.

[0029] In the present invention, the first combination of markers can be used to effectively diagnose whether a patient is sensitive to tumor radiotherapy. Further, after the first combination of markers is used to effectively diagnose patients who are sensitive to tumor radiotherapy, in order to determine the degree of sensitivity of the above-mentioned patients to tumor radiotherapy, the present invention further uses the second combination of markers to construct a logistic regression model of stool samples to be tested with a critical value ≥ 0.95 vs stool samples of colorectal cancer patients with moderate sensitivity to radiotherapy, effectively diagnosing whether the patient is highly sensitive to tumor radiotherapy or moderately sensitive to tumor radiotherapy.

[0030] The present invention provides an application of a combined marker in the preparation of a product for diagnosing tumor radiotherapy sensitivity.

[0031] The present invention provides an application of a combined marker in the preparation of a product for diagnosing tumor radiotherapy sensitivity grading.

[0032] In the present invention, the tumor radiotherapy sensitivity is graded into high sensitivity to tumor radiotherapy and moderate sensitivity to tumor radiotherapy. The present invention divides tumor radiotherapy sensitivity into high sensitivity to tumor radiotherapy, moderate sensitivity to tumor radiotherapy and low sensitivity to tumor radiotherapy (also known as insensitive to tumor radiotherapy) according to the NAR (Neoadjuvant Rectal) score. When the NAR score is less than 8, it means that the patient is highly sensitive to tumor radiotherapy, when the NAR score is 8-16, it means that the patient is moderately sensitive to tumor radiotherapy, and when the NAR score is greater than 16, it means that the patient is insensitive to tumor radiotherapy.

[0033] The present invention provides a system for predicting tumor radiotherapy sensitivity using the above-mentioned combined markers, comprising a nucleic acid sample separation unit for separating a fecal flora nucleic acid sample from a test sample;

[0034] A data processing unit, using the above-mentioned combined markers to establish a logistic regression model to obtain a critical value;

[0035] The result determination unit is used to compare the critical value obtained by the data processing unit with the set diagnosis value.

[0036] In the present invention, the critical value is also called the optimal threshold for predicting P. When judging whether the stool sample to be tested is sensitive to tumor radiotherapy, the diagnostic value is 0.95, and when judging that the stool sample to be tested is sensitive to tumor radiotherapy, and further judging its sensitivity to tumor radiotherapy, the diagnostic value is 0.52.

[0037] In the present invention, the logistic regression model is a first logistic regression model and a second logistic regression model; the first logistic regression model is a logistic regression model of stool samples to be tested vs stool samples of colorectal cancer patients who are insensitive to radiotherapy constructed using the first combination marker mentioned above, when the critical value is <0.95, it indicates that the tumor patient is insensitive to radiotherapy, and when the critical value is ≥0.95, it indicates that the tumor patient is moderately sensitive to radiotherapy or highly sensitive to radiotherapy;

[0038] The second logistic regression model is a logistic regression model constructed by using the above-mentioned second combination marker to construct a critical value ≥0.95 for the stool sample to be tested vs the stool sample of the colorectal cancer patient with moderate sensitivity to radiotherapy. When the critical value is <0.52, it indicates that the tumor patient is moderately sensitive to radiotherapy, and when the critical value is ≥0.52, it indicates that the tumor patient is highly sensitive to radiotherapy.

[0039] The present invention utilizes the above-mentioned combined markers to effectively diagnose whether a tumor patient is sensitive to tumor radiotherapy and the degree of sensitivity to tumor radiotherapy, and the diagnosis is highly accurate.

[0040] In the present invention, unless otherwise specified, all raw material components are commercially available products well known to those skilled in the art.

[0041] The technical solutions provided by the present invention are described in detail below in conjunction with the embodiments, but they should not be construed as limiting the protection scope of the present invention.

[0042] Example 1

[0043] A method for screening and detecting metabolic markers of tumor radiotherapy sensitivity, comprising the following steps:

[0044] (1) Metabolomics sample collection plan

[0045] The blood was collected in a centrifuge tube and allowed to stand at 37°C (or room temperature) for 1 hour for coagulation and stratification. Then, the blood was centrifuged at 3000 rpm for 10 minutes at room temperature, and the supernatant was transferred to a clean centrifuge tube. The blood was centrifuged at 12000 rpm for 10 minutes at 4°C, and the supernatant was dispensed into 1.5 mL centrifuge tubes to obtain serum, which was stored at -80°C.

[0046] (2) Metabolomics serum sample pre-treatment plan

[0047] The serum sample of step (1) was transferred to an EP tube, and an extract (Baiqu Biotechnology, Shanghai) (containing an isotope-labeled internal standard mixture) was added, vortexed for 30 seconds, ultrasonicated in an ice-water bath for 10 minutes, and allowed to stand at -40°C for 1 hour; the sample was centrifuged at 4°C for 15 minutes to obtain the supernatant, which was placed in a sample injection bottle for detection;

[0048] (3) Metabolomics serum sample experimental detection plan

[0049] Ultra-high performance liquid chromatography can efficiently separate various mixtures and perform qualitative analysis on each component in the sample by measuring peak area or peak height.

[0050] The supernatant obtained in step (2) was chromatographed using a Vanquish (Thermo Fisher Scientific) ultra-high performance liquid chromatograph with a Waters ACQUITY UPLCHSS T3 (2.1 mm × 100 mm, 1.8 μm) liquid chromatography column. Phase A of the liquid chromatography was an aqueous phase containing 5 mmol / L ammonium acetate and 5 mmol / L acetic acid, and phase B was acetonitrile. Sample plate temperature: 4°C, injection volume: 2 μL. The Orbitrap Exploris 120 mass spectrometer was capable of primary and secondary mass spectrometry data acquisition under the control of the control software (Xcalibur, version: 4.4, Thermo).

[0051] The raw data were converted into mzXML format using ProteoWizard software, and then processed using the R package (with XCMS as the kernel) for peak identification, peak extraction, peak alignment, and integration. They were then matched with the BiotreeDB (V2.1) mass spectrometry database for substance annotation, and the cutoff value for algorithm scoring was set to 0.3.

[0052] In practical applications, ① multiple cohorts of serum samples were obtained and classified into 6 categories, including prHS (pre-radiotherapy high sensitivity), prMS (pre-radiotherapy medium sensitivity), prLS (pre-radiotherapy low sensitivity), poHS (post-radiotherapy high sensitivity), poMS (post-radiotherapy medium sensitivity), and poLS (post-radiotherapy low sensitivity). Specifically, prHS stands for: sensitive data before radiotherapy (pre-radiotherapy highly sensitive metabolite subset), prMS stands for: moderately sensitive data before radiotherapy (pre-radiotherapy moderately sensitive metabolite subset), prLS stands for: insensitive data before radiotherapy (pre-radiotherapy insensitive metabolite subset), poHS stands for: sensitive data after radiotherapy (post-radiotherapy highly sensitive metabolite subset), poMS stands for: moderately sensitive data after radiotherapy (post-radiotherapy moderately sensitive metabolite subset), and poLS stands for: insensitive data after radiotherapy (post-radiotherapy insensitive metabolite subset).

[0053] ② Combined analysis of positive and negative ion modes.

[0054] The data of 185 serum samples were divided into positive ion mode and negative ion mode, including 19 samples in poHS group, 5 samples in poLS group, 21 samples in poMS group, 56 samples in prHS group, 30 samples in prLS group, and 54 samples in prMS group. Among them, there were 958 substances in positive ion mode and 475 substances in negative ion mode.

[0055] The serum sample data results include MS2 name (the substance name obtained by secondary mass spectrometry qualitative matching analysis) and MS2 score (the score of secondary mass spectrometry qualitative matching, with a value of [0, 1]). The data are merged in positive and negative ion modes. For substances with duplicate names, i.e., repeated MS2 names, substances with high MS2 score secondary mass spectrometry qualitative matching scores are retained, and those with low MS2 scores are deleted, and the data in the sample is retained; when the scores of substances with duplicate names are the same, substances with high peak areas in QC (quality control) samples, i.e., substances with high QC sample data, are retained, and those with low peak areas are deleted. MS2 name and MS2 score are retained, and only the data in the sample is retained. The positive and negative ions are merged to obtain the merged data set, i.e., the initial sample set. There are 1328 substances in total after the positive and negative ions are merged.

[0056] Step S1: normalizing the initial sample set to obtain a processed sample set, specifically including:

[0057] A. Filter the deviation value of the initial sample set to obtain a first filtered sample set:

[0058] The deviation values ​​were filtered based on the relative standard deviation, and the relative standard deviation (rsd) was calculated for the QC group, i.e., the standard deviation divided by the corresponding mean value multiplied by 100%. The detection in the QC group with rsd greater than 0.3 was considered unstable, and all detection data of the metabolite were deleted to obtain the first filtered sample set.

[0059] B. Filter missing values ​​on the first filtered sample set to obtain a second filtered sample set:

[0060] Missing value filtering: only retain a single group of null values, i.e., peak area data with null values ​​of no more than 50% in each group of prHS, prMS, prLS, poHS, poMS, and poLS or null values ​​of no more than 50% in all groups, to obtain the second filtered sample set;

[0061] C. Fill missing values ​​in the sample set after the second filtering to obtain a filled sample set:

[0062] Performing minimum half filling on the missing values ​​in the second filtered sample set to obtain a filled sample set;

[0063] D. Perform normalization processing on the padded sample set to obtain a processed sample set:

[0064] The total ion current of each sample in the filled sample set was used for normalization to obtain the processed sample set. After normalization, there were 1192 substances in total.

[0065] Then, the processed sample set is used to train the initial model, specifically including:

[0066] Step S2: using principal component analysis and orthogonal partial least squares method to process the processed sample set to obtain an initial differential metabolite sample set;

[0067] Through the processed sample set, principal component analysis was performed for three different comparison strategies: prHS vs. prMS, prHS vs. prLS, and prMS vs. prLS. The differences between groups and the distribution of samples were reflected by the scatter plot of the principal component scores. The closer the sample points were, the more similar the content and type of metabolites in the samples were. Conversely, the farther the samples were, the greater the difference in their overall metabolic levels. The results were further analyzed by the statistical method of orthogonal partial least squares discriminant analysis. Through orthogonal partial least squares discriminant analysis, orthogonal variables that were not related to the categorical variables in the metabolites can be filtered out, and non-orthogonal variables and orthogonal variables can be analyzed separately, so as to obtain more reliable information on the correlation between the inter-group differences of metabolites and the experimental groups. The difference multiples between the two groups were compared by calculating the Fold Change value. A difference multiple between the two groups greater than 1 represents upregulation, and less than 1 represents downregulation. After the above analysis, the results of the statistical analysis of unit variables and multivariate variables were combined to screen out the differential metabolites, and the initial differential metabolite sample set was obtained.

[0068] Step S3: using a rank sum test to preliminarily screen the initial differential metabolite sample set to obtain a second differential metabolite sample set.

[0069] First, establish the null hypothesis: the two groups of samples come from the same population. Mix the two groups of data and rank them uniformly from small to large. When encountering the same data, take the average rank and calculate the statistic. Set the significance level a=0.05, and obtain the probability of the current statistic value under the condition that the null hypothesis is true. If p-value<0.05, reject the null hypothesis, indicating that there is a significant difference between the two samples.

[0070] Specifically, the rank sum test was performed for the four comparisons of prHS and prMS, prHS and prLS, prMS and prLS, and prHS+prMS and prLS. If p-value < 0.05, it means that there is a significant difference between the two samples. After screening, 47, 34, 40, and 36 differential metabolites were obtained between prHS and prMS, prHS and prLS, prMS and prLS, and prHS+prMS and prLS, respectively, as shown in Table 1.

[0071] Table 1 The second differential metabolite sample set after screening between prHS and prMS, prHS and prLS, prMS and prLS, and prHS+prMS and prLS

[0072]

[0073]

[0074] Step S4: Based on the logistic regression model and P-value established between the differential samples in the second differential metabolite sample set and the significant clinical indicators, the differential substances after eliminating the interference of clinical factors are obtained, that is, the third differential metabolite sample set is obtained.

[0075] The initial clinical indicators were screened using multiple comparisons and PerMANOVA multivariate analysis of variance, and significant clinical indicators were obtained, including:

[0076] a. Obtaining initial clinical indicators:

[0077] A total of 185 clinical samples were collected, and the clinical indicator information was extracted, including treatment response group, age, gender, BMI, smoking, drinking, T, N, GTV, distance from the anus in cm (MR / CT), CRM (MR), EMVI (MR), NLR, concurrent chemotherapy regimen, hormone use, and antibiotic use, a total of 16 items.

[0078] b. Screening the initial clinical indicators using the multiple comparison method to determine the second clinical indicator:

[0079] The multiple comparison method was used to retain the samples with p-value < 0.05 in the statistical test results of the above 16 clinical indicators. The p-value was obtained by performing multiple comparisons on the clinical indicators among multiple experimental groups, firstly testing their homogeneity by Bartlett Test (corresponding to the R language program function bartlett.test) (taking p-value < 0.05 as a non-null hypothesis), and then calculating their p-values ​​by the oneway.test function of the R language program package. If p < 0.05, it means that the clinical indicator is significant.

[0080] c. Use the PerMANOVA multivariate analysis of variance method to screen the second clinical indicator and determine the significant clinical indicator.

[0081] The permaonva analysis method was used to further screen out clinical indicator information that had a significant relationship with the metabolic profile expression of human serum samples (i.e., the second clinical indicator);

[0082] d. Using the logistic regression model, the significant clinical index information obtained in step c is used as the response variable to perform a final correction on the second differential metabolite sample set obtained in step S3, and the differential substances whose p-value after regression is still less than 0.05 are screened out to obtain the differential substances (third differential metabolite sample set) after eliminating the interference of clinical factors, specifically including:

[0083] The clinically significant index information obtained in step c is used as the response variable, and the second differential metabolite sample set obtained in step S3 is used as the independent variable. A logistic regression model is established to obtain the p-value. If the p-value of the differentially expressed substance is less than 0.05, it means that these substances are not affected by the clinical indicators, that is, they are differential substances after eliminating the interference of clinical factors.

[0084] In the prHS+prMS and prLS groups, the clinical indicator smoking was found to be significant using the permaonva analysis method, and the 36 differentially expressed substances screened were respectively used to establish logistic regression with the clinical indicator smoking. Among these 36 differentially expressed substances, 18 had p values ​​less than 0.05, namely ACar(11:1), H-PHE-PRO-OH, Thyroxine, ACar(13:0), ACar(9:0), ACar(13:1), SM(d14:0 / 18:2), 2-Methylglutaric acid, N1-Methyl-2-pyridone-5-carboxamide, N1-Methyl-4-pyridone-3-carboxamide, Dethiobiotin, 3-Hydroxyisovaleric acid, [7]-Paradol, I-Urobilin, (-)-Catechin 3-O-gallate, Isopentenyl acetate, Desglucocoroloside, and Spirotaccagenin.

[0085] All clinical indicators obtained by permaonva analysis method for prHS and prMS, prHS and prLS, and prMS and prLS groups were not significant. The 47, 34, and 40 differentially expressed substances in the initial screening did not need to establish logistic regression with all clinical indicators respectively. Therefore, the 47, 34, and 40 differentially expressed substances in the second differential metabolite sample set were the differential substances after excluding the interference of clinical factors.

[0086] e. For the above four comparison strategies of prHS and prMS, prHS and prLS, prMS and prLS, and prHS+prMS and prLS, the differential substances after eliminating the interference of clinical factors were merged to form the third differential metabolite sample set, and three different machine learning algorithms, random forest method, support vector machine method, and LASSO, were used for further screening and analysis. Based on the key candidate differential substance combinations finally screened out by these three different algorithms, joint ROC analysis was carried out, and finally the final differential substance combination panel was determined according to the obtained AUC value, specifically:

[0087] First, the third differential metabolite sample set was divided into a training set and a test set (the ratio of training set to test set was 8:2), and different algorithms were used to screen differential substances, as follows:

[0088] Random forest method: Use the R package randomForest to fit the random forest model to the training set, and finally determine the treeNum in the fitting model through parameter tuning. Fit the training set with the optimal parameters, and test the test set to obtain the confusion matrix of the prediction results. Further, use the rfcv function in the R package randomForest to perform 10-fold cross validation to obtain the relationship between variable selection and the average error rate of model classification, and obtain the optimal differential material based on MeanDecreaseAccuracy. The larger the value, the greater the importance of the variable.

[0089] Taking prHS and prMS as an example, there are 110 samples. According to the 47 differentially expressed substances obtained above, the training set (88) and the test set (22) are distinguished according to 8:2. The treeNum in the fitting model is finally determined to be 500 through parameter tuning. The rfcv function performs 10-fold cross validation to obtain the lowest average error rate of 34.1% when the number of variables is 47, which are Ovalicin, 3-Hydroxydecanoic acid, Glutamylthreonine, Diethylfumarate, Tetracosahexaenoic acid, 3-Hydroxymeth ylglutaricacid, Estrone, SM(d14:0 / 28:2), (3beta,5alpha,6beta,9alpha,22E,24R)-23-Methylergosta-7,22-diene-3,5,6,9-tetrol, (10E,12Z)-9-HODE, FAHFA(18:0 / 22:3), Pisatoside, 2-Isopropyl-3,5-dimethoxy-6-methylpyrazine, PC(16:3 / 24:4), Hydroxyisocaproi cacid, Apo-12'-violaxanthal, Fexofenadine, Gossypetin8-glucuronide3-sulfate, Cholicacid, 4-Acetylbutyrate, 3-(2-Furanylmeth ylene)pyrrolidine, Lauroyldiethanolamide, Cuscohygrine, PC(14:0e / 18:0), PE(22:6(4Z,7Z,10Z,13Z,16Z,19Z) / 16:0), Riboflavin, (- )-(E)-1-(4-Hydroxyphenyl)-7-phenyl-6-hepten-3-ol, (R)-3-Hydroxy-tetradecanoicacid, Vulgarole, Isoliquiritigenin, Pyroglut amicacid, PC(22:2(13Z,16Z) / 16:1(9Z)), Serylphenylalanine, ProstaglandinF1a, PC(16:1e / 20:4), Cholesterol, Mizolastine, (4E)-1,7-bis(4-hydroxyphenyl)hept-4-en-3-one, 4-Pentenoicacid, Nortriptyline, 8-IsoprostaglandinE1, Pipernonaline, 4-Deoxyerythronicacid, PiperoleinB, Triethyl citrate, 15-Keto-prostaglandin E2, PC(16:0 / 20:3). ,

[0090] Taking prHS and prLS as an example, there are 86 samples. According to the 34 differentially expressed substances obtained above, the training set (70) and the test set (17) are distinguished according to 8:2. The treeNum in the fitting model is finally determined to be 500 through parameter tuning. The rfcv function performs 10-fold cross validation to obtain the lowest average error rate of 22.1% when the number of variables is 34, which are ACar (11:1), D-Glutamine, Thyroxine, 1-Methyl-L-Histidine, Uridine, ACar (13:0), PC (14:0e / 18:0), ACar (9:0), ACar (13:1), SM (d14:0 / 18:2), Deoxycholic acid, Aspartame, Tetracosahexaenoic acid acid, PC(22:6(4Z,7Z,10Z,13Z,16Z,19Z) / 20:3(5Z,8Z,11Z)), ACar(9:1), 2-Methylglutaric acid, PC(22:2(13Z,16Z) / 16:1(9Z)), Butylparaben, Riboflavin, Uracil, PC(22:6(4Z,7Z,10Z,13Z,16Z,19Z) / 18:3(6Z,9Z,12Z)), ACar(15:0), 3-Hy droxyisovalerylcarnitine, Nicotinamide, Dehydrophytosphingosine, Ricinoleicacid, S-(2-carboxypropyl)-Cysteamine, 3-Hydroxyisovaleric acid, beta-Citraurinepoxide, Selenomethionine, (-)-Catechin3-O-gallate, Butyl levulinate, Desglucocoroloside, Spirotaccagenin.

[0091] Taking prMS and prLS as an example, there are 84 samples. According to the 40 differentially expressed substances obtained above, the training set (67) and the test set (17) are distinguished according to 8:2. The treeNum in the fitting model is finally determined to be 500 through parameter tuning. The rfcv function performs 10-fold cross validation to obtain the lowest average error rate of 29.9% when the number of variables is 40, which are Isopentenyl acetate, 2-Methylglutaric acid, Nicotinamide, ACar (9:0), Vulgarole, Spirotaccagenin, Dhelwangin, Phenylglyoxylic acid, 7-Ketocholesterol, (3beta,5alpha,6beta,9alpha,22E,24R)-23-Methylergosta-7,22-diene-3,5,6,9-tetrol, Carbendazim, Oleamide, H-PHE-PRO-OH, I-Urobilin, MethylIndole-3-acetate, 15-Keto-prostaglandin E2, Desglucocoroloside, PC(9:0 / 26:1), Aspartame, ACar(11:1), 3-Methyl-2-oxopentanoate, 1-Hydroxy-6-methoxypyrene, Dihydrofolic acid, ACar(13:0), LysoPC(P-16:0), Dethiobiotin, beta-Citraurin epoxide, FAHFA(18:3 / 22:3), SM(d14:0 / 22:2), N-Acetylglutamic acid, Xanthurenic acid, [7]-Paradol, 8,15-DiHETE, 2-Isopropyl-3,5-dimethoxy-6-methylpyrazine, 7-Methylguanosine, N1-Methy l-2-pyridone-5-carboxamide, ACar(13:1), PC(16:0e / 22:6), N1-Methyl-4-pyridone-3-carboxamide, L-Urobilin.

[0092] Taking prHS+prMS and prLS as examples, there are 140 samples. According to the 18 differentially expressed substances obtained above, the training set (112) and the test set (28) are distinguished according to 8:2. Through parameter tuning, the treeNum in the fitting model is finally determined to be 500. The rfcv function performs 10-fold cross validation to obtain the lowest average error rate of 17.9% when the number of variables is 18, namely ACar (11:1), H-PHE-PRO-OH, Thyroxine, ACar (13:0), ACar (9:0), ACar (13:1), SM (d14:0 / 18:2), 2-Methylglutaric acid, N1-Methyl-2-pyridone-5-carboxamide, N1-Methyl-4-pyridone-3-carboxamide, Dethiobiotin, 3-Hydroxyisovaleric acid, [7]-Paradol, I-Urobilin, (-)-Catechin 3-O-gallate, Isopentenyl acetate, Desglucocoroloside, Spirotaccagenin.

[0093] Support vector machine (SVM): Support vector machine is a two-class classification model. Its basic model is a linear classifier that finds the separating hyperplane with the maximum margin in the feature space. SVM is an algorithm for the post-selection of a sequence based on the maximum margin principle. In the first iteration, the SVM model is optimized for all feature sets in the data set. Then the score of each feature is calculated and sorted in descending order. The feature set with the smallest score is recorded, and then the feature with the smallest score is deleted. It is iterated again until only one feature is left. The training set is trained using the svm function of the R language to obtain the relationship between variable selection and the average error rate of model classification, and the feature set with the lowest error rate is selected.

[0094] Taking prHS and prMS as an example, there were 110 samples. According to the 47 differentially expressed substances obtained above, the training set (88) and the test set (22) were distinguished according to 8:2. The training set was trained using the svm function of R language, and the lowest error rate was 35.2% when the variables were Diethyl fumarate, Pisatoside, Pyroglutamic acid, SM (d14:0 / 28:2), 4-Pentenoicacid, Pipernonaline, 4-Deoxyerythronic acid, FAHFA (18:0 / 22:3), PC (16:1e / 20:4), 4-Acetylbutyrate, Fexofenadine, PC (16:0 / 20:3), 3-(2-Furanylmethylene)pyrrolidine, PC (16:3 / 24:4), Cholesterol, Mizolastine, Glutamylthreonine, and Apo-12'-violaxanthal.

[0095] Taking prHS and prLS as an example, there were 86 samples in total. According to the 34 differentially expressed substances obtained above, the training set (70) and the test set (17) were distinguished according to 8:2. The training set was trained using the svm function of R language, and the lowest error rate was 20.6% when the variables were Spirotaccagenin, Selenomethionine, Uracil, Uridine, Desglucocoroloside, (-)-Catechin 3-O-gallate, PC (22:6 (4Z, 7Z, 10Z, 13Z, 16Z, 19Z) / 18:3 (6Z, 9Z, 12Z)), Butylparaben, beta-Citraurin epoxide, 3-Hydroxyisovalerylcarnitine, 1-Methyl-L-Histidine, Thyroxine, 2-Methylglutaric acid, Nicotinamide, ACar (9:1), and ACar (13:0).

[0096] Taking prMS and prLS as an example, there are 84 samples. According to the 40 differentially expressed substances obtained above, the training set (67) and the test set (17) were distinguished according to the 8:2 ratio. The training set was trained using the R language svm function, and the variables Spirotaccagenin, Desglucocoroloside, Dihydrofolic acid, Dhelwangin, Isopentenylacetate, 1-Hydroxy-6-methoxypyrene, I-Urobilin, Carbendazim, [7]-Paradol, and beta-Citraurin were obtained. epoxide, 2-Isopropyl-3,5-dimethoxy-6-methylpyrazine, Dethiobiotin, N1-Methyl-4-pyridone-3-carboxamide, (3beta,5alpha,6beta,9alpha,22E,24R)-23-Methylergosta-7,22-diene-3,5,6,9-tetrol, N-Acetylglutamic acid, Nicotinamide, 15-Keto-prostaglandin E2, 7-Methylguanosine, 7-Ketocholesterol, Xanthurenicacid, Vulgarole, N1-Methyl-2-pyridone-5-carboxamide, 3-Methyl-2-oxopentanoate, LysoPC(P-16:0), 2-Methylglutaric acid, L-Urobilin, Oleamide, Aspartame, Phenylglyoxylic The lowest error rate was 16.4% when using acid, MethylIndole-3-acetate, PC(9:0 / 26:1), PC(16:0e / 22:6), ACar(13:1), ACar(9:0), 8,15-DiHETE, ACar(13:0), FAHFA(18:3 / 22:3), SM(d14:0 / 22:2), H-PHE-PRO-OH, and ACar(11:1).

[0097] Taking prHS+prMS and prLS as examples, there were 140 samples. According to the 18 differentially expressed substances obtained above, the training set (112) and the test set (28) were distinguished according to 8:2. The training set was trained using the svm function of R language, and the lowest error rate was 16.1% when the variables were Isopentenyl acetate, (-)-Catechin 3-O-gallate, 2-Methylglutaric acid, Desglucocoroloside, ACar (13:1), H-PHE-PRO-OH, Thyroxine, [7]-Paradol, and N1-Methyl-2-pyridone-5-carboxamide.

[0098] LASSO: LASSO imposes regression penalties on all variables, making the coefficients of relatively unimportant variables become 0, thus being excluded from modeling, and then selecting independent variables that have a greater impact on the dependent variable and calculating the corresponding regression coefficients, and finally obtaining a prediction model. The model is trained on the training set using the R language glmnet. The lambda value with the smallest mean square error is selected through the mean square error under different parameters lambda. A vertical line is drawn through the position of the corresponding penalty value of the change trajectory diagram of each independent variable coefficient. Each curve represents a variable, and the variable that intersects with the penalty value is the variable finally included in the model. The ordinate corresponding to the variable is the regression coefficient of the variable, and the regression coefficient is also the contribution of the variable.

[0099] Taking prHS and prMS as an example, there are 110 samples. According to the 47 differentially expressed substances obtained above, the training set (88) and the test set (22) were distinguished according to 8:2. The training set was trained using the R language glmnet, and 28 differential metabolites were obtained: (R)-3-Hydroxy-tetradecanoic acid, Lauroyl diethanolamide, FAHFA (18:0 / 22:3), Hydroxyisocaproic acid, PC(16:3 / 24:4), PC(14:0e / 18:0), Cholicacid, 3-(2-Furanylmethylene)pyrrolidine, SM(d14:0 / 28:2), Serylphenylalanine , PC(16:0 / 20:3), Tetracosahexaenoicacid, PC(22:2(13Z,16Z) / 16:1(9Z)), Vulgarole, Riboflavin, 4-Acetylbutyrate, Gossypetin 8-glucuronide 3-sulfate, Estrone, 3-Hydroxymethylglutaric acid, (4E)-1,7-bis(4-hydroxyphenyl)hept-4-en-3-one, Pisatoside, (3beta,5alpha,6beta,9alpha,22E,24R)-23-Methylergosta-7,22-diene-3,5,6,9-tetrol, Diethyl fumarate, 2-Isopropyl-3,5-dimethoxy-6-methylpyrazine, PE (22: 6 (4Z, 7Z, 10Z, 13Z, 16Z, 19Z) / 16: 0), Ovalicin, Glutamylthreonine, Isoliquiritigenin, the specific regression equation is: Y = 0.176630112-0.481373480 × (R) -3-Hydroxy-tetradecanoic acid + 0.189970030 × Lauroyldiethanolamide + 0.294612153 × FAHFA (18: 0 / 22: 3) - 0.114201224 × Hydroxyisocaproic acid-0.329202341×PC(16:3 / 24:4)+0.063337880×PC(14:0e / 18:0)-0.202945506×Cholic acid+0.402645810×3-(2-Furanylmethylene)pyrrolidine+0.068413260×SM(d14:0 / 28:2)-0.453305087×Serylphenylalanine+0.245675693×PC(16:0 / 20:3)-0.547027447×Tetracosahexaenoicacid+0.020912311×PC(22:2(13Z,16Z) / 16:1(9Z))+0.220321552×Vulgarole+0.049755862×Riboflavin+0.325673143×4-Acetylbutyrate+0.112288678×Gossypetin8-glucuronide 3-sulfate-0.014118167×Estrone-0.621134893×3-Hydroxymethylglutaric acid+0.158254558×(4E)-1,7-bis(4-hydroxyphenyl)hept-4-en-3-one+0.237417433×Pisatoside-0.052592560×(3beta,5alpha,6beta,9alpha,22E,24R)-23-Methylergosta-7,22-diene-3,5,6,9-tetrol+0.265102112×Diethyl.

[0100] fumarate-0.303018854×2-Isopropyl-3,5-dimethoxy-6-methylpyrazine+0.131173867×PE(22:6(4Z,7Z,10Z,13Z,16Z,19Z) / 16:0)+0.196723739×Ovalicin-0.581576171×Glutamylthreonine-0.371153252×Isoliquiritigenin。

[0101] Taking prHS and prLS as an example, there are 86 samples. According to the 34 differentially expressed substances obtained above, the training set (70) and the test set (17) were distinguished according to 8:2. The training set was trained by R language glmnet, and 21 differential metabolites were obtained: Thyroxine, Uridine, ACar (13:0), PC (14:0e / 18:0), SM (d14:0 / 18:2), Deoxycholicacid, Aspartame, Tetracosahexaenoic acid, PC (22:6 (4Z, 7Z, 10Z, 13Z, 16Z, 19Z) / 20:3 (5Z, 8Z, 11Z)), ACar (9:1), 2-Methylglutaricacid, Butylparaben, Riboflavin, Nicotinamide, Dehydrophytosphingosine, Ricinoleic acid acid, beta-Citraurinepoxide, (-)-Catechin3-O-gallate, Butyllevulinate, Desglucocoroloside, Spirotaccagenin, the specific regression equation is: Y = 0.945592741 + 0.425529358 × Thyroxine + 0.434334101 × Uridine + 0.1416733 × ACar (13: 0) + 0.122174358 × PC (14: 0e / 18: 0) + 0.023755087 × SM (d14: 0 / 18: 2) - 0.433318642 × Deoxycholic acid + 0.139953171 × Aspartame - 0.042631334 × Tetracosahexaenoic acid acid+0.01596018×PC(22:6(4Z,7Z,10Z,13Z,16Z,19Z) / 20:3(5Z,8Z,11Z))+0.040489289×ACar(9:1)-0.472540206×2-Methylglutaric

[0102] acid-0.331792329×Butylparaben+0.019591005×Riboflavin+0.014836255×Nicotinamide-0.002452764×Dehydrophytosphingosine-0.475405577×Ricinoleicacid-0.025951223×beta-Citraurinepoxide+0.489435332×(-)-Catechin 3-O-gallate-0.058483768×Butyllevulinate+0.003892541×Desglucocoroloside+0.056270747×Spirotaccagenin。

[0103] Taking prMS and prLS as an example, there are 84 samples. According to the 40 differentially expressed substances obtained above, the training set (67) and the test set (17) are distinguished according to 8:2. The training set is trained by using R language glmnet, and the 24 differential metabolites are obtained: The specific regression equation is: Y = 2.99887369337435 + 1.10976468382751 × H-PHE-PRO- OH+0.920014445806788×ACar(13:0)+0.459808787978451×8,15-DiHETE+1.1340621781099 1×ACar(9:0)+1.17299463874812×PC(16:0e / 22:6)-0.959473151493842×Phenylglyoxylic acid+2.92897693190931×Oleamide-1.83958237987782×L-Urobilin-3.65659024396626×2-Methylglutaricacid+0.047319849 5987288×N1-Methyl-2-pyridone-5-carboxamide-1.2340219781403×Vulgarole+0.704837527183466×15-Keto-prostaglandin E2+0.894224083216816×Nicotinamide-0.354283446737299×(3beta,5alpha,6beta,9alpha,22E,24R)-23-Methylergosta-7,22-diene- 3,5,6,9-tetrol+2.69118679210089×Dethiobiotin-0.300262274515251×2-Isopropyl-3,5-dimethoxy-6-methylpyrazine-0.81664507 6823208×beta-Citraurinepoxide-0.84713756840258×I-Urobilin+1.62670013495577×1-Hydroxy-6-methoxypyrene+1.5842157663272 5×Isopentenylacetate-2.5391338060763×Dhelwangin+1.36756431423767×Dihydrofolicacid+0.055139707565556×Spirotaccagenin.

[0104] Taking prHS+prMS and prLS as examples, there are 140 samples. According to the 18 differentially expressed substances obtained above, the training set (112) and the test set (28) were distinguished according to 8:2. The training set was trained using the R language glmnet model, and 15 differential metabolites were obtained: ACar (11:1), H-PHE-PRO-OH, Thyroxine, ACar (13:0), ACar (13:1), 2-Methylglutaricacid, N1-Methyl-2-pyridone-5-carboxamide, Dethiobiotin, 3-Hydroxyisovalericacid, [7]-Par adol, I-Urobilin, (-)-Catechin3-O-gallate, Isopentenylacetate, Desglucocoroloside, Spirotaccagenin, the specific regression equation is: Y = 2.408973956 + 0.059529776 × ACar (11: 1) + 0.173874078 × H-PHE-PRO-OH + 0.581385965 × Thyroxine + 0.343568892 × ACar (13: 0) + 0.144166563 × ACar (13: 1) - 0.979911099 × 2-Methylglutaric acid acid+0.112680502×N1-Methyl-2-pyridone-5-carboxamide+0.20221917×Dethiobiotin+0.551895565×3-H ydroxyisovalericacid-0.710781395×[7]-Paradol-0.298712382×I-Urobilin+1.325241453×(-)-Catechin 3-O-gallate+0.673871343×Isopentenyl acetate+0.038857388×Desglucocoroloside+0.446201899×Spirotaccagenin.

[0105] Step S5: The materials obtained from the logistic regression, LASSO, support vector machine, and random forest models are combined to draw a joint ROC curve. The larger the AUC value, the better the effect of the model. The optimal model is determined to obtain a tumor radiotherapy sensitivity marker.

[0106] The present invention screens and detects 13 kinds of tumor radiotherapy sensitivity metabolic markers: oleamide (CAS: 3072-13-7), dihydrofolicacid (CAS: 4033-27-6), H-PHE-PRO-OH (CAS: 7669-65-0), 2-Methylglutaricacid (CAS: 617-62-9), phenylglyoxylic acid (CAS: 611-73-4), thyroxine (CAS: 51-48-9), 1-Methyl-L-histidine (CAS: 332-80-9), uridine (CAS: 58-96-8), D-glutamine (CAS: 5959-95-5), deoxycholic acid (CAS: 83-44-3), lauroyl diethanolamide (CAS: 120-40-1), (R)-3-hydroxy-tetradecanoic Acid (CAS: 28715-21-1) and Cholic acid (CAS: 81-25-4). The CAS number in brackets is the internationally used CAS number.

[0107] The panel composed of the above 13 metabolic markers had the largest AUC (area under the curve) value, with prHS vs prLS AUC reaching 0.932, prHS vs prMS AUC being 0.817, prMS vs prLS AUC being 0.931, and prHS+prMS vs prLS AUC reaching 0.914.

[0108] The above 13 serum metabolite markers obtained by the present invention are expected to become fixed markers for determining the sensitivity of colorectal cancer to radiotherapy, and the judgment is made by collecting patient serum and detecting the above metabolites.

[0109] Example 2

[0110] A method for screening and detecting bacterial flora markers for tumor radiotherapy sensitivity, comprising the following steps:

[0111] 1.1 Fecal DNA extraction

[0112] (1) Add 0.25-0.5 g of sample to a 2 mL centrifuge tube. If it is a liquid sample, transfer 200 μL to the centrifuge tube, add 500 μL buffer SA, 100 μL buffer SC and 0.25 g grinding beads. (There may be residual RNA in the fecal sample. If RNA needs to be removed, it is recommended to add 10 μL RNase A (TIANGEN, RT405-02, self-prepared)) Vortex to mix or use a tissue homogenizer to mix, then heat and lyse at 70°C for 15 minutes to improve lysis efficiency. Centrifuge at 12000 rpm for 1 minute, and transfer the first supernatant (about 500 μL) to a new 2 mL centrifuge tube.

[0113] (2) Add 200 μL of buffer SH to the first supernatant, mix well, vortex for 5 seconds, and place at 4°C for 10 minutes.

[0114] (3) Centrifuge at 12000 rpm for 3 min, transfer the second supernatant to a new 2 mL centrifuge tube, add 500 μL of buffer GFA (please check whether isopropanol has been added before use), and mix by inversion.

[0115] (4) Add 10 μL of magnetic bead suspension G and vortex to mix for 5 min.

[0116] (5) Place the centrifuge tube on the magnetic rack and let it stand for 30 seconds. After the magnetic beads are completely adsorbed, carefully remove the liquid.

[0117] (6) Remove the centrifuge tube from the magnetic rack, add 700 μL of deproteinized solution RD (please check whether anhydrous ethanol has been added before use), and shake to mix for 5 minutes.

[0118] (7) Place the centrifuge tube on the magnetic rack and let it stand for 30 seconds. After the magnetic beads are completely adsorbed, carefully remove the liquid.

[0119] (8) Remove the centrifuge tube from the magnetic rack, add 700 μL of PWD (please check whether anhydrous ethanol has been added before use), and shake to mix for 3 min.

[0120] (9) Place the centrifuge tube on the magnetic rack and let it stand for 30 seconds. After the magnetic beads are completely adsorbed, carefully remove the liquid.

[0121] (10) Repeat steps (8) and (9) once.

[0122] (11) Place the centrifuge tube on a magnetic rack and dry it at room temperature for 5 to 10 minutes.

[0123] (12) Remove the centrifuge tube from the magnetic rack, add 50-100 μL of elution buffer TB, vortex to mix, and incubate at 56°C for 5 min. During this time, vortex to mix three times, 3-5 times each time.

[0124] (13) Place the centrifuge tube on a magnetic rack and let it stand for 2 min. After the magnetic beads are completely adsorbed, carefully transfer the DNA solution to a new centrifuge tube and store it under appropriate conditions to obtain fecal DNA.

[0125] 1.2 Introduction to full-length amplicon

[0126] 16S rDNA is a gene that encodes the ribosome 30S small subunit component (16S rRNA) in prokaryotes. It is about 1500bp in length and includes 9 variable regions (V1-V9) and 10 conserved regions. The conserved regions reflect the relationship between species, while the variable regions reflect the differences between species. The degree of variation is closely related to the bacterial phylogeny and is considered to be the most suitable indicator for bacterial phylogeny and classification identification. Figure 1 .

[0127] Unlike the second-generation amplicons that only target 1-2 hypervariable regions, the third-generation amplicons sequence the full length of 16S (the amplification primer pair is 27F: AGRGTTYGATYMTGGCTCAG (SEQ ID NO.1) and 1492R: RGYTACCTTGTTACGACTT (SEQ ID NO.2)), where R represents A / G, Y represents C / T, and M represents A / C, avoiding the influence of different hypervariable regions and PCR preference. More hypervariable region information can significantly improve the resolution and accuracy of species annotation, and has been widely used in the study of microbial communities in different habitats.

[0128] 1.3 Introduction to Sequencing Principle

[0129] The single molecule real-time (SMRT) technology of PacBio (Pacific BioSciences) has the advantages of long read length and high accuracy; the Sequel system is the current mainstream third-generation sequencing platform. There are a large number of nanoscale holes on the SMRT cell of its sequencing reaction, called zero-mode waveguides (ZMWs). Usually, a template DNA and a molecule of DNA polymerase are fixed in each hole, and the sequence information is determined by detecting the fluorescent signal of dNTPs during the chain extension process. The principle of PacBio sequencing can be found in Figure 2 .

[0130] The Circular Consensus Sequencing (CCS) mode performs multiple sequencing on each template molecule and obtains accurate sequence information through subread correction, namely CCS reads (also called HiFi reads); it is currently mainly used for libraries with insert fragments less than 2kb, which is highly consistent with the requirements of 16S rDNA full-length sequencing. Figure 3 .

[0131] 2. Workflow

[0132] 2.1 Experimental Procedure

[0133] In order to ensure the accuracy and reliability of sequencing data from the source, there are rigorous and reliable sample testing and quality control processes from DNA extraction to sequencing. The sample quality is strictly controlled at every stage to ensure the authenticity and reliability of the sequencing data. Figure 4 .

[0134] 2.2 Information Analysis Process

[0135] After the sequencing is completed, the original data Raw CCS obtained will be quality controlled to obtain high-quality ValidCCS. In order to meet the needs of different customers, two different analysis modes can be provided in the future. One is based on Qiime2's DADA2 (Divisive Amplicon Denoising Algorithm), which obtains representative sequences ASVs (Amplicon Sequence Variants) with single-base accuracy through "dereplication"; the other is to use VSEARCH for OTUs (Operational Taxonomic Units) clustering (97% similarity). At present, the former is more widely recognized and recommended due to its higher phylogenetic resolution and comparability between different data sets. The project uses the ASVs method by default. After obtaining the ASVs / OTUs table, further diversity analysis, species classification annotation and difference analysis are carried out.

[0136] 3. Standard Analysis

[0137] 3.1 Sequencing data quality control statistics

[0138] After sequencing, first obtain Raw CCS (circular consensus sequencing), and then perform further data quality control to obtain ValidCCS. The steps are as follows:

[0139] (1) SMRTLinks identification to obtain Raw CCS, parameters: minPasses ≥ 3, minPredictedAccuracy ≥ 0.99;

[0140] (2) limma uses default parameters to identify barcode split samples;

[0141] (3) cutadapt identified and removed the primers and then performed length screening to retain reads of 1200-1650 bp;

[0142] (4) The final Valid CCS was denoised and chimeras were filtered using the qiime dada2 denoise-single method to obtain the ASV (feature) feature sequence and abundance table, and singletons ASVs were removed to obtain the processed Valid CCS.

[0143] 3.2 Screening of markers

[0144] The processed Valid CCS is then used to screen markers.

[0145] The processed Valid CCS was classified into 6 categories according to the NAR (Neoadjuvant Rectal) score and before and after radiotherapy: prHS, prMS, prLS, poHS, poMS and poLS. Specifically, prHS stands for sensitive data before radiotherapy (highly sensitive bacterial community subset before radiotherapy), prMS stands for moderately sensitive data before radiotherapy (moderately sensitive bacterial community subset before radiotherapy), prLS stands for insensitive data before radiotherapy (insensitive bacterial community subset before radiotherapy), poHS stands for sensitive data after radiotherapy (highly sensitive bacterial community subset after radiotherapy), poMS stands for moderately sensitive data after radiotherapy (moderately sensitive bacterial community subset after radiotherapy), and poLS stands for insensitive data after radiotherapy (insensitive bacterial community subset after radiotherapy), which were used as the initial sample set.

[0146] Step S1: The initial sample set is processed by principal component analysis and principal coordinate analysis (PCoA) to obtain an initial differential flora sample set; the initial differential flora sample set is preliminarily screened by rank sum test to obtain a second differential flora sample set.

[0147] First, establish the null hypothesis: the two groups of samples come from the same population. Mix the two groups of data and rank them uniformly from small to large. When encountering the same data, take the average rank and calculate the statistic. Set the significance level a=0.05, and obtain the probability of the current statistic value under the condition that the null hypothesis is true. If p-value<0.05, reject the null hypothesis, indicating that there is a significant difference between the two samples.

[0148] Step S2: Based on the logistic regression model and P-value established between the differential samples in the second differential flora sample set and the significant clinical indicators, the differential flora after eliminating the interference of clinical factors is obtained, that is, the third differential flora sample set is obtained.

[0149] The initial clinical indicators were screened using multiple comparisons and PerMANOVA multivariate analysis of variance, and significant clinical indicators were obtained, including:

[0150] a. Obtaining initial clinical indicators:

[0151] A total of 271 clinical samples were collected, and the clinical indicator information was extracted, including treatment response group, age, gender, BMI, smoking, drinking, T, N, GTV, distance from the anus in cm (MR / CT), CRM (MR), EMVI (MR), NLR, concurrent chemotherapy regimen, hormone use, and antibiotic use, a total of 16 items.

[0152] b. Screening the initial clinical indicators using the multiple comparison method to determine the second clinical indicator:

[0153] The multiple comparison method was used to retain the samples with p-value < 0.05 in the statistical test results of the above 16 clinical indicators. The p-value was obtained by performing multiple comparisons on the clinical indicators among multiple experimental groups, firstly testing their homogeneity by Bartlett Test (corresponding to the R language program function bartlett.test) (taking p-value < 0.05 as a non-null hypothesis), and then calculating their p-values ​​by the oneway.test function of the R language program package. If p < 0.05, it means that the clinical indicator is significant.

[0154] c. Use the PerMANOVA multivariate analysis of variance method to screen the second clinical indicator and determine the significant clinical indicator.

[0155] The permaonva analysis method was used to further screen out clinical indicator information that had a significant relationship with the expression of the flora profile of human fecal samples (i.e., the second clinical indicator);

[0156] d. The significant clinical index information obtained in step c is used as the response variable to perform a final correction on the second differential flora sample set obtained in step S3 above, and the differential flora whose p-value after regression is still less than 0.05 is screened out to obtain the differential flora (third differential flora sample set) after eliminating the interference of clinical factors, specifically including:

[0157] The clinically significant index information obtained in step c is used as the response variable, and the second differential flora sample set obtained in step S2 is used as the independent variable. A logistic regression model is established to obtain the p-value. If the p-value of the differentially expressed flora is less than 0.05, it means that these floras are not affected by the clinical indicators, that is, they are differential floras after eliminating the interference of clinical factors.

[0158] e. For the above four comparison strategies of prHS and prMS, prHS and prLS, prMS and prLS, and prHS+prMS and prLS, the differential flora after excluding the interference of clinical factors was obtained, that is, the third differential flora sample set was obtained. Three different machine learning algorithms, namely random forest method, support vector machine method, and LASSO, were used for further screening and analysis. Based on the key candidate differential flora combinations finally screened out by these three different algorithms, a joint ROC analysis was carried out. Finally, the final differential flora combination panel was determined according to the obtained AUC value, specifically:

[0159] First, the third differential flora sample set was divided into a training set and a test set (the ratio of training set to test set was 8:2), and different algorithms were used to screen differential flora, as follows:

[0160] Random forest method: Use the R package randomForest to fit the random forest model to the training set, and finally determine the treeNum in the fitting model through parameter tuning. Fit the training set with the optimal parameters, and test the test set to obtain the confusion matrix of the prediction results. Further, use the rfcv function in the R package randomForest to perform 10-fold cross validation to obtain the relationship between variable selection and the average error rate of model classification, and obtain the optimal differential flora based on MeanDecreaseAccuracy. The larger the value, the greater the importance of the variable.

[0161] Support vector machine (SVM): Support vector machine is a two-class classification model. Its basic model is a linear classifier that finds the separating hyperplane with the maximum margin in the feature space. SVM is an algorithm for the post-selection of a sequence based on the maximum margin principle. In the first iteration, the SVM model is optimized for all feature sets in the data set. Then the score of each feature is calculated and sorted in descending order. The feature set with the smallest score is recorded, and then the feature with the smallest score is deleted. It is iterated again until only one feature is left. The training set is trained using the svm function of the R language to obtain the relationship between variable selection and the average error rate of model classification, and the feature set with the lowest error rate is selected.

[0162] LASSO: LASSO imposes regression penalties on all variables, making the coefficients of relatively unimportant variables become 0, thus being excluded from modeling, and then selecting independent variables that have a greater impact on the dependent variable and calculating the corresponding regression coefficients, and finally obtaining a prediction model. The model is trained on the training set using the R language glmnet. The lambda value with the smallest mean square error is selected through the mean square error under different parameters lambda. A vertical line is drawn through the position of the corresponding penalty value of the change trajectory diagram of each independent variable coefficient. Each curve represents a variable, and the variable that intersects with the penalty value is the variable finally included in the model. The ordinate corresponding to the variable is the regression coefficient of the variable, and the regression coefficient is also the contribution of the variable.

[0163] Secondly, the differential bacterial communities were screened out through random forest, support vector machine and LASSO. The data of these differential bacterial communities were found in the normalized data. The differentially expressed bacterial communities were used as the response variables, and the groups were divided into 1 as the experimental group and 0 as the control group as the dependent variables to establish a logistic regression model.

[0164] Finally, the probability given by logistic regression was calculated to draw the joint ROC curve. The larger the AUC value, the better the classification effect of the model. The optimal differential flora combination panel was determined based on the size of the AUC.

[0165] Step S3: Using the optimal differential flora combination panel and combining the k-fold cross-validation method, the logistic regression model, the random forest model, the support vector machine and the LASSO model are trained respectively to obtain a trained logistic regression model, a trained random forest model, a trained support vector machine and a trained LASSO model.

[0166] First, the best differential flora combination panels were determined for the three different comparison strategies of prHS and prMS, prHS and prLS, and prMS and prLS. Then, the best differential flora combination panels were merged and randomly sampled into training sets and test sets, with a ratio of 7:3 for training sets: test sets.

[0167] Secondly, for the training sets of different comparison strategies, the caret package was used to perform 10-fold cross-validation for the logistic regression, LASSO, support vector machine, and random forest models, respectively, with 1000 repetitions. The cross-validation was mainly performed using trainControl in the caret package in the R language.

[0168] The steps involved in the 10-fold cross validation are:

[0169] (1) Randomly divide the dataset into 10 subsets, (2) For each selected subset of data points, use that subset as a validation set; use all remaining subsets for training purposes; train the model and evaluate it on the validation set or test set; calculate the prediction error, (3) repeat the above steps 10 times, (4) generate the overall prediction error by taking the average of the prediction errors.

[0170] Step S4: Using ROC analysis, according to the AUC value, use the train function to train the logistic regression, LASSO, support vector machine, and random forest models respectively, calculate the prediction probability of the test set under the trained model, and draw the ROC curves of different models at the same time. The larger the AUC value, the better the model effect, and determine the optimal model. Finally, merge the microbial flora obtained by the logistic regression, LASSO, support vector machine, and random forest models to obtain microbial flora markers for screening and detecting tumor radiotherapy sensitivity.

[0171] Finally, the following 12 microbial combinations were obtained: s__Schaalia_odontolytica, s__Roseburia_inulinivorans, s__Mogibacterium_diversum, s__Campylobacter_gracilis, s__Alistipes_obesi, s__bacterium_New, s__Prevotella_nigrescens, s__Clostridioides_difficile, s__Bacteroides_clarus, s__Enterobacter_cloacae, s__Anaerostipes_unclassified, and s__Prevotella_stercorea. The panel had the largest AUC (area under the curve) value. The AUC value of the 12 differential bacterial communities in prHS vs prLS was 0.890, the AUC value in prHS vs prMS was 0.552, and the AUC value in prMS vs The AUC value in prLS can reach 0.798, revealing that 12 intestinal flora have potential predictive value for radiotherapy sensitivity in patients with colorectal cancer. It is expected that they can be used as fixed markers for prediction by collecting patient feces and performing 16s rRNA high-throughput sequencing analysis.

[0172] Example 3

[0173] A combined marker for detecting tumor radiotherapy sensitivity, the combined marker comprising a first combined marker and a second combined marker;

[0174] The first combination marker is Dihydrofolic acid, (R)-3-Hydroxy-tetradecanoic acid, 2-Methylglutaric acid in the serum metabolic markers described in Example 1 and s__Roseburia_inulinivorans in the flora markers described in Example 2.

[0175] The second combination marker is Cholic acid, (R)-3-Hydroxy-tetradecanoic acid, Dihydrofolic acid, D-Glutamine, Deoxycholic acid, Thyroxine, Phenylglyoxylic acid among the serum metabolic markers described in Example 1 and s__Roseburia_inulinivorans and s__Prevotella_nigrescens among the microbial markers described in Example 2.

[0176] This example uses the critical value of the logistic regression model constructed using the above combination of markers to predict the radiotherapy sensitivity of colorectal cancer patients.

[0177] Among them, the reagents are divided into serum and stool samples to be tested, samples of colorectal cancer patients who are insensitive to radiotherapy, and samples of colorectal cancer patients who are moderately sensitive to radiotherapy.

[0178] Judgment method: The logistic regression model constructed by Metaboanalyst6.0 (https: / / www.metaboanalyst.ca / ) was used to predict the optimal threshold of P for the samples to be tested vs. the samples of colorectal cancer patients who were insensitive to radiotherapy. When P < 0.95, it indicated that the tumor patient was insensitive to radiotherapy, and when P ≥ 0.95, it indicated that the tumor patient was moderately sensitive to radiotherapy or highly sensitive to radiotherapy.

[0179] In order to further determine whether the tumor patients are moderately sensitive or highly sensitive to radiotherapy, a logistic regression model was constructed by comparing the test samples with P ≥ 0.95 with the samples of colorectal cancer patients with moderate sensitivity to radiotherapy to predict the optimal threshold of P. When P < 0.52, it indicates that the tumor patients are moderately sensitive to radiotherapy, and when P ≥ 0.52, it indicates that the tumor patients are highly sensitive to radiotherapy.

[0180] The inclusion criteria for the study subjects (tumor patients) included: (1) patients who received surgical treatment after neoadjuvant radiotherapy or chemoradiotherapy; (2) patients whose radiotherapy regimen was long-term radiotherapy. The exclusion criteria included: (1) patients with any other type of malignant tumor or severe immune, nervous, digestive, or blood system diseases; (2) patients whose plasma samples could not be obtained; (3) patients with colorectal adenocarcinoma that had metastasized; and (4) patients whose radiotherapy regimen was short-term radiotherapy.

[0181] In this embodiment, 24 stool samples were tested, 24 serum samples were tested, 32 serum samples of colorectal cancer patients who were insensitive to radiotherapy, 31 stool samples of colorectal cancer patients who were insensitive to radiotherapy, 59 serum samples of colorectal cancer patients who were moderately sensitive to radiotherapy, and 60 stool samples of colorectal cancer patients who were moderately sensitive to radiotherapy. Among them, colorectal cancer patients who were insensitive to radiotherapy refer to patients with NAR scores > 16, colorectal cancer patients who were moderately sensitive to radiotherapy refer to patients with NAR scores of 8 to 16, and NAR scores are significantly correlated with overall survival (OS).

[0182] The logistic regression model constructed using the first combination marker for the sample to be tested vs the samples of colorectal cancer patients who are insensitive to radiotherapy is logit(P)=log(P / (1-P))=2.903+4.886×Dihydrofolic acid+5.667×s__Roseburia_inulinivorans+3.344×(R)-3-Hydroxy-tetradecanoic acid-6.263×2-Methylglutaric acid. The optimal threshold (or cutoff value) for predicting P is 0.95. When P<0.95, it indicates that the tumor patient is poorly sensitive to radiotherapy. When P≥0.95, it indicates that the tumor patient is highly or moderately sensitive to radiotherapy.

[0183] The logistic regression model of the samples to be tested with P ≥ 0.95 vs the samples of colorectal cancer patients with moderate sensitivity to radiotherapy constructed using the second combination marker is logit(P) = log(P / (1-P)) = -0.219 + 188.491 × Cholic acid + 76.134 × (R)-3-Hydroxy-tetradecanoic acid + 42.197 × Dihydrofolic acid + 109.16 × D-Glutamine + 2373.987 × s__Roseburia_inulinivorans + 261.411 × Deoxycholic acid + 34.567 × Thyroxine + 32.912 × Phenylglyoxylic acid+2367.977×s__Prevotella_nigrescens, the optimal threshold (or cutoff value) for predicting P is 0.52. When P < 0.52, it indicates that the tumor patient is moderately sensitive to radiotherapy, and when P ≥ 0.52, it indicates that the tumor patient is highly sensitive to radiotherapy. Colorectal cancer patients with high sensitivity to radiotherapy refer to those with NAR scores < 8.

[0184] At the same time, the AUC curve is drawn according to the logistic regression model fitting method. Figure 5 The results showed that the sample AUC for the tested samples vs. the colorectal cancer patients who were insensitive to radiotherapy reached 1.000, and the sample AUC for the tested samples with P ≥ 0.95 vs. the colorectal cancer patients who were moderately sensitive to radiotherapy was 0.949.

[0185] Draw the AUC curve based on the substances involved in the logistic regression model equation. Figure 6 The results showed that the AUC for diagnosing prHS+prMS vs prLS using the combined markers in the present invention was 0.838, and the AUC for prHS vs prMS was 0.828.

[0186] The relative levels of serum metabolites and flora of the group of colorectal cancer patients with radiotherapy insensitive in Example 1 were brought into the logistic regression model of the above-mentioned samples to be tested vs. the samples of colorectal cancer patients with radiotherapy insensitive to verify their grouping, and the P values ​​of each sample were 0.882075, 0.882076, 0.882076, 0.882075, 0.88207, 0.88207, and 0.88207, respectively, all meeting P < 0.95. The verification results also prove that tumor patients have low sensitivity to radiotherapy.

[0187] In summary, the combined markers of the present invention have high accuracy in detecting tumor radiotherapy sensitivity.

[0188] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A combined marker for detecting tumor radiotherapy sensitivity, characterized in that: The combination marker is a first combination marker and a second combination marker; The first combination marker is Dihydrofolic acid, (R)-3-Hydroxy-tetradecanoic acid, 2-Methylglutaric acid and s__Roseburia_inulinivorans; The second combination marker is Cholic acid, (R)-3-Hydroxy-tetradecanoic acid, Dihydrofolic acid, D-Glutamine, Deoxycholic acid, Thyroxine, Phenylglyoxylicacid, s__Roseburia_inulinivorans and s__Prevotella_nigrescens.

2. Use of the combined marker according to claim 1 in the preparation of a product for diagnosing tumor radiotherapy sensitivity.

3. Use of the combined marker according to claim 1 in the preparation of a product for diagnosing and grading tumor radiotherapy sensitivity.

4. The use according to claim 3, characterized in that: The tumor radiotherapy sensitivity classification is divided into tumor radiotherapy high sensitivity and tumor radiotherapy moderate sensitivity.

5. A system for predicting tumor radiotherapy sensitivity using the combined marker of claim 1, characterized in that: It includes a nucleic acid sample separation unit, which is used to separate a fecal flora nucleic acid sample from a test sample; A data processing unit, using the combined markers of claim 1 to establish a logistic regression model to obtain a critical value; The result determination unit is used to compare the critical value obtained by the data processing unit with the set diagnosis value.

6. The system according to claim 5, characterized in that The logistic regression model is a first logistic regression model and a second logistic regression model; The first logistic regression model is a logistic regression model for constructing stool samples to be tested vs stool samples of colorectal cancer patients who are insensitive to radiotherapy using the first combination marker described in claim 1, when the critical value is <0.95, it indicates that the tumor patient is insensitive to radiotherapy, and when the critical value is ≥0.95, it indicates that the tumor patient is moderately sensitive to radiotherapy or highly sensitive to radiotherapy; The second logistic regression model is a logistic regression model constructed by using the second combination marker described in claim 1 to construct a stool sample to be tested with a critical value ≥0.95 vs. a stool sample of a colorectal cancer patient with moderate sensitivity to radiotherapy. When the critical value is <0.52, it indicates that the tumor patient is moderately sensitive to radiotherapy, and when the critical value is ≥0.52, it indicates that the tumor patient is highly sensitive to radiotherapy.

Images