Method and system for screening biomarkers of intestinal flora and metabolic products of overweight or obese patients for weight loss effect of dietary intervention, and application thereof

Abstract

The present application relates to a method and system for screening biomarkers of gut flora and metabolic products of the effect of dietary intervention on weight loss of overweight or obese patients and application, belonging to the technical field of microbiome and related metabolome and human health. Based on the random controlled intervention test design, the biomarker combination is screened out as 6 kinds of bacteria and 3 kinds of fecal metabolites which are significantly related to weight loss effect, wherein the relative abundance of 6 kinds of bacteria and the concentration of 2 metabolites are negatively correlated with weight loss effect, and the concentration of 1 metabolite is positively correlated with weight loss effect. In addition, an efficient weight loss effect prediction model is established by using the screened biomarker combination, which provides a new strategy for realizing effective weight loss and helps to guide personalized weight management.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of microbiome and related metabolome and human health, more specifically, relates to a screening method and system for gut flora and metabolic product biomarkers of diet intervention weight loss effect in overweight or obese patients, and application thereof. BACKGROUND

[0002] Obesity is a chronic disease caused by multiple factors and is a global public health challenge. The prevalence of overweight and obesity in Chinese adults is about 46%, and the total number of overweight and obese people ranks first in the world (Wang et al., 2019; Pan et al., 2021). Overweight and obesity are risk factors for many diseases, including hypertension, diabetes, cardiovascular disease, and certain cancers (such as breast cancer, colorectal cancer, etc.) (Wang et al., 2021). Obesity not only increases the risk of adverse health outcomes for individuals, but also brings a huge medical burden to society. Studies have shown that a 5%-10% weight loss can help improve blood glucose control, reduce blood pressure, and increase high-density lipoprotein cholesterol concentration, thereby reducing the risk of related diseases (Elmaleh-Sachs et al., 2023).

[0003] Lifestyle intervention is the preferred treatment for overweight and obesity, and other treatments include weight loss drugs and weight loss surgery. The Chinese Clinical Management and Treatment of Obesity states that dietary intervention is a core component of lifestyle intervention in weight management (Zeng et al., 2021). Studies have shown that compared with regular diets, dietary patterns such as calorie-restricted diets, low-carbohydrate diets, time-restricted diets, and Mediterranean diets can effectively reduce the body weight of overweight or obese individuals, but the weight loss effect is influenced by various factors such as environment and genetic background (Thom & Lean, 2017). Therefore, finding predictors that can be used to predict the impact of weight loss and establishing a prediction model for weight loss effect is crucial for preventing obesity and improving weight loss effect and reducing the risk of obesity-related complications.

[0004] Metagenomics obtains all genomic information contained in microbial communities through high-throughput sequencing technology, enabling a more in-depth study of the structure and function of microorganisms in samples and providing new clues for disease mechanisms. However, metagenomics technology cannot directly measure the functional activity and key active molecules of microorganisms. Metabolomics is a qualitative and quantitative detection of small molecule metabolites in samples, which helps to elucidate the pathogenesis of diseases. Combined analysis of metagenomics and metabolomics can overcome the limitations of single-omics research to some extent, helping to comprehensively understand the relationship between microorganisms, metabolites, and diseases, and identify potential biomarkers and therapeutic targets.

[0005] The etiology of obesity is complex and can include adverse lifestyle (high-calorie diet, low physical activity level), genetic factors and social environmental factors, among which gut microbiota and their metabolites can play an important role (Wu et al., 2021). Experimental and observational studies suggest (Le Chatelier et al., 2013; Ley et al., 2006; Turnbaugh et al., 2009) that the composition of the gut microbiota is associated with obesity, and the diversity of the gut microbiota of obese individuals is reduced. In addition, gut microbes can also regulate the metabolism and body weight status of the host through different mechanisms, such as directly affecting the absorption of calories from food by the body, increasing the permeability of the gut, and indirectly affecting the metabolism of the host through metabolites produced by the microbiota or regulating dietary, host-derived metabolites, thereby affecting the body weight of the host et al., 2004; Cardinelli et al., 2015; Van Hul & Cani, 2023).

[0006] In recent years, many studies have begun to look for biomarkers for predicting the effect of dietary weight loss. In European Patent Application EP3464621, a small molecule RNA (miRNA) is disclosed for predicting the degree of weight loss obtained by applying one or more dietary interventions to an individual. However, miRNA is greatly affected by ethnic and environmental factors, and its expression may differ in different populations. In the past decade, studies have suggested that the individual differences in weight loss may depend partly on the composition of the gut microbiota (Van Hul & Cani, 2023). Studies have suggested that the gut microbiota is an important predictor of the individual's weight loss trajectory (Jie et al., 2021; Roager & Christensen, 2022). In Chinese Patent Publication No. CN112980945, a neural network model is used to predict the weight loss effect of a low-carbon dietary intervention by the relative abundance of Bacteroides bacteria before weight loss, but this patent may be limited by the small sample size, and the prediction model is not divided into training and validation sets. And only the baseline relative abundance of Bacteroides Bacteroides was found to be positively correlated with weight loss, and the AUC value of the prediction model established by this bacterium was 73.2%, and the prediction ability needs to be improved. In addition, due to the limited amount of fecal samples, fecal metabolomics detection could not be performed.

[0007] Therefore, there is still a need to find biomarkers suitable for the Chinese population and to establish more accurate prediction models for predicting the weight loss effect of individuals. SUMMARY

[0008] The present application aims to screen biomarkers related to intestinal flora and fecal metabolites of overweight or obese patients after dietary intervention. Finally, 9 biomarkers significantly related to weight loss effect are screened, which can be combined with traditional clinical weight loss effect prediction model to optimize existing weight loss strategies. The present application provides scientific basis for evaluating the weight loss effect of dietary intervention and developing personalized weight loss programs, and promotes the application of precision medicine in obesity management. In addition, the present application also provides a new idea for the research and development of weight loss related drugs.

[0009] According to the first aspect of the present application, a method for screening intestinal flora and metabolite biomarkers of overweight or obese patients after dietary intervention is provided, comprising the following steps:

[0010] (1) Collecting baseline fecal samples of patients, performing metagenomic sequencing, intestinal microbiome analysis and fecal metabolome detection;

[0011] (2) Preprocessing the data of intestinal flora abundance and fecal metabolite concentration, using one tenth of the minimum detection value of the analyte to fill in the missing values below the detection limit, and log transforming the concentration of each marker;

[0012] (3) Using a covariate-corrected conditional logistic model to screen biomarkers related to weight loss effect, the covariants of the model include age, gender, whether to adopt healthy low-carbohydrate dietary intervention, whether to adopt time-limited dietary intervention and baseline body weight;

[0013] (4) using a LASSO-logistic regression model to reduce the dimensionality of the related biomarkers, and performing ten-fold cross-validation at each step of screening, to finally obtain a biomarker combination significantly related to the weight loss effect after the caloric restriction diet intervention; the biomarker combination includes 9 biomarkers, which are 6 bacterial species and 3 metabolites; the 6 bacterial species are: Actinomyces sp. ICM47, Blautia sp. AF19-10LB, Lachnospira sp. NSJ-4, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis, and Bacteroides nordii; the 3 metabolites are: Vanillic acid, Hyocholic acid, and Isocaproic acid; the abundance of the 6 bacterial species is lower in the weight loss group than in the control group without weight loss; the concentration of Vanillic acid and Hyocholic acid among the 3 metabolites is lower in the weight loss group than in the control group without weight loss, and the concentration of Isocaproic acid among the 3 metabolites is higher in the weight loss group than in the control group without weight loss.

[0014] Preferably, in step (1), shotgun method is used for metagenomic sequencing and intestinal microbiome analysis.

[0015] Preferably, in step (1), the detection of fecal metabolites uses ultra-high performance liquid chromatography-mass spectrometry.

[0016] According to another aspect of the present application, a biomarker screening system for intestinal flora and metabolic products of overweight or obese patients after caloric restriction diet intervention is provided, comprising:

[0017] Data acquisition module: collect baseline fecal samples of patients, perform metagenomic sequencing, intestinal microbiome analysis, and fecal metabolome detection;

[0018] Data preprocessing module: preprocess the intestinal flora abundance and fecal metabolite concentration data, use one-tenth of the minimum detection value of the analyte to fill in the missing values below the detection limit, and perform log transformation on the concentration of each marker;

[0019] The candidate biomarker screening module: the biomarkers related to the weight loss effect are screened using a covariate-corrected conditional logistic model, and the covariates of the model include age, gender, whether a healthy low-carbohydrate diet intervention is adopted, whether a time-limited diet intervention is adopted, and baseline body weight;

[0020] The machine learning dimension reduction biomarker screening module: the above candidate biomarkers are subjected to dimension reduction processing using a LASSO-logistic regression model that corrects for confounding factors, each step of screening is subjected to ten-fold cross-validation, and finally a biomarker combination significantly related to the weight loss effect after the energy-restricted diet intervention is obtained: the biomarker combination includes 9 biomarkers, the 9 biomarkers are 6 bacterial species and 3 metabolites; the 6 bacterial species are: Actinomyces sp. ICM47, Blautia sp. AF19-10LB, Lachnospira sp. NSJ-4, Mediterranean bacterium glycyrrhizinilyticus, Dysosmobacter welbionis and Bacteroides nordii; the 3 metabolites are: Vanillic acid, Hyocholic acid and Isocaproic acid; the abundance of the 6 bacterial species in the weight loss group is lower than that in the control group without weight loss; the concentration of Vanillic acid and Hyocholic acid in the 3 metabolites in the weight loss group is lower than that in the control group without weight loss, and the concentration of Isocaproic acid in the 3 metabolites in the weight loss group is higher than that in the control group without weight loss.

[0021] According to another aspect of the present application, there is provided an application of intestinal flora and metabolite biomarkers in constructing a weight loss effect prediction model of a diet intervention for an overweight or obese patient, at least one of the screened intestinal flora and metabolite biomarkers is added to the weight loss effect prediction model after traditional clinical intervention to reconstruct a new weight loss effect prediction model.

[0022] The biomarkers are 6 bacterial species and 3 metabolites; the 6 bacterial species are: Actinomyces sp. ICM47, Blautia sp. AF19-10LB, Lachnospira sp. NSJ-4, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis and Bacteroides nordii; the 3 metabolites are: Vanillic acid, Hyocholic acid and Isocaproic acid;

[0023] The 6 bacterial species are all reduced in abundance in the weight loss group compared with the non-weight loss control group; Vanillic acid and Hyocholic acid of the 3 metabolites are reduced in concentration in the weight loss group compared with the non-weight loss control group, and Isocaproic acid of the 3 metabolites is increased in concentration in the weight loss group compared with the non-weight loss control group.

[0024] Preferably, the traditional prediction model is based on factors such as age, gender, intervention measures and baseline initial body weight; the new weight loss effect prediction model is based on the traditional prediction model, and then the newly identified biomarker combination is further included in the model as a prediction factor; the prediction performance of the model is evaluated and compared by the area under the receiver operating characteristic curve and the DeLong test.

[0025] Preferably, the dietary characteristic index includes baseline total energy intake and baseline total dietary fiber intake.

[0026] In summary, the above technical solutions of the present application have the following technical advantages compared with the prior art:

[0027] (1) Based on high-quality clinical research: the present application relies on a randomized controlled intervention study on the improvement of body weight and blood sugar of overweight or obese people by health low-carbohydrate diet and time-limited diet, which minimizes potential bias by randomly assigning subjects, ensures the reliability and objectivity of the results, and provides a solid scientific basis for the present application.

[0028] (2) The biomarker screening process is rigorous and comprehensive: First, the innovative use of metagenomics and metabolomics combined analysis can overcome the limitations of single-omics research to some extent, helping to comprehensively understand the relationship between microorganisms-metabolites-diseases, and identify potential biomarkers and therapeutic targets. Second, the patients are grouped according to whether they achieve clinically meaningful weight loss, and their intestinal microorganisms and fecal metabolites are analyzed. Then, compared with the method of using conditional logistic model to screen biomarkers alone, the combined use of conditional logistic model with covariate correction, machine learning dimension reduction technology and cross-validation can consider the collinearity between biomarkers, retain important features and improve the stability and reliability of the results. Finally, through the above screening methods, 9 biomarkers significantly related to weight loss are screened. These markers provide a scientific basis for future personalized weight loss strategy development and promote the application of precision medicine in obesity management.

[0029] (3) Significantly improve the performance of the prediction model and have wide clinical application potential: Based on the biomarker combination screened above, the prediction model is constructed, and through the evaluation and comparison of the prediction performance of the model, it is found that the biomarker combination significantly improves the prediction accuracy of the traditional clinical prediction model. In addition, the prediction model has good interpretability, verifiability and clinical practicability, can provide personalized intervention recommendations for obese patients, improve the success rate of weight loss, and has wide clinical application prospects. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 The clinical experiment design and statistical methods involved in the source of the invention are used to screen biomarkers significantly related to weight loss after 12 weeks of dietary intervention, and to establish a prediction model for weight loss after 12 weeks of dietary intervention.

[0031] Figure 2 Clinical experiment design and weight loss effect of 12-week dietary intervention: A, research design of 12-week dietary intervention. The dietary intervention provides subjects with an energy-reduced diet of about 25%, male: 1600kcal / day, female 1300kcal / day; B, the broken line chart shows the body weight changes of the 4 groups during the 12-week intervention. Compared with the baseline, the body weight of the subjects after 12 weeks of dietary intervention is significantly reduced. The invention groups according to whether the weight loss effect of 12-week dietary intervention reaches clinically meaningful weight loss (5% of body weight, calculation formula weight loss percentage = weight loss kg / initial body weight*100%), the response group is the clinically meaningful weight loss, i.e. weight loss percentage ≥5%, and the non-response group is the clinically meaningful weight loss (i.e. <5%).

[0032] Figure 3Differential analysis between response and non-response groups was performed using conditional logistic model for baseline gut microbiota species level. The volcano plot was used to display the results, with the horizontal axis representing the fold change and the vertical axis representing the negative logarithm of P value. When the fold change was 0, the right side of the vertical dotted line represented the gut microbiota species level that was significantly up-regulated in the response group, and the left side represented the gut microbiota species level that was significantly down-regulated. The horizontal dotted line represented the significance P value, and the gut microbiota species level above the horizontal dotted line represented a significant change, and the gut microbiota species level below the horizontal dotted line represented no significant difference.

[0033] Figure 4 Differential analysis between response and non-response groups was performed using conditional logistic model for baseline gut microbiota species level. The volcano plot was used to display the results, with the horizontal axis representing the fold change and the vertical axis representing the negative logarithm of P value. When the fold change was 0, the right side of the vertical dotted line represented the gut microbiota species level that was significantly up-regulated in the response group, and the left side represented the gut microbiota species level that was significantly down-regulated. The horizontal dotted line represented the significance P value, and the gut microbiota species level above the horizontal dotted line represented a significant change, and the gut microbiota species level below the horizontal dotted line represented no significant difference.

[0034] Figure 5 The penalty term parameter lambda in LASSO regression was selected according to 10-fold cross-validation. Figure 5 A in FIG. 1 shows the change trajectory of the coefficients of each independent variable in the model corresponding to the change of lambda, Figure 5 B in FIG. 1 shows the lambda value within 1 standard error of the minimum value of the mean square error according to cross-validation.

[0035] Figure 6 The receiver operating characteristic curve (ROC) was used to compare the prediction ability of the newly constructed prediction model with the traditional model including only age, gender, intervention method and baseline body weight, and the model adding baseline dietary characteristics index for the weight loss effect of 12-week dietary intervention. The prediction performance of the traditional regression model was 0.672 (95% CI 0.557, 0.788); when the dietary quality index (baseline total energy intake and baseline total dietary fiber intake) was added, the prediction performance of the prediction model reached 0.722 (95% CI 0.614, 0.830); further adding 9 newly identified biomarkers as prediction factors, the prediction performance was significantly improved to 0.939 (95% CI 0.891, 0.988). DETAILED DESCRIPTION

[0036] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application. In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0037] Figure 1 The clinical trial design and statistical methods involved in the present application are screened through the process to screen biomarkers significantly related to the weight loss effect of 12-week dietary intervention, and a prediction model for the weight loss effect of 12-week dietary intervention is established accordingly.

[0038] Example 1

[0039] I. Experimental methods

[0040] 1. Research subjects

[0041] The subjects of the present study come from the control group, time-limited diet group, healthy low-carbohydrate diet group and healthy low-carbohydrate diet combined with time-limited diet intervention group in the prospective clinical intervention study “Randomized Controlled Intervention Study on Health Low-carbohydrate Diet and Time-limited Diet for Weight Loss and Blood Glucose Improvement in Overweight or Obese Population” (hereinafter referred to as RCT) conducted by the School of Public Health of Huazhong University of Science and Technology. The present study complies with the ethical guidelines of the Helsinki Declaration, is approved by the Ethics Committee of Tongji Medical College of Huazhong University of Science and Technology, and is registered in advance in the China Clinical Trial Registry (clinical trial approval number: ChiCTR2200056363). The subjects voluntarily participate in the present study and sign the written informed consent form. The trial period is from March 2022 to April 2023. The design idea of the clinical trial of the present study is shown in Figure 2 .

[0042] 2. Diagnostic criteria

[0043] The diagnosis of overweight / obesity is based on the diagnostic criteria of “Guidelines for Prevention and Treatment of Overweight and Obesity in Chinese Adults” (2003). Overweight: 24.0 kg / m 2 ≤ BMI < 28.0 kg / m 2 , obesity: BMI ≥ 28.0 kg / m 2 .

[0044] 2.1 Inclusion criteria

[0045] 1) Age 20-60 years old;

[0046] 2) Overweight or obese: body mass index ≥ 24 kg / m 2 ;

[0047] 3) Engage in light physical activity.

[0048] 2.2 Exclusion criteria

[0049] 1) More than once per week night shift work (whole night shift: 8:00 pm-8:00 am the next day)

[0050] 2) Regular fasting (15 hours of fasting per day or 12 times of 24 hours fasting in the past year)

[0051] 3) Adopting a restricted carbohydrate diet in the past 3 months, or taking a special or prescribed diet for other reasons

[0052] 4) Significant change in body weight (≥3 kg) in the past 3 months

[0053] 5) Currently participating in a weight loss or weight management program or taking any drugs with known effects on appetite

[0054] 6) Suffering from serious digestive system diseases, history of gastrointestinal surgery, impaired nutrient absorption or eating disorders

[0055] 7) Suffering from diabetes or serious cardiovascular, renal, hepatic, pulmonary or nervous system diseases, etc. acute or chronic diseases, infectious diseases, cancer, etc.

[0056] 8) Having received surgical treatment in the past 3 months or planning to receive surgery within 3 months

[0057] 9) Taking antibiotics or any drugs that may affect the study endpoints such as blood glucose, blood lipids and body composition metabolism (such as anti-diabetic drugs, steroids, beta blockers, adrenergic stimulants, etc.) in the past 3 months

[0058] 10) Taking dietary supplements (containing n-3 unsaturated fatty acids, prebiotics, probiotics, etc.) in the past 3 months

[0059] 11) Suffering from serious mental disorders or unable to understand or follow instructions

[0060] 12) Daily high-intensity physical activity

[0061] 13) Heavy drinking habits (more than 100g of Baijiu per day, more than 5 bottles of beer) or heavy smoking habits (more than 2 packs per day)

[0062] 14) Pregnancy or pregnancy or planning to become pregnant or lactating women within three months

[0063] 15) Participating in other clinical trials

[0064] 3 Clinical trial research methods

[0065] The research method follows the RCT research protocol.

[0066] 3.1 Study grouping

[0067] First, the volunteers were randomly divided into 4 groups in a ratio of 1:1:1:1 by computer-generated random numbers, and the randomization grouping was performed by a person not involved in the study.

[0068] (1) Control group: The volunteers were provided with general meals made according to the Dietary Guidelines for Chinese Residents without restriction on total daily energy intake.

[0069] (2) Healthy low-carbohydrate diet group: Healthy low-carbohydrate diet was provided without restriction on eating time.

[0070] (3) Time-restricted diet group: General meals were provided and time-restricted diet was adopted.

[0071] (4) Combined intervention group: Healthy low-carbohydrate diet was provided and time-restricted diet was adopted.

[0072] 3.2 Dietary intervention program

[0073] The intervention mainly provided meals (working days) + supplemented with meal guidance (weekend rest days)

[0074] a) General meals: meals made according to the Dietary Guidelines for Chinese Residents without restriction on total energy intake;

[0075] b) Healthy low-carbohydrate diet: whole grains as the main source of carbohydrates, accounting for 35% of total daily energy, with energy supply ratios of fat and protein being 45-50% and 15-20%, respectively, with unsaturated fatty acids as the main source of fat and plant proteins as the main source of protein with a certain proportion of animal proteins;

[0076] c) Time-restricted diet: 14 hours of fasting and 10 hours of eating. Three meals were provided at regular intervals, and daily energy intake was concentrated in the eating period. No high-calorie food could be eaten during the fasting period, and volunteers were recommended to drink water or consume 1-2 servings of each of the following: less than 4kcal of chewing gum, mint candy, tea leaves, black coffee, 0-calorie drinks, etc.

[0077] d) Non-time-restricted diet: i.e., without restriction on eating time, in addition to the provision of three meals, volunteers could consume snacks such as fruits, but needed to be recorded.

[0078] 3.3 Study plan

[0079] According to the inclusion and exclusion criteria of the research subjects, 96 overweight or obese patients meeting the inclusion and exclusion criteria were selected, signed the informed consent form, and then the baseline measurement including 24-hour dietary record, questionnaire, biological measurement and physical examination were conducted, and blood and fecal samples were collected. After the baseline measurement and randomization grouping (4 groups, 24 people in each group), the subjects will receive the meal overview manual. During the formal intervention period, all volunteers will be provided with three meals a day on weekdays, and dietary guidance on weekends. The intervention will last for 12 weeks, and the endpoint measurement (consistent with the baseline measurement) will be conducted after the intervention ends.

[0080] 3.4 Collection of clinical data

[0081] (1) Body weight was detected by standard electronic body weight scale, and was detected at baseline (before intervention) and endpoint (after intervention). The body weight detection data at baseline and endpoint were used in the present application.

[0082] (2) The fecal sample collection time points were baseline (before intervention) and endpoint (after intervention), which were collected by the subjects themselves.

[0083] (3) Shotgun metagenomic sequencing and gut microbiome analysis: Microbial DNA was extracted from 758 fecal samples using Mag Pure Fast Stool DNA KF Kit B reagent, and the extracted DNA was subjected to shotgun metagenomic 100 bp paired-end (PE) sequencing using BGI platform. The quality control process of whole genome shotgun sequencing data used KneadData (version 0.7.2), Trimmomatic (version 0.33) and Bowtie2 (version 2.3.4.3). By aligning the reads to the human reference genome (GRCh37) database and the SILVA 128 database, human reads and rDNA reads were filtered. After quality control, an average of 36.9 million (minimum 14.8 million, maximum 54.1 million) high-quality reads were obtained for each sample. Taxonomic features were determined by MetaPhlan (version 3.0.3), and microbial functional features, including MetaCyc pathways and enzyme commission gene families, were determined by HUMAnN (version 3.0.1). Microbial species with a relative abundance lower than 0.01% and not appearing in more than 90% of the samples were excluded from subsequent analysis. Finally, a total of 205 microbial species were included.

[0084] (4) Fecal metabolomics detection: Ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) was used for detection. A total of 222 fecal metabolites associated with gut microbiota were identified. After excluding metabolites with a missing proportion exceeding 30% or a coefficient of variation >30% in all samples, 217 fecal microbial host co-metabolites were quantified and included in the data analysis. In addition, the missing values ​​of metabolites included in the analysis were estimated using 1 / 10 of the minimum detectable value. The 217 metabolites included 40 amino acids, 42 fatty acids, 19 carbohydrates, 34 bile acids, 6 indole compounds, 10 short-chain fatty acids, and other categories of metabolites, including benzoic acid, organic acids, and phenols.

[0085] (5) This invention uses only baseline gut microbiome species abundance and baseline fecal metabolite detection data.

[0086] II. Statistical Analysis and Results

[0087] 1. After the 12-week intervention, a total of 88 subjects completed the 12-week dietary intervention. After excluding subjects with poor adherence and those lacking gut microbiota measurement data, the baseline gut microbiota metagenomic data, fecal metabolomics data, baseline dietary survey data, and baseline and endpoint clinical and physical examination data of 84 subjects were included in this invention. The 12-week dietary intervention produced good weight loss in all subjects. Figure 2 (B) Previous studies have considered a 5% weight loss percentage to be clinically significant, calculated as weight loss percentage = weight loss (kg) / initial body weight * 100%. Therefore, to facilitate the dissemination of this invention, this invention uses whether the weight loss effect of a 12-week dietary intervention reaches a clinically significant level (5% of body weight) as the screening biomarker and predictive outcome. The participants are grouped according to whether the 12-week dietary intervention achieves clinically significant weight loss: the response group achieves clinically significant weight loss (≥5%), while the non-response group does not achieve clinically significant weight loss (<5%).

[0088] 2. Biomarkers significantly associated with the weight loss effect of the 12-week dietary intervention were screened using commonly used epidemiological statistical methods and machine learning algorithms. All analyses below were performed in SAS 9.4 (SAS Institute, Cary, NC) and R software 4.3.1. The detailed analysis workflow is as follows:

[0089] 2.1 Normalization of potential biomarker data

[0090] For the 205 gut microbiota species level abundance data and 217 fecal gut microbiota related metabolite concentration data, missing values below the limit of detection will be imputed with one-tenth of the minimum detectable value for the analyte. Since the above markers are usually skewed, the concentrations will be log-transformed before analysis. The above markers will be used as potential biomarkers for the following screening module

[0091] 2.2 Screening of candidate biomarkers

[0092] Multiple factor regression analysis will be used to screen potential biomarkers, with a statistical significance threshold of 0.05. Specifically, a conditional logistic model with covariate adjustment will be used to perform regression screening of biomarkers related to weight loss, with covariates including age, gender, whether to use a healthy low-carbohydrate diet intervention, whether to use a time-limited diet intervention, and baseline body weight. The results of the analysis are as follows Figure 3 and Figure 4As shown, 6 species and 10 fecal metabolites were found to be significantly different between the groups with clinically meaningful weight loss and those without. Compared with the control group without clinically meaningful weight loss, the baseline abundance of 6 species was significantly down-regulated in the group with clinically meaningful weight loss, mainly including Actinomyces sp. ICM47, Blautia sp. AF19-10LB, Lachnospira sp. NSJ-43, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis, and Bacteroides nordii. Among the 10 fecal metabolites, the concentration of 8 metabolites was significantly up-regulated in the group with clinically meaningful weight loss, mainly including Beta_D_Fucose, N-Acetylneuraminic acid, N-Acetyl-L-aspartic acid, L-Malic acid, Dodecanoic acid, Myristic acid, Pyroglutamic acid, and Isocaproic acid, and the concentration of 2 metabolites was significantly down-regulated (Vanillic acid and Hyocholic acid). The 16 biomarkers associated with weight loss were further screened by machine learning algorithms.

[0093] 2.3 Machine learning dimension reduction screening and verification

[0094] The metagenomic and metabolomic data have the characteristics of high dimension and high collinearity. The Least Absolute Shrinkage and Selection Operator (LASSO) regression model is a regularization method for avoiding overfitting and achieving dimension reduction. In this design, the LASSO-logistic regression method was used to correct confounding factors for data dimension reduction, and multiple repeated cross-validation was performed at each step of screening. The 16 candidate biomarkers were further screened.

[0095] To prevent overfitting, the LASSO penalty parameter, lambda, was determined by 10-fold cross-validation using the mean squared error of cross-validation as the loss function. Specifically, all samples were randomly divided into 10 equal-sized parts. For each different lambda value, 9 of the 10 parts were used as the training dataset to fit the model parameters, and the remaining 1 part was used as the validation dataset to compute the mean squared error of the parameters. This process was repeated 10 times for each lambda, and different parts were used as the validation dataset each time. The mean squared error of the model was estimated each time, and the average of the 10 mean squared errors was taken after 10-fold cross-validation. The minimum mean squared error of cross-validation corresponded to the minimum value of lambda. To obtain a prediction model with a relatively small number of predictor variables, we selected lambda.min as the penalty parameter. Figure 5 Figure A in the accompanying drawings illustrates the trajectory of each coefficient of the independent variable with respect to the value of lambda in a LASSO regression, with the vertical coordinate being the value of the coefficient of the independent variable, and the lower horizontal coordinate being log(lambda) and the upper horizontal coordinate being the number of variables with non-zero coefficients in the model for each lambda. Since the covariates need to be corrected in the model, the covariates are not penalized, and all covariates are assigned non-zero regression coefficients and are always retained in the model. When the coefficient of a variable is equal to zero, it means that the variable is not included in the model at the corresponding lambda parameter. Figure 5 Figure B in the accompanying drawings illustrates the change of the mean squared error of the model with respect to lambda in a LASSO regression, and the optimal model is selected according to the mean squared error of cross-validation. Considering that different partitions of the dataset in cross-validation can affect the results and easily show high variance, the LASSO regression of 10-fold cross-validation is repeated 100 times, and the regression coefficient is selected as non-zero and 100% reproducible as a biomarker.

[0096] The analysis screen obtained a combination of biomarkers significantly associated with weight loss effect of 12 weeks dietary intervention; the combination of biomarkers includes 9 biomarkers, consisting of 6 bacterial species (Actinomyces sp. ICM47, Blautia sp. AF19-10LB, Lachnospira sp. NSJ-4, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis and Bacteroides nordii) and 3 fecal metabolites (Vanillic acid and Hyocholic acid and Isocaproic acid).

[0097] The analysis can be implemented by R package "glmnet".

[0098] 3. Establishment and evaluation of prediction model

[0099] The biomarker combination of the weight loss effect of the 12-week dietary intervention obtained based on the above screening is applied to the optimization and improvement of the weight loss effect prediction model after traditional clinical intervention. The currently reported traditional prediction model of weight loss effect after intervention is mainly established by factors such as age, gender, intervention measures and baseline initial body weight. Improved model 1 incorporates dietary characteristic indicators (baseline total energy intake and baseline total dietary fiber intake) into the traditional prediction model. Improved model 2 further incorporates 9 newly identified biomarker combinations as prediction factors into the model on the basis of improved model 1 (traditional prediction factors and dietary characteristic indicators). The present application uses the receiver operating characteristic curve (ROC) to evaluate the accuracy of the prediction model. The abscissa X represents (1-specificity), i.e. the false positive rate; the ordinate Y represents sensitivity, i.e. the true positive rate. The area under the curve (AUC) is used to represent the accuracy of the prediction, with a value between 0 and 1, and the larger the AUC, the higher the accuracy of the prediction. AUC can be obtained by trapezoidal rule, and in general cases, AUC of 0 indicates that the method has no ability to distinguish between responders and non-responders, 0.7-0.8 is considered to have good discrimination, and more than 0.8 is considered to have excellent discrimination. When evaluating the prediction ability of a binary logistic regression model, the value of C-statistic is equal to AUC. DeLong test is used to compare the statistical differences in AUC between models. In this design, the traditional prediction model of weight loss effect is composed of age, gender, intervention measures and baseline initial body weight, and the C-statistic is 0.672 (95% CI 0.557, 0.788), and when the dietary characteristic indicators (baseline total energy intake and baseline total dietary fiber intake) are added, the C-statistic increases to 0.722 (95% CI 0.614, 0.830) Figure 6 ). On the basis of the foregoing model, 9 newly identified biomarkers are further added as prediction factors, and the C-statistic is further improved to 0.939 (95% CI 0.891, 0.988), and DeLong test shows that compared with the traditional model and the clinical model, the prediction performance of the newly established prediction model is significantly improved: ΔAUC is 26.7% and 21.7% (P<0.001), respectively.

[0100] Those skilled in the art will readily understand that the above description is only the preferred embodiment of the present application and is not intended to limit the present application, and any modifications, equivalent replacements and improvements made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for screening gut microbiota and metabolite biomarkers for the effect of dietary intervention on weight loss in overweight or obese patients, characterized by, Comprising the following steps: (1) Collecting baseline fecal samples of patients, performing metagenomic sequencing, intestinal microbiome analysis and fecal metabolome detection; (2) Preprocessing the abundance data of intestinal flora and the concentration data of fecal metabolites, using one-tenth of the minimum detection value of the analyte to fill in the missing values below the detection limit, and log transforming the concentration of each marker; (3) Using a covariate-corrected conditional logistic model to screen biomarkers related to weight loss effect, the covariates of the model including age, gender, whether to adopt a healthy low-carbohydrate diet intervention, whether to adopt a time-limited diet intervention and baseline weight; (4) Using a LASSO-logistic regression model corrected for confounding factors to reduce the dimensionality of the related biomarkers, each step of screening is cross-validated by ten-fold, and finally a biomarker combination significantly related to weight loss effect after energy-restricted diet intervention is obtained; the biomarker combination comprises 9 biomarkers, the 9 biomarkers are 6 species and 3 metabolites; the 6 species are: Actinomyces sp. ICM47, Blautia sp. AF19-10LB, Lachnospira sp. NSJ-4, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis and Bacteroides nordii; the 3 metabolites are: Vanillic acid, Hyocholic acid and Isocaproic acid; the abundance of the 6 species is reduced in the weight loss group compared with the control group without weight loss; the concentration of Vanillic acid and Hyocholic acid in the 3 metabolites is reduced in the weight loss group compared with the control group without weight loss, and the concentration of Isocaproic acid in the 3 metabolites is increased in the weight loss group compared with the control group without weight loss.

2. The biomarker screening method of claim 1, wherein, In step (1), shotgun metagenomic sequencing and intestinal microbiome analysis are performed.

3. The biomarker screening method of claim 1, wherein, In step (1), the detection of fecal metabolites uses ultra-high performance liquid chromatography-mass spectrometry.

4. A biomarker screening system for gut microbiota and metabolites for weight loss effect in an overweight or obese patient after a reduced energy diet intervention, characterized in that, Comprising: A data acquisition module: collecting baseline fecal samples of patients, performing metagenomic sequencing, intestinal microbiome analysis and fecal metabolome detection; A data preprocessing module: preprocessing the abundance data of intestinal flora and the concentration data of fecal metabolites, using one-tenth of the minimum detection value of the analyte to fill in the missing values below the detection limit, and log transforming the concentration of each marker; The candidate biomarker screening module: a covariate-corrected conditional logistic model is used to screen biomarkers related to weight loss effect, and the covariates of the model include age, gender, whether to adopt a healthy low-carbohydrate diet intervention, whether to adopt a time-limited diet intervention, and baseline body weight; The machine learning dimension reduction biomarker screening module: a LASSO-logistic regression model is used to process the above candidate biomarkers, and ten-fold cross-validation is performed at each step of screening, and finally a biomarker combination significantly related to weight loss effect after energy-restricted diet intervention is obtained: the biomarker combination includes 9 biomarkers, which are 6 bacterial species and 3 metabolites; the 6 bacterial species are: Actinomycessp.ICM47, Blautia sp.AF19-10LB, Lachnospira sp.NSJ-4, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis and Bacteroides nordii; the 3 metabolites are: Vanillic acid, Hyocholic acid and Isocaproic acid; the abundance of the 6 bacterial species in the weight loss group is lower than that in the control group without weight loss; the concentrations of Vanillic acid and Hyocholic acid in the 3 metabolites in the weight loss group are lower than those in the control group without weight loss, and the concentration of Isocaproic acid in the 3 metabolites in the weight loss group is higher than that in the control group without weight loss.

5. Use of gut microbiota and metabolite biomarkers in the construction of a predictive model of the effectiveness of a dietary intervention for weight loss in an overweight or obese patient, characterized in that, The combination of the screened intestinal flora and metabolite biomarkers is added to the weight loss effect prediction model after traditional clinical intervention to reconstruct a new weight loss effect prediction model; The machine learning dimension reduction biomarker screening module: a LASSO-logistic regression model is used to process the above candidate biomarkers, and ten-fold cross-validation is performed at each step of screening, and finally a biomarker combination significantly related to weight loss effect after energy-restricted diet intervention is obtained: the biomarker combination includes 9 biomarkers, which are 6 bacterial species and 3 metabolites; the 6 bacterial species are: Actinomycessp.ICM47, Blautia sp.AF19-10LB, Lachnospira sp.NSJ-4, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis and Bacteroides nordii; the 3 metabolites are: Vanillic acid, Hyocholic acid and Isocaproic acid; the abundance of the 6 bacterial species in the weight loss group is lower than that in the control group without weight loss; the concentrations of Vanillic acid and Hyocholic acid in the 3 metabolites in the weight loss group are lower than those in the control group without weight loss, and the concentration of Isocaproic acid in the 3 metabolites in the weight loss group is higher than that in the control group without weight loss. The biomarker combination is 6 bacterial species and 3 metabolites; the 6 bacterial species are: Actinomyces sp. ICM47, Blautia sp. AF19-10LB, Lachnospira sp. NSJ-4, Mediterraneibacter glycyrrhizinilyticus, Dysosmobacter welbionis and Bacteroides nordii; the 3 metabolites are: Vanillic acid, Hyocholic acid and Isocaproic acid; The abundance of the 6 bacterial species is reduced in the weight loss group compared with the non-weight loss control group; the concentration of Vanillic acid and Hyocholic acid among the 3 metabolites is reduced in the weight loss group compared with the non-weight loss control group, and the concentration of Isocaproic acid among the 3 metabolites is increased in the weight loss group compared with the non-weight loss control group.

6. The use according to claim 5, wherein the compound is ###0002### The traditional prediction model is based on factors such as age, gender, intervention measures and baseline initial body weight; the new weight loss effect prediction model is based on the traditional prediction model, and then the newly identified biomarker combination is further included in the model as a prediction factor; the prediction performance of the model is evaluated and compared by the area under the receiver operating characteristic curve and the DeLong test.

7. Use according to claim 6, wherein The dietary characteristic index includes baseline total energy intake and baseline total dietary fiber intake.

Images