WO2020244017A1 - Intestinal flora-based schizophrenia biomarker combination, and applications thereof and motu screening method therefor - Google Patents

Abstract

Disclosed are an intestinal flora-based schizophrenia biomarker combination and applications. On the basis of the influence on biomarkers before and after use of candidate drugs, whether the candidate drugs can be used for treating or preventing schizophrenia is determined.

Description

A schizophrenia biomarker combination based on intestinal flora, its application and mOTU screening method Technical field

The invention belongs to the technical field of biomedicine, and relates to a biomarker for schizophrenia based on intestinal flora, a kit for diagnosing or predicting the risk of schizophrenia, and applications.

Background technique

Schizophrenia (English: Schizophrenia) is a group of severe mental illnesses of unknown etiology, mostly slow or subacute onset in young adults, clinically often manifested as syndromes with different symptoms, involving perception, thinking, emotion and behavior Various obstacles and incoordination of mental activities. Patients generally have clear consciousness and normal intelligence, but some patients will suffer from cognitive impairment during the course of the disease. The course of the disease is generally protracted, showing repeated attacks, aggravation or deterioration. Most patients will eventually experience decline and mental disability, and only a few patients will be cured or basically cured after treatment.

The global prevalence of schizophrenia is about 0.3-0.7%. As of 2016, there are estimated to be more than 21 million people with schizophrenia in the world, and their average life expectancy is 10 to 25 years shorter than that of normal people.

Although previous studies have shown that the onset of schizophrenia is caused by a combination of genetic and environmental factors, and that the patient’s serum and brain tissue have some abnormal changes, the current diagnosis of schizophrenia still relies on symptomatic evaluation. Reliable biomarker identification. In addition, the existing diagnostic criteria cannot early predict the onset, efficacy and prognosis of schizophrenia.

Summary of the invention

The problem solved by the present invention is to provide a combination of biomarkers for schizophrenia based on intestinal flora and its application, which can overcome the inability of the existing diagnosis of schizophrenia to achieve early warning, fail to predict the incidence and development trend of schizophrenia, etc. Disadvantages, it can help the pathological classification of diseases and the study of drug targets, precise medication, and pathogenesis.

The present invention is realized through the following technical solutions:

A biomarker combination for schizophrenia based on intestinal flora, which is used to provide relative abundance information, and includes one or more selected from the following:

Biomarker 1: Lachnospiraceae bacterium 3_1_57FAA_CT1;

Biomarker 2: Cronobacter sakazakii;

Biomarker 3: Lactobacillus acidophilus;

Biomarker 4: Veillonella parvula;

Biomarker 5: Lactococcus lactis;

Biomarker 6: Alkaliphilus oremlandii;

Biomarker 7: Pseudoflavonifractor capillosus;

Biomarker 8: Streptococcus gallolyticus;

Biomarker 9: Dialister invisus;

Biomarker 10: Lactobacillus johnsonii;

Biomarker 11: Methanobrevibacter smithii.

The relative abundance information provided by the biomarker combination is used for comparison with reference values.

The relative abundance information of the biomarkers 1-11 is provided based on the gene sequence for which abundance calculation can be performed.

The combination of biomarkers for schizophrenia based on intestinal flora is used as a detection target or a detection target in the preparation of a detection kit.

The method for screening a combination of biomarkers for schizophrenia based on intestinal flora includes the following steps:

1) Sample collection: After collecting stool samples, they are frozen and transported and quickly transferred to -80°C for storage. DNA extraction is performed to obtain extracted DNA samples. The sample subjects include schizophrenia patients and healthy people;

2) Metagenomic sequencing and assembly

3) Conservative single-copy gene alignment and abundance calculation

3) Input high-quality sequencing fragments into the software mOTU to calculate the relative abundance of species:

3.1) Align the high-quality sequencing fragments to the reference single-copy gene;

3.2) Count the number of inserts according to the comparison results;

3.3) Normalize the number of inserts to the length of a single copy gene to obtain the corresponding abundance.

4) Randomly select schizophrenic patients and healthy people from the sample set as the training set, and the remaining samples as the validation set. Calculate the relative abundance of mOTU in each sample in the training set, and then input the mOTU of the training set into the random forest classifier. Perform 5 10-fold cross-validation on the classifier, use the relative abundance of mOTU screened by the RF model to calculate the risk of schizophrenia for each individual, draw the ROC curve, and calculate the AUC as the performance evaluation parameter of the discriminant model, and select the marker The number of combination of substances is less than 30, and the combination with the best discrimination performance, the importance index of each mOTU is output in the model. The higher the importance index, it represents the importance of the marker used to distinguish schizophrenia and non-schizophrenia Higher.

In the sample collection, the sample subjects included 90 schizophrenia patients and 81 healthy people, and the verification collection, the sample subjects included 10 schizophrenia patients and 10 healthy people.

The method of using the above markers, that is, to diagnose whether the subject has schizophrenia or predict whether the subject has the risk of schizophrenia is:

1) Collect samples from the subject;

2) Determine the relative abundance information of the biomarkers in the sample obtained in step 1);

3) Compare the relative abundance information described in step 2) with the reference data set or reference value. The method can be used not only for disease diagnosis in the sense of patent law, but also for non-disease diagnosis such as scientific research or other personal genetic information enrichment and genetic information database enrichment. The relative abundance information of each biomarker in the test subject is compared with a reference data set or reference value to determine whether the subject has schizophrenia, or predict the risk of schizophrenia.

The reference data set includes relative abundance information of biomarkers in samples from multiple schizophrenia patients and multiple healthy controls.

The reference data set refers to the relative abundance information of each biomarker obtained by operating the samples of the diagnosed diseased individuals and healthy individuals, which is used as a reference for the relative abundance of each biomarker . Specifically, the reference data set may refer to the training data set. According to the present invention, the training set refers to and the verification set has a well-known meaning in the art. In an embodiment of the present invention, the training set refers to a data set containing the content of each biomarker in the test samples of a certain number of schizophrenic subjects and non-schizophrenic subjects. The verification set is an independent data set used to test the performance of the training set.

The reference value in the present invention refers to the reference value or normal value of healthy controls. Those skilled in the art know that when the sample volume is large enough, detection and calculation methods known in the art can be used to obtain the normal value (absolute value) range of each biomarker in the sample. When an assay method is used to detect the level of a biomarker, the absolute value of the biomarker level in the sample can be directly compared with a reference value to assess the risk of disease and diagnose or early diagnose schizophrenia. Optionally, you can Including statistical methods.

The step of comparing the relative abundance information with the reference data set in step 2) further includes executing a multivariate statistical model to obtain the disease probability. The use of multivariate statistical models can achieve fast and efficient detection. Specifically, the multivariate statistical model is a random forest model.

The probability of illness being greater than the threshold indicates that the subject has schizophrenia or related diseases or is at risk of suffering from schizophrenia or related diseases. Specifically, the threshold is 0.5.

The relative abundance information of the biomarker in step 2) is obtained by a sequencing method, and further includes: isolating a nucleic acid sample from the sample of the subject, and constructing a DNA library based on the obtained nucleic acid sample , Sequencing the DNA library to obtain a sequencing result; and comparing the sequencing result with a reference gene set based on the sequencing result to determine the relative abundance information of the biomarker.

According to an embodiment of the present invention, at least one of SOAP2 and MAQ can be used to compare the sequencing result with the reference gene set, thereby improving the efficiency of the comparison, and thus the efficiency of detecting schizophrenia. According to the embodiment of the present invention, multiple (at least two) biomarkers can be detected at the same time, which can improve the efficiency of schizophrenia detection.

The reference gene set includes performing metagenomic sequencing from samples of multiple schizophrenia patients and multiple healthy controls to obtain a non-redundant gene set, and then combining the non-redundant gene set with the gut microbial gene set, Obtain the reference gene set. The reference gene set in the present invention can be an existing gene set, such as an existing intestinal microbial reference gene set that has been published; it can also be metagenomic sequencing of samples from multiple schizophrenia patients and multiple healthy controls , Obtaining a non-redundant gene set, and then combining the non-redundant gene set with the gut microbial gene set to obtain the reference gene set, and the obtained reference gene set information is more comprehensive and the detection result is more reliable.

The non-redundant gene set is explained as a person skilled in the art generally understands, and in simple terms is the set of remaining genes after the redundant genes are removed. Redundant genes usually refer to multiple copies of a gene that appear on a chromosome.

Specifically, the sample is a stool sample. The sequencing method is performed by a second-generation sequencing method or a third-generation sequencing method. The means for sequencing is not particularly limited. Fast and efficient sequencing can be achieved by second-generation or third-generation sequencing methods.

The sequencing method is performed by at least one selected from the group consisting of Hiseq2000, SOLiD, 454, and single molecule sequencing devices. As a result, the high-throughput and deep sequencing characteristics of these sequencing devices can be utilized, thereby facilitating the analysis of subsequent sequencing data, especially the accuracy and accuracy of statistical testing.

The present invention proposes the application of the schizophrenia biomarker combination based on the intestinal flora as a detection target or a detection target in the preparation of a detection kit for diagnosing whether a subject has schizophrenia Or related diseases or predict whether the subject has a risk of schizophrenia or related diseases.

That is, the present invention proposes a kit that includes reagents for detecting biomarkers. With the kit, the relative abundance of these markers in the intestinal flora can be determined, and thus, the obtained The relative abundance value is used to determine whether the subject has or is susceptible to schizophrenia, and the efficiency of the treatment effect for monitoring schizophrenia patients.

The present invention proposes the application of a combination of biomarkers for schizophrenia based on intestinal flora as a target in the screening of drugs for the treatment and/or prevention of schizophrenia. The biomarkers are the above-mentioned biomarkers of the present invention, and the influence of the candidate drug on these biomarkers before and after use can be used to determine whether the candidate drug can be used to treat or prevent schizophrenia.

The change in the relative abundance of the biomarker combination provides a basis for determining whether the candidate drug is effective.

Compared with the prior art, the present invention has the following beneficial technical effects:

The present invention is based on the discovery and recognition of the following facts and problems: intestinal microbes are the microbial community that exists in the human intestinal tract and are the "second genome" of the human body. The human intestinal flora and host constitute an interconnected whole. Gut microbes can produce most of the neurotransmitters found in the human brain. More and more evidence supports the view that intestinal microbes affect central neurochemistry and behavior. Irritable bowel syndrome is considered a typical case of brain-gut microbial axis regulation disorders. Translational studies have shown that certain specific flora may have an impact on stress response and cognitive function. Using probiotics or antibiotics to change the intestinal microbiota provides a new method for improving brain function and treating depression and autism and other intestinal-brain axis diseases. Therefore, the present invention analyzes the intestinal flora and gene sequences of patients with schizophrenia and healthy people, thereby screening biomarkers with high correlation with schizophrenia, and using the markers can accurately diagnose schizophrenia Or predict the risk of disease, and can be used to monitor the effect of treatment.

Stool is a metabolite of the human body. It not only contains metabolites of the human body, but also includes intestinal microbes that are closely related to changes in our body's metabolism and immunity and brain function. Research on feces found that there are obvious differences in the composition of the intestinal flora between schizophrenia patients and healthy people, which can accurately assess the risk of schizophrenia and make early diagnosis. The present invention is based on the comparison and analysis of the intestinal flora of schizophrenia patients and healthy people, and obtains a variety of related intestinal microbes. Combining high-quality mOTU of schizophrenia and healthy people as a training set, it can accurately treat mental Patients with schizophrenia undergo risk assessment and early diagnosis. Compared with the current commonly used diagnostic methods, this method has the characteristics of convenience and speed.

The schizophrenia-related biomarkers proposed by the present invention are valuable for early diagnosis. First, the marker of the present invention has high specificity and sensitivity. Second, the analysis of stool ensures accuracy, safety, affordability and patient compliance. And the stool sample is transportable. Polymerase chain reaction (PCR)-based tests are comfortable and non-invasive, so it is easier for people to participate in a given screening procedure. Third, the marker of the present invention can also be used as a tool for treatment monitoring of patients with schizophrenia to detect the response to treatment. Due to the reasons of abundance measurement and population selection, the combination of 11 markers in the present invention is particularly suitable for the measurement of abundance based on the mOTU method of fecal mOTU of people in northwestern China.

Description of the drawings

Figure 1 shows the β diversity of patients with schizophrenia and healthy controls at the species level according to an embodiment of the present invention. The graphic shows that there is a significant difference between schizophrenic patients and healthy controls.

Fig. 2 shows the error rate distribution of five 10-fold cross-validation in a random forest classifier according to an embodiment of the present invention.

Fig. 3 is a receiver operating characteristic (ROC) curve and curve of a training set composed of schizophrenia patients and healthy controls based on a random forest model (11 intestinal markers) according to an embodiment of the present invention Area under Curve (AUC).

Figure 4 shows the ROC curve and AUC of a validation set composed of schizophrenia patients and healthy controls based on a random forest model (11 gut markers) according to an embodiment of the present invention.

Detailed ways

The terms used in the present invention have meanings commonly understood by those of ordinary skill in the relevant art. However, in order to better understand the present invention, some definitions and related terms are explained as follows:

"Schizophrenia" is a group of severe mental illnesses of unknown etiology, mostly in young adults with slow or subacute onset, clinically often manifested as syndromes with different symptoms, involving perception, thinking, emotion, and behavior. Obstacles and incoordination of mental activities.

"Biomarker", also known as "biological marker", refers to a measurable indicator of the biological state of an individual. Such biomarkers can be any substance in the individual, as long as they are related to the specific biological state (e.g., disease) of the subject, for example, nucleic acid markers (also called genetic markers, such as DNA), Protein markers, cytokine markers, chemokine markers, carbohydrate markers, antigen markers, antibody markers, species markers (species/genus markers) and functional markers (KO/OG markers), etc. Among them, the meaning of nucleic acid markers is not limited to existing genes that can be expressed as biologically active proteins, but also includes any nucleic acid fragments, which can be DNA, RNA, modified DNA or RNA, or It is unmodified DNA or RNA, and a collection of them. In this context, nucleic acid markers can sometimes be referred to as characteristic fragments. In the present invention, biomarkers can also be expressed as "intestinal markers", because the found biomarkers related to schizophrenia are all present in the intestine of the subject. Biomarkers are measured and evaluated, often used to check normal biological processes, pathogenic processes, or pharmacological responses to therapeutic interventions, and are useful in many scientific fields.

The biomarkers can be used for high-throughput sequencing to batch analyze stool samples of healthy people and patients with schizophrenia. Based on the high-throughput sequencing data, the healthy population is compared with the schizophrenia patient group to determine the specific nucleic acid sequence related to the schizophrenia patient group. In short, the steps are as follows:

Sample collection and processing: collect stool samples from healthy people and schizophrenia patients, and use the kit for DNA extraction to obtain nucleic acid samples;

Library construction and sequencing: DNA library construction and sequencing are performed by high-throughput sequencing to obtain the nucleic acid sequence of gut microbes contained in stool samples;

Through bioinformatics analysis methods, the nucleic acid sequences of specific gut microbes related to schizophrenia patients are determined. First, the reads are combined with a reference gene set (also called a reference gene set, which can be a newly constructed gene set or any database of known sequences, for example, using known non-redundant genes of the human gut microbial community Set) for comparison. Next, based on the comparison results, the relative abundance of each gene in the nucleic acid samples from the stool samples of healthy people and schizophrenia patients are determined. By comparing the sequencing sequence with the reference gene set, the corresponding relationship between the sequencing sequence and the genes in the reference gene set can be established, so that for a specific gene in a nucleic acid sample, the number of corresponding sequencing sequences can effectively reflect the gene Relative abundance. Therefore, the relative abundance of genes in the nucleic acid sample can be determined through the comparison results and conventional statistical analysis. Finally, after the relative abundance of each gene in the nucleic acid sample is determined, the relative abundance of each gene in the nucleic acid samples from the feces of healthy people and schizophrenic patients is statistically tested. From this, it can be judged in healthy people and mental health. Whether there is a gene with a significant difference in relative abundance in the population of schizophrenia patients, if the gene is significantly different, the gene is regarded as a biomarker of an abnormal state, that is, a nucleic acid marker.

In addition, for a known or newly constructed reference gene set, it usually contains gene species information and functional annotations. Therefore, on the basis of determining the relative abundance of genes, the gene species information and functional annotations can be further classified, In order to determine the relative abundance of the species and the relative abundance of the functions of the microorganisms in the intestinal flora, the species markers and functional markers of the abnormal state can be further determined. In short, the method for determining species markers and functional markers further includes: comparing the sequencing sequences of healthy people and schizophrenia patients with a reference gene set; based on the comparison results, determining healthy people and schizophrenia respectively The relative abundance of species and relative abundance of each gene in nucleic acid samples of patients with diseases; the relative abundance of species and relative abundance of each gene in nucleic acid samples from healthy people and schizophrenia patients Test; and respectively determine species markers and functional markers that have significant differences in the relative abundance of nucleic acid samples between healthy people and schizophrenia patients. According to the embodiment of the present invention, the relative abundance of genes from the same species and the relative abundance of genes with the same functional annotation can be used to perform statistical tests, such as summation, average value, median value, etc., to determine the function Relative abundance and relative abundance of species.

Unless otherwise specified, the technical means used in the examples are conventional means well known to those skilled in the art, and can be carried out with reference to the third edition of the "Molecular Cloning Experiment Guide" or related products. The reagents and products used are also available. Commercially acquired. The various processes and methods that are not described in detail are conventional methods well known in the art. The source of the reagents used, the trade name, and those whose components are necessary to be listed are all indicated when they first appear, and the same reagents used thereafter are not special The description is the same as the content indicated for the first time.

The present invention adopts the analysis method of Metagenome-Wide Association Study (MWAS) to analyze the bacterial composition and functional difference of stool samples by sequencing; the random forest discriminant model is used to discriminate the schizophrenic group and the non-schizophrenic group, Obtain the probability of illness, which can be used for risk assessment, diagnosis, early diagnosis of schizophrenia or to find potential drug targets.

According to the present invention, the term "individual" refers to animals, especially mammals, such as primates, and preferably humans.

According to the present invention, terms such as "a", "an" and "this" not only refer to a singular individual, but include the general category that can be used to describe a particular embodiment.

In the present invention, the sequencing (second-generation sequencing) and MWAS are well known in the art, and those skilled in the art can make adjustments according to specific conditions. According to the embodiment of the present invention, it can be performed according to the method described in the literature (Jun Wang, and Huijue Jia. Metagenome-wide association studies: fine-mining the microbiome. Nature Reviews Microbiology 14.8 (2016): 508-522.).

According to the present invention, the term "mOTU" refers to the operational classification unit (metagenomics Operational Taxonomic Units) (Sunagawa S, Mende D R, Zeller G, et al. Metagenomic species profiling using universal phylogenetic marker genes [J]. Nature, methods, 2013 10(12):1196-1199.), in phylogenetic research or population genetics research, in order to facilitate analysis, artificially set the same mark for a certain taxa (line, species, genus, group, etc.). The sequence is usually divided into different mOTUs according to the similarity threshold, and each mOTU is usually regarded as a microbial species.

In the present invention, the method of using the random forest model and the ROC curve is well known in the art, and those skilled in the art can set and adjust the parameters according to specific conditions. According to the embodiment of the present invention, it can be based on the literature (Drogan D, Dunn WB, Lin W, Buijsse B, Schulze MB, Langenberg C, Brown M, Floegel a., Dietrich S, Rolandsson O, Wedge DC, Goodacre R, Forouhi NG ,Sharp SJ,Spranger J,Wareham NJ,Boeing H:Untargeted Metabolic Profiling Identifies Altered Serum Metabolites of Type 2-Diabetes Melitus in a Prospective, Nested Case Control Study.Clin 487-Study, Milik-Study. SF, de las Heras J, Bacha F, Lee S, Chace DH, DeJesus VR, Vockley J, Arslanian SA: Metabolomic profiling of fatty acid and amino acid metabolism in youth with diabetes for disease and type 2 diabetes: Care 2012, 35:605-611., the full text is incorporated herein by reference).

In the present invention, a training set of biomarkers for schizophrenia subjects and non-schizophrenia subjects is constructed, and based on this, the biomarker content value of the sample to be tested is evaluated.

Those skilled in the art know that when the sample size is further expanded, the normal content value range (absolute value) of each biomarker in the sample can be obtained by using sample detection and calculation methods known in the art. The absolute value of the detected biomarker content can be compared with the normal content value. Optionally, statistical methods can be combined to obtain risk assessment and diagnosis of schizophrenia, and for monitoring patients with schizophrenia The efficiency of the treatment effect and so on.

Without wishing to be bound by any theory, the inventor pointed out that these biomarkers are the intestinal flora present in the human body. Through the method of the present invention, the intestinal flora of the subject is associated with the analysis, and it is obtained that the biomarker of the schizophrenia population shows a certain content range value in the detection of the flora.

The present invention will be further described in detail below in conjunction with specific embodiments, which are to explain rather than limit the present invention.

Example 1

1.1 Sample collection

Refer to the method described in the literature A metagenome-wide association study of gut microbiota in type 2 diabetes (Qin J et al. Nature.2012,490,55-60), the stool samples are collected, frozen, transported and quickly transferred to -80°C for storage, Perform DNA extraction to obtain extracted DNA samples. The stool samples of the schizophrenic and non-schizophrenic subjects of the present invention are from China. A total of 171 cases, including 81 healthy samples and 90 schizophrenia samples.

1.2 Metagenomic sequencing and assembly

The extracted DNA samples were used to construct a sequencing library, and the paired-end metagenomic sequencing (insert 350bp, read length 100bp) was performed on the Illumina HiSeq2000 sequencing platform. The data generated by sequencing is filtered (quality-controlled) to remove adapter-contaminated sequences, low-quality sequences, and host genome-contaminated sequences to obtain high-quality reads. .

1.3 Conservative single-copy gene alignment and abundance calculation

Input the high-quality reads of the above-mentioned "1.2 Metagenomic Sequencing and Screening" into the software mOTU (http://www.bork.embl.de/software/mOTU/download.html) to calculate the relative species Abundance. Refer to the method described in the literature Metagenomic species profiling using universal phylogenetic marker genes (Sunagawa S et al. Nature methods. 2013, 10(12), 1196-9). The abundance calculation steps are as follows: 1) Compare the high-quality sequencing fragments to the reference single-copy gene; 2) Count the number of inserts according to the comparison results; 3) Compare the number of inserts to the length of the single-copy gene Standardization (standardize according to the average gene length, and get the abundance of the corresponding mOTU by rounding down) to get the corresponding abundance.

1.4 Use random forest (ROC/AUC) to screen potential biomarkers for the development of schizophrenia

In order to further screen the intestinal biomarkers of potential diseases, this example constructs a training set of biomarkers for schizophrenia subjects and non-schizophrenia subjects, and uses this as a benchmark to determine the biomarkers of the sample to be tested. The material content value is evaluated. Wherein, in the present invention, the training set and the verification set have meanings known in the art. In the embodiment of the present invention, the training set refers to a data set containing the content of each biomarker in the sample to be tested for a certain number of schizophrenic subjects and non-schizophrenic subjects. The validation set is an independent data set used to test the performance of the training set. Among them, non-schizophrenic subjects are subjects with good mental states, and the subjects may be humans or model animals. In this embodiment, humans are used as subjects for experiments.

Specifically include the following steps:

The present invention randomly selects 80 schizophrenia patients and 71 healthy people from 171 samples (90 schizophrenia patients and 81 healthy people) as the training set (Table 1-1, 1-2), and the rest The sample is used as a validation set (10 schizophrenic patients and 10 healthy people).

1.4.1 Biomarkers obtained by screening using training set data

First, calculate the relative abundance of mOTU in each sample in the training set according to the method described in 1.3. Then input the mOTU of the training set into the random forest (RandomForest 4.6-12 in R 3.2.5, RF) classifier. The classifier was cross-validated 5 times, and the relative abundance of mOTU screened by the RF model was used to calculate the risk of schizophrenia for each individual, draw the ROC curve, and calculate the AUC as the performance evaluation parameter of the discriminant model. Select the combination of markers <30 and the combination with the best discrimination performance is the combination of the invention. Output the importance index of each mOTU in the model. The higher the importance index, the higher the importance of the marker used to distinguish schizophrenia and non-schizophrenia.

Finally, the RF classifier obtained in the present invention contains 11 metabolites (ie 11 biomarkers). The relative abundances of these 11 biomarkers are shown in Tables 1-1 and 1-2, respectively. The detailed information is shown in Table 2 shows. Table 3 shows the combination of 11 biomarkers to predict the probability of illness in the training set, where the probability of illness ≥ 0.5 confirms that the individual is at risk of schizophrenia or has schizophrenia.

Figure 2 shows the error rate distribution of the 5-fold 10-fold cross-validation in the random forest classifier. The model is trained with the training set samples on the relative abundance of mOTU that meets the target obtained through the MWAS process. The thick black solid curve represents the average value of 5 trials (the thin black curve represents 5 trials), and the black vertical line represents the selected The number of mOTUs in the best combination.

Figure 3 shows the ROC curve and AUC of a training set composed of schizophrenia patients and healthy controls based on the random forest model (11 biomarkers), where the specific characterization of the probability of correcting the disease is sensitive to Sex refers to the probability of correcting the disease. The discriminating power of the training set samples is: AUC=82.1%, 95% confidence interval CI=75.3-88.89%. The results show that the metabolite combination obtained from this model can be used as a potential biomarker to distinguish schizophrenia from non-schizophrenia.

1.4.2 Use validation set data to verify the biomarkers obtained by screening

The present invention then uses an independent population to verify the model, and the disease probability is greater than or equal to 0.5 to predict that the individual is at risk of schizophrenia or suffers from schizophrenia. First, calculate the relative abundance of each biomarker in each sample in the training set according to the method described in 1.3. Then use the random forest model to verify the validation set data according to the method of 1.4.1.

Based on the model:

The relative abundance of the 11 biomarkers in the validation set is shown in Table 4-1 and 4-2. Table 5 shows the probability of disease based on the 11 biomarker prediction validation set.

Figure 4 shows the ROC curve and AUC of an independent validation set composed of schizophrenia patients and healthy controls based on the random forest model (11 biomarkers). The model's discrimination AUC=77% (95%CI=55.71%- 98.29%).

In version R 3.2.5, use "randomForest 4.6-12 package" for random forest model classification and regression. The input includes training set data (that is, the relative abundance of the selected mOTU markers in the training sample, see Tables 1-1 and 1-2), sample disease status (the sample disease status of the training sample is a vector, and '1' represents the spirit Schizophrenia, '0' stands for healthy control), and a validation set (the relative abundance of the selected mOTU markers in the validation set, see Table 4-1, 4-2). Then, the inventor uses the random forest function of the random forest package in the R software to establish classification and prediction functions to predict the verification set data, and the output is the prediction result (probability of disease); the threshold is 0.5, if the probability of disease is ≥0.5, then Think that you are at risk of schizophrenia or suffer from schizophrenia.

Table 1-1 Relative abundance data of intestinal markers (mOTU) in the training set of random forest model

*SZ: patients with schizophrenia; H: healthy controls

Table 1-2 Relative abundance data of intestinal markers (mOTU) in the training set of random forest model

Table 2 Detailed information of 11 biomarkers

#Validation set AUC indicates the degree of discrimination of the validation set data under the model obtained from the training set data.

Table 3 Prevalence probability based on 11 kinds of marker combination training set

Sample number	Probability of illness	To	Sample number	Probability of illness
H_1	0.918918919	To	SZ_1	0.870056497
H_10	0.042328042	To	SZ_10	0.710382514
H_11	0.259668508	To	SZ_11	0.615789474
H_12	0.049180328	To	SZ_12	0.56043956
H_13	0.166666667	To	SZ_13	0.783783784
H_14	0.099447514	To	SZ_14	0.944785276
H_15	0.12972973	To	SZ_15	0.957446809
H_16	0.068571429	To	SZ_16	0.888268156
H_17	0.31	To	SZ_17	0.428571429
H_18	0.069148936	To	SZ_18	0.823529412
H_19	0.374301676	To	SZ_19	0.614130435
H_2	0.891304348	To	SZ_2	0.625766871
H_20	0.224043716	To	SZ_20	0.508287293
H_21	0.167539267	To	SZ_21	0.632124352
H_22	0.329896907	To	SZ_22	0.735
H_23	0.387755102	To	SZ_23	0.796407186
H_24	0.529100529	To	SZ_24	0.898876404
H_25	0.247311828	To	SZ_25	0.494565217
H_26	0.156424581	To	SZ_26	0.824120603
H_27	0.076923077	To	SZ_27	0.494736842
H_28	0.167464115	To	SZ_28	0.314285714
H_29	0.360946746	To	SZ_29	0.465838509

H_3	0.728643216	To	SZ_3	0.903061224
H_30	0.278688525	To	SZ_30	0.920212766
H_31	0.253807107	To	SZ_31	0.827225131
H_32	0.375661376	To	SZ_32	0.723529412
H_33	0.395	To	SZ_33	0.701149425
H_34	0.344632768	To	SZ_34	0.613259669
H_35	0.15819209	To	SZ_35	0.73255814
H_36	0.248618785	To	SZ_36	0.177142857
H_37	0.248587571	To	SZ_37	0.324175824
H_38	0.391304348	To	SZ_38	0.194594595
H_39	0.355670103	To	SZ_39	0.733009709
H_4	0.444444444	To	SZ_4	0.63253012
H_40	0.086294416	To	SZ_40	0.916167665
H_41	0.497326203	To	SZ_41	0.277486911
H_42	0.747252747	To	SZ_42	0.768421053
H_43	0.204878049	To	SZ_43	0.983957219
H_44	0.317365269	To	SZ_44	0.220338983
H_45	0.059701493	To	SZ_45	0.715976331
H_46	0.225433526	To	SZ_46	0.876344086
H_47	0.551020408	To	SZ_47	0.481081081
H_48	0.142857143	To	SZ_48	0.808917197
H_49	0.207865169	To	SZ_49	0.25443787
H_5	0.694300518	To	SZ_5	0.611940299
H_50	0.202247191	To	SZ_50	0.444444444
H_51	0.103092784	To	SZ_51	0.794444444
H_52	0.491712707	To	SZ_52	0.513513514
H_53	0.481283422	To	SZ_53	0.80239521
H_54	0.666666667	To	SZ_54	0.950549451
H_55	0.384615385	To	SZ_55	0.91
H_56	0.563535912	To	SZ_56	0.905263158
H_57	0.762430939	To	SZ_57	0.979274611
H_58	0.045454545	To	SZ_58	0.773255814
H_59	0.717647059	To	SZ_59	0.705202312
H_6	0.901639344	To	SZ_6	0.594736842
H_60	0.251282051	To	SZ_60	0.710227273
H_61	0.13368984	To	SZ_61	0.96969697
H_62	0.231213873	To	SZ_62	0.741935484
H_63	0.174603175	To	SZ_63	0.056179775
H_64	0.478787879	To	SZ_64	0.055865922
H_65	0.371428571	To	SZ_65	0.345360825
H_66	0.453038674	To	SZ_66	0.169590643
H_67	0.47761194	To	SZ_67	0.941489362
H_68	0.580310881	To	SZ_68	0.557471264
H_69	0.596153846	To	SZ_69	0.929906542
H_7	0.015075377	To	SZ_7	0.462427746

H_70	0.044198895	To	SZ_70	0.802955665
H_71	0.041025641	To	SZ_71	0.693989071
H_8	0.551020408	To	SZ_72	0.591954023
H_9	0.244565217	To	SZ_73	0.666666667
To	To	To	SZ_74	0.972677596
To	To	To	SZ_75	0.766497462
To	To	To	SZ_76	0.169230769
To	To	To	SZ_77	0.603550296
To	To	To	SZ_78	0.196969697
To	To	To	SZ_79	0.809782609
To	To	To	SZ_8	0.815217391
To	To	To	SZ_80	0.762711864
To	To	To	SZ_9	0.720670391

Table 4-1 Random Forest Model Validation Set Intestinal Marker (mOTU) Relative Abundance Data

Table 4-2 Relative abundance data of intestinal markers (mOTU) in the validation set of random forest model

Table 5 Prevalence probability based on the validation set of 11 marker combinations

Sample number	Probability of illness	To	Sample number	Probability of illness
H_72	0.746	To	SZ_81	0.562
H_73	0.454	To	SZ_82	0.378
H_74	0.306	To	SZ_83	0.634
H_75	0.272	To	SZ_84	0.96
H_76	0.322	To	SZ_85	0.6
H_77	0.114	To	SZ_86	0.3
H_78	0.124	To	SZ_87	0.36
H_79	0.104	To	SZ_88	0.88
H_80	0.506	To	SZ_89	0.898
H_81	0.394	To	SZ_90	0.292

The above results show that the biomarker disclosed in the present invention has high accuracy and specificity, and has a good prospect of being developed as a diagnostic method, so as to find potential drugs for risk assessment, diagnosis, and early diagnosis of schizophrenia The target provides the basis.

Therefore, the present invention proposes the following applications:

The combination of biomarkers for schizophrenia based on intestinal flora is used as a detection target or a detection target in the preparation of a detection kit.

The application of the combination of biomarkers for schizophrenia based on intestinal flora as a target in screening drugs for the treatment and/or prevention of schizophrenia.

The change in the relative abundance of the biomarker combination provides a basis for determining whether the candidate drug is effective.

Although the embodiments of the present invention have been shown and described above, it can be understood that the above-mentioned embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can comment on the foregoing The embodiment undergoes changes, modifications, substitutions and modifications.

Claims (8)

1. A biomarker combination for schizophrenia based on intestinal flora, characterized in that the biomarker combination is used to provide relative abundance information, and it contains one or more selected from the following:

   Biomarker 1: Lachnospiraceae bacterium 3_1_57FAA_CT1;

   Biomarker 2: Cronobacter sakazakii;

   Biomarker 3: Lactobacillus acidophilus;

   Biomarker 4: Veillonella parvula;

   Biomarker 5: Lactococcus lactis;

   Biomarker 6: Alkaliphilus oremlandii;

   Biomarker 7: Pseudoflavonifractor capillosus;

   Biomarker 8: Streptococcus gallolyticus;

   Biomarker 9: Dialister invisus;

   Biomarker 10: Lactobacillus johnsonii;

   Biomarker 11: Methanobrevibacter smithii.
2. The schizophrenia biomarker combination based on intestinal flora of claim 1, wherein the relative abundance information provided by the biomarker combination is used for comparison with a reference value.
3. The biomarker combination for schizophrenia based on the intestinal flora of claim 1, wherein the relative abundance information of the biomarkers 1-11 is based on genes whose abundance can be calculated Sequence to provide.
4. Application of the schizophrenia biomarker combination based on intestinal flora of claim 1 as a detection target or a detection target in the preparation of a detection kit.
5. The application of the schizophrenia biomarker combination based on the intestinal flora of claim 1 as a target in screening drugs for the treatment and/or prevention of schizophrenia.
6. The application according to claim 5, wherein the change in the relative abundance of the biomarker combination provides a basis for determining whether the candidate drug is effective.
7. The method for screening a combination of biomarkers for schizophrenia based on intestinal flora of claim 1, wherein the steps are as follows:

   1) Sample collection: After collecting stool samples, they are frozen and transported and quickly transferred to -80°C for storage. DNA extraction is performed to obtain extracted DNA samples. The sample subjects include schizophrenia patients and healthy people;

   2) Metagenomic sequencing and assembly

   3) Conservative single-copy gene alignment and abundance calculation

   3) Input high-quality sequencing fragments into the software mOTU to calculate the relative abundance of species:

   3.1) Align the high-quality sequencing fragments to the reference single-copy gene;

   3.2) Count the number of inserts according to the comparison results;

   3.3) Normalize the number of inserts to the length of a single copy gene to obtain the corresponding abundance.

   4) Randomly select schizophrenic patients and healthy people from the sample set as the training set, and the remaining samples as the validation set. Calculate the relative abundance of mOTU in each sample in the training set, and then input the mOTU of the training set into the random forest classifier. Perform 5 10-fold cross-validation on the classifier, use the relative abundance of mOTU screened by the RF model to calculate the risk of schizophrenia for each individual, draw the ROC curve, and calculate the AUC as the performance evaluation parameter of the discriminant model, and select the marker The number of combination of substances is less than 30, and the combination with the best discrimination performance, the importance index of each mOTU is output in the model. The higher the importance index, it represents the importance of the marker used to distinguish schizophrenia and non-schizophrenia Higher.
8. The screening method according to claim 7, characterized in that the sample is concentrated, the sample subjects include 90 schizophrenia patients and 81 healthy people, and the verification concentration, the sample subjects include 10 schizophrenia patients and 10 A healthy person.

Images