A biomarker and risk prediction model for Alzheimer's disease based on intestinal flora

Abstract

The present invention relates to an Alzheimer's disease marker and risk prediction model based on intestinal flora. The Alzheimer's disease marker based on intestinal flora includes intestinal flora, and the intestinal flora includes anaerobic Lawsonia sibiricum, Escherichia coli, Streptococcus salivarius, Asahidiella pseudomallei, Kluyveromyces pedis, intestinal Roseburia, Roseburia faecalis, Roseburia hominis, Firmicutes CAG424 bacteria and Clostridium CAG58 bacteria. The present invention mines Alzheimer's disease markers based on intestinal flora and further constructs an Alzheimer's disease risk prediction model based on intestinal flora, which has the characteristics of high detection accuracy, convenience, speed, safety and non-invasiveness, and has important clinical guidance significance for assisting the diagnosis of Alzheimer's disease-related indicators.

Description

Technical Field

[0001] The present invention belongs to the field of biotechnology and relates to an Alzheimer's disease marker and risk prediction model based on intestinal flora. Background Art

[0002] Alzheimer's disease (AD) is a chronic, progressive neurodegenerative disorder characterized by irreversible brain damage. It can lead to memory loss, loss of social and occupational function, executive dysfunction, speech and motor deficits, personality changes, and behavioral and psychological disturbances. The course of AD typically lasts 8-10 years, and its pathological hallmarks include neurofibrillary tangles, senile plaques, neuronal loss, and brain atrophy, accompanied by defects in acetylcholine synthesis at the cellular level.

[0003] Although there is no cure at present, Alzheimer's disease is not untreatable. Existing treatments can alleviate symptoms and slow the progression of the disease. In addition, early diagnosis and treatment are crucial to improving patients' quality of life. Currently, the diagnosis of Alzheimer's disease usually requires medical history, physical examination, cognitive function assessment, and a series of diagnostic tests, such as blood tests, biochemical tests, thyroid function tests, vitamin B12 levels, CT, MRI, cerebrospinal fluid analysis, etc. The early diagnosis of Alzheimer's disease faces some challenges and shortcomings, mainly including the following: Low diagnostic rate: The public's lack of awareness of Alzheimer's disease has resulted in many patients failing to receive timely diagnosis and treatment; Hidden early symptoms: Alzheimer's disease is latent in its early stages and difficult to detect; The misdiagnosis rate is high; Limitations of biomarker detection: Although based on the ATN diagnostic criteria for Alzheimer's disease, biomarkers can They are divided into Aβ pathological biomarkers, tau pathological biomarkers and biomarkers of neurodegenerative diseases, but these detection methods have certain limitations. For example, the specificity of lumbar puncture is relatively lower than that of PET-CT, and it is traumatic and has the risk of infection; limitations of imaging examinations: although neuroimaging examinations such as MRI and PET-CT are important, they are expensive and have high requirements for equipment and professional skills, which limits their widespread application in early diagnosis; limitations of blood marker detection: although blood marker detection provides the possibility of non-invasive detection, the specificity and sensitivity of these detection methods are still under study, and have not yet been widely used in clinical diagnosis.

[0004] In summary, discovering new AD early diagnosis biomarkers and developing corresponding disease risk prediction tools are of great significance in the field of AD early diagnosis. Summary of the Invention

[0005] In response to the shortcomings of existing technologies and actual needs, the present invention provides an Alzheimer's disease marker and risk prediction model based on intestinal flora, in order to solve the current problems of lack of early diagnostic tools for Alzheimer's disease, complex detection, and low accuracy.

[0006] To achieve this object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides an Alzheimer's disease marker based on intestinal flora, wherein the marker includes intestinal flora, and the intestinal flora includes Lawsonibacter asaccharolyticus, Escherichia coli, Streptococcus salivarius, Asaccharobacter celatus, Adlercreutzia equolifaciens, Roseburia intestinalis, Roseburia faecis, Roseburia hominis, Firmicutes bacterium CAG 424, and Clostridium sp CAG 58.

[0008] In the present invention, the relative abundance information of intestinal flora in people with genetic risk of AD was analyzed based on sequencing technology, and Alzheimer's disease markers based on intestinal flora were mined. It was found that compared with healthy people, the abundance of Anaerobic Bacteria Lawsoniae in people with genetic risk of AD was reduced, the abundance of Escherichia coli was reduced, the abundance of Streptococcus salivarius was increased, the abundance of Assassinella wiggling was reduced, the abundance of Kluyveromyces pedis was reduced, the abundance of intestinal Roseburia was increased, the abundance of Roseburia faecalis was reduced, the abundance of Roseburia hominis was reduced, the abundance of Firmicutes CAG424 was increased, and the abundance of Clostridium CAG58 was reduced.

[0009] In a second aspect, the present invention provides the use of the intestinal flora-based Alzheimer's disease marker and / or its detection reagent described in the first aspect in the preparation of a product for predicting the risk of Alzheimer's disease.

[0010] It is understood that reagents and equipment that can detect the level of intestinal flora in the art are all applicable to the present invention.

[0011] In a third aspect, the present invention provides a kit for predicting the risk of Alzheimer's disease, wherein the kit comprises a reagent for detecting the abundance of the intestinal flora-based Alzheimer's disease markers described in the first aspect.

[0012] In a fourth aspect, the present invention provides a method for constructing a model for predicting the risk of Alzheimer's disease, the method comprising:

[0013] Detect the abundance of the intestinal flora-based Alzheimer's disease markers described in the first aspect in healthy subjects and subjects at risk of Alzheimer's disease; use the obtained abundance data as a training set, and construct a model for predicting Alzheimer's disease risk based on a machine learning algorithm.

[0014] In the present invention, a model for predicting the risk of Alzheimer's disease is further developed based on the mined markers for early risk prediction of Alzheimer's disease.

[0015] Preferably, the machine learning algorithm includes any one of random forest, naive Bayes, support vector machine or decision tree.

[0016] In a fifth aspect, the present invention provides a model for predicting the risk of Alzheimer's disease, wherein the model for predicting the risk of Alzheimer's disease is constructed by the method for constructing a model for predicting the risk of Alzheimer's disease described in the fourth aspect.

[0017] Preferably, the input data of the model is the abundance of the Alzheimer's disease marker based on intestinal flora described in the first aspect; the output variable is the probability of having the risk of Alzheimer's disease; and the judgment criterion for having the risk of Alzheimer's disease is: the predicted probability of Alzheimer's disease risk is greater than 0.5.

[0018] In the present invention, a machine learning model constructed using a combination of 10 intestinal flora markers and their abundance levels is used to predict the early risk of Alzheimer's disease. The AUC value can reach 0.859. It has the characteristics of high detection accuracy, convenience, speed, safety and non-invasiveness, and has important clinical guidance significance for assisting the diagnosis of Alzheimer's disease-related indicators.

[0019] In a sixth aspect, the present invention provides a device for predicting the risk of Alzheimer's disease, the device comprising a detection unit and an evaluation unit;

[0020] The detection unit is configured to perform the following steps:

[0021] Detecting the abundance of the intestinal flora-based Alzheimer's disease markers described in the first aspect in the test sample;

[0022] The evaluation unit is configured to perform the following steps:

[0023] The abundance detected by the detection unit is input into the model for predicting the risk of Alzheimer's disease described in the fifth aspect, and the probability of having the risk of Alzheimer's disease is output; the judgment standard for having the risk of Alzheimer's disease is: the predicted probability of the risk of Alzheimer's disease is greater than 0.5.

[0024] Preferably, the sample to be tested includes a stool sample.

[0025] In a seventh aspect, the present invention provides an electronic device comprising one or more processors and a memory for storing executable instructions, wherein the one or more processors are configured to call the executable instructions stored in the memory to implement the functions of the device for predicting the risk of Alzheimer's disease as described in the sixth aspect.

[0026] In an eighth aspect, the present invention provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the functions of the device for predicting the risk of Alzheimer's disease as described in the sixth aspect.

[0027] In a ninth aspect, the present invention provides the use of the intestinal flora-based Alzheimer's disease markers described in the first aspect as targets in screening drugs for treating or preventing Alzheimer's disease.

[0028] Specifically, the screening includes selecting drugs for treating or preventing Alzheimer's disease by analyzing the effects of candidate drugs on the intestinal flora-based Alzheimer's disease markers in the subject.

[0029] Compared with the prior art, the present invention has at least the following beneficial effects:

[0030] In the present invention, the relative abundance information of the intestinal flora of AD genetic risk individuals is analyzed based on sequencing technology, and Alzheimer's disease markers based on the intestinal flora are mined. A machine learning model constructed with a specific combination of intestinal flora markers and their abundance levels is used for early risk prediction of Alzheimer's disease. It has the characteristics of high detection accuracy, convenience, speed, safety and non-invasiveness, and has important clinical guidance significance for assisting the diagnosis of Alzheimer's disease-related indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 ROC curve for predicting the risk of Alzheimer's disease. DETAILED DESCRIPTION

[0032] The technical solution of the present invention will be further described below with reference to the accompanying drawings and through specific embodiments. However, the following examples are merely simplified examples of the present invention and do not represent or limit the scope of protection of the present invention. The scope of protection of the present invention shall be subject to the claims.

[0033] If no specific techniques or conditions are specified in the examples, the experiments were carried out according to the techniques or conditions described in the literature in the field or according to the product instructions. If no manufacturer is specified for the reagents or instruments used, they are all conventional products that can be purchased through regular channels.

[0034] In a specific embodiment of the present invention, information on the relative abundance of intestinal flora is obtained by high-throughput sequencing analysis of the intestinal flora genome using the Illumina NovaSeq 6000 sequencing platform. DNA from the intestinal flora is randomly fragmented using a DNA ultrasonic fragmenter, and the fragmented DNA fragments are amplified and constructed. Qualified sample libraries are used for library construction and high-throughput sequencing on the Illumina NovaSeq 6000 sequencer. The sequencing data undergoes a series of bioinformatics analysis and machine learning methods, including filtering low-quality data, removing host sequences, species annotation, differential analysis, and machine learning model training. This yields analytical results such as the sample's microbial composition and abundance, species with significant differences between groups, an Alzheimer's disease risk prediction model, and species that contribute significantly to the model. Based on the relative abundance of intestinal flora at the genus level in the diseased and normal control groups, a machine learning approach is applied to construct an Alzheimer's disease risk prediction model at the genus level, and to screen for intestinal flora that contribute significantly to the Alzheimer's disease risk prediction model.

[0035] In a specific embodiment of the present invention, a device for predicting the risk of Alzheimer's disease is also provided, the device comprising a detection unit and an evaluation unit; the detection unit is used to perform the following operations: detecting the abundance of the intestinal flora-based Alzheimer's disease marker in the sample to be tested; the evaluation unit is used to perform the following operations: inputting the abundance detected by the detection unit into the model for predicting the risk of Alzheimer's disease, and outputting the probability of having the risk of Alzheimer's disease; the judgment criterion for having the risk of Alzheimer's disease is: the predicted probability of the risk of Alzheimer's disease is greater than 0.5.

[0036] In a specific embodiment of the present invention, an electronic device is also provided, which includes a memory, a processor, and a computer program (instructions) stored in the memory and executable on the processor. When the processor executes the computer program, the device realizes the function of predicting the risk of Alzheimer's disease.

[0037] In another specific embodiment of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and the computer program has a program code, which realizes the function of a device for predicting the risk of Alzheimer's disease when running in a corresponding processor, controller, computing device or terminal.

[0038] Those skilled in the art should understand that the solutions proposed in the embodiments of the present invention may be implemented in various forms of hardware, software, firmware, dedicated processors, or a combination thereof.

[0039] Example 1

[0040] In this example, the intestinal flora of healthy people and AD patients were analyzed and markers were mined.

[0041] The fecal samples of 49 healthy individuals (Control group) and 49 individuals with genetic risk of AD but without cognitive impairment (Pre-AD) (samples and data were from Knight Alzheimer's Disease Research Center) were analyzed for gut microbiota abundance.

[0042] Fecal samples were collected from diseased individuals and normal healthy control donors, and total DNA was isolated using a nucleic acid isolation kit. Primers were designed based on conserved regions, sequencing adapters were added to the ends of the primers, PCR amplification was performed, and the products were purified, quantified, and normalized to form a sequencing library. The constructed library was first subjected to library quality inspection, and the library passing the quality inspection was sequenced by the Illumina NovaSeq 6000 system for high-throughput sequencing. Reads containing more than 10 low-quality (<Q20) bases were filtered from the raw data. Then, the MetaPhlAn3 software was used to annotate the microbial species of the filtered reads. The Shannon index was calculated using the "vegan" package in R to detect the bacterial α-diversity. After obtaining the genus classification profile, the ANCOM-BC bias correction method was used to compare the abundance differences at the genus level between the AD genetic risk and normal control groups (p<0.05).

[0043] The specific results are shown in Table 1. Compared with the healthy individuals, the abundances of Anaerotruncus labsonii, Escherichia coli, Streptococcus salivarius, Aspergillus awamori, Kluyvera equi, Roseburia intestinalis, Rothia faecis, Rothia hominis, CAG424 of Firmicutes, and CAG58 of Clostridium were decreased in the individuals with AD genetic risk.

[0044] Table 1

[0045] Fungus Healthy control group (mean ± standard deviation) AD genetic risk group (mean ± standard deviation) Anaerobic bacteria Lawsonia Pasteurella 0.10±0.15 0.07±0.16 Escherichia coli 1.53±3.67 0.25±1.07 Streptococcus salivarius 0.76±1.57 1.18±2.90 Asahini wig 0.28±0.47 0.11±0.13 Kluyveromyces pediatrics 0.20±0.30 0.13±0.21 Roseburia intestinalis 0.65±1.20 1.00±1.75 Roseburia faecalis 3.06±4.79 2.03±3.67 Roseburia hominis 0.56±0.89 0.55±0.74 Firmicutes CAG424 0.16±0.52 0.17±0.98 Clostridium CAG58 0.25±0.34 0.23±0.46

[0046] Example 2

[0047] In this example, an AD risk prediction model was constructed based on the gut microbiota abundances detected in Example.

[0048] The 49 healthy people and 49 people at genetic risk of AD in Example 1 were randomly divided into three groups, and the groups had the same or similar number of people. First, two groups of samples were randomly selected as training sets, and the remaining group of samples was used as a test set. Secondly, the grouping information of the training set samples and the relative abundance of the genus level of the intestinal flora were used to train random forest, naive Bayes, support vector machine, and decision tree models to construct a risk prediction model for AD. Then, the Bootstrap method was used to select 40 samples with replacement from the training set to test the four AD risk prediction models. Finally, the random forest model based on the relative abundance of the genus level of the intestinal flora showed that the intestinal flora of the Pre-AD group was significantly different from that of the Control group, and the AUC value of the model reached 1, the sensitivity was 1, and the specificity was 1, confirming that the random forest model was an AD risk prediction model.

[0049] Example 3

[0050] This example tests and verifies the AD risk prediction model constructed in Example 2.

[0051] The relative abundance of the intestinal flora genus level of the test set samples in Example 2 was input into the AD risk prediction model trained in Example 2, and the probability of having AD risk was output. When the predicted probability of AD risk is greater than 0.5, it is determined to be an AD risk population. Based on the actual grouping information of the test set samples and the AD risk prediction results, it was found that the random forest model of the relative abundance of the intestinal flora genus level can significantly distinguish the intestinal flora of the Pre-AD group in the test set from that of the Control group, and the AUC value of the model reached 0.859 ( Figure 1 ), with a sensitivity of 0.667 and a specificity of 1.000.

[0052] In summary, the present invention analyzes the relative abundance information of the intestinal flora of AD genetic risk individuals based on sequencing technology, mines Alzheimer's disease markers based on the intestinal flora, and constructs a machine learning model based on a specific combination of intestinal flora markers and their abundance levels for early risk prediction of Alzheimer's disease. The model has the characteristics of high detection accuracy, convenience, speed, safety and non-invasiveness, and has important clinical guidance significance for assisting the diagnosis of Alzheimer's disease-related indicators.

[0053] The applicant declares that the above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention fall within the scope of protection and disclosure of the present invention.

Claims

1. Use of a gut microbiota-based Alzheimer's disease marker detection reagent in the preparation of a product for predicting Alzheimer's disease risk, characterized in that: The Alzheimer's disease marker based on intestinal flora is Anaerobic Bacillus Lawsoniae ( Lawsonibacter asaccharolyticus ), Escherichia coli ( Escherichia coli ), Streptococcus salivarius ( Streptococcus salivarius ), Assahibacterium wigglingii ( Asaccharobacter celatus ), Kluyveromyces pedis ( Adlercreutzia equolifaciens ), enteric Roseburia ( Roseburia intestinalis ), Roseburia faecalis ( Roseburia faecis ), Roseburia hominis ( Roseburia hominis )、Firmicutes CAG424 bacteria ( Firmicutes bacterium CAG 424 ) and Clostridium CAG58 ( Clostridium sp CAG 58 ).

2. Application of intestinal flora-based Alzheimer's disease marker detection reagents in the preparation of kits for predicting Alzheimer's disease risk; The Alzheimer's disease marker based on intestinal flora is Anaerobic Bacillus Lawsoniae ( Lawsonibacter asaccharolyticus ), Escherichia coli ( Escherichia coli ), Streptococcus salivarius ( Streptococcus salivarius ), Assahibacterium wigglingii ( Asaccharobacter celatus ), Kluyveromyces pedis ( Adlercreutzia equolifaciens ), enteric Roseburia ( Roseburia intestinalis ), Roseburia faecalis ( Roseburia faecis ), Roseburia hominis ( Roseburia hominis )、Firmicutes CAG424 bacteria ( Firmicutes bacterium CAG 424 ) and Clostridium CAG58 ( Clostridium sp CAG 58 ).

3. A method for constructing a model for predicting the risk of Alzheimer's disease, characterized in that: The method comprises: Detecting the abundance of gut microbiota-based Alzheimer's disease markers in healthy individuals and those at risk for Alzheimer's disease; Using the obtained abundance data as a training set, a model for predicting Alzheimer's disease risk was constructed based on a machine learning algorithm; The Alzheimer's disease marker based on intestinal flora is Anaerobic Bacillus Lawsoniae ( Lawsonibacter asaccharolyticus ), Escherichia coli ( Escherichia coli ), Streptococcus salivarius ( Streptococcus salivarius ), Assahibacterium wigglingii ( Asaccharobacter celatus ), Kluyveromyces pedis ( Adlercreutzia equolifaciens ), enteric Roseburia ( Roseburia intestinalis ), Roseburia faecalis ( Roseburia faecis ), Roseburia hominis ( Roseburia hominis )、Firmicutes CAG424 bacteria ( Firmicutes bacterium CAG 424 ) and Clostridium CAG58 ( Clostridium sp CAG 58 ).

4. The method for constructing a model for predicting the risk of Alzheimer's disease according to claim 3, wherein: The machine learning algorithm includes any one of random forest, naive Bayes, support vector machine or decision tree.

5. A device for predicting the risk of Alzheimer's disease, characterized in that: The device comprises a detection unit and an evaluation unit; The detection unit is configured to perform the following steps: Detect the abundance of Alzheimer's disease markers based on gut microbiota in the test samples; The Alzheimer's disease marker based on intestinal flora is Anaerobic Bacillus Lawsoniae ( Lawsonibacter asaccharolyticus ), Escherichia coli ( Escherichia coli ), Streptococcus salivarius ( Streptococcus salivarius ), Assahibacterium wigglingii ( Asaccharobacter celatus ), Kluyveromyces pedis ( Adlercreutzia equolifaciens ), enteric Roseburia ( Roseburia intestinalis ), Roseburia faecalis ( Roseburia faecis ), Roseburia hominis ( Roseburia hominis )、Firmicutes CAG424 bacteria ( Firmicutes bacterium CAG 424 ) and Clostridium CAG58 ( Clostridium sp CAG 58 ); The evaluation unit is configured to perform the following steps: inputting the abundance detected by the detection unit into the model for predicting the risk of Alzheimer's disease as claimed in claim 3, and outputting the probability of having the risk of Alzheimer's disease; The criterion for determining the risk of Alzheimer's disease is: the predicted probability of Alzheimer's disease risk is greater than 0.

5.

6. An electronic device comprising one or more processors and a memory for storing executable instructions, characterized in that: The one or more processors are configured to call the executable instructions stored in the memory to implement the functions of the device for predicting the risk of Alzheimer's disease as claimed in claim 5.

7. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the functions of the device for predicting the risk of Alzheimer's disease according to claim 5 are realized.

Images