Infant micro-ecological flora development evaluation method and system

Abstract

The invention provides an infant micro-ecological flora development evaluation method and system, relates to a set of judgment models and standards, and comprises an infant micro-ecological development maturity classification model, an infant micro-ecological development age regression model, a micro-ecological age development curve and a reference interval. According to the method, the DNA of the microorganisms in the feces of the infants is extracted and sequenced, the 27-month-old infants are analyzed on the basis of a statistical method and are key nodes for micro-ecological development maturity of the Chinese infants, a machine learning classification model is constructed with the key nodes as a boundary to judge whether flora development is mature or not, and the AUC can reach 97% or above. And on this basis, the micro-ecological development age of the infant is predicted according to a machine learning regression model. A regression prediction model is applied to predict and set a development judgment standard reference range on a large-scale data set, and a new reference standard is provided for infant micro-ecological development maturity evaluation.

Description

Technical Field

[0001] The present invention relates to the field of intestinal microecological analysis, and in particular to a method and system for evaluating the age development of infants' microecological information using microecological information. Background Art

[0002] Intestinal microorganisms are closely related to human health, especially in the early development stage of infants and young children, and play an important role in the growth, development and maturation of immunity. In the first 1000 days of life, the richness and diversity of the intestinal flora of infants and young children are in a process of continuous growth and dynamic maturation, which is the best time to intervene in the development of the flora. Therefore, early evaluation of the development of the intestinal flora of infants and young children is of great significance to the subsequent healthy growth of the individual, and can guide subsequent nutritional health and dietary feeding methods.

[0003] At present, the judgment of the maturity of infant microecology is relatively simple, without considering the gradual maturity of infant microecology before the age of 3, or predicting age through one or two microbial markers, or predicting intestinal microecological age through simple regression models, which are out of line with the characteristics of the mature development of microecological flora. Therefore, a scientific method and standard for judging the development of infant microecological flora is urgently needed. Summary of the invention

[0004] Purpose of the invention: In view of the deficiencies of the above-mentioned prior art, the purpose of the present invention is to provide a method and system for evaluating the development of infant microecological flora, propose a classification model for the maturity of infant microecological development, an age regression model for infant microecological development, and a microecological age development curve, so as to conduct a scientific and reasonable assessment of whether the infant microecological development is normal.

[0005] Technical solution: To achieve the above-mentioned invention object, the present invention adopts the following technical solution:

[0006] The first aspect of the present invention provides a method for evaluating the development of infant microecological flora, comprising the following steps:

[0007] Extract and sequence fecal microorganisms of infants and young children to obtain microecological flora composition data, and classify them using a microecological maturity judgment model for infants and young children to determine whether the microecological flora is mature; the microecological maturity judgment model for infants and young children is constructed based on the best classification age obtained by statistics;

[0008] If it is not mature, the microecological age is predicted using a machine learning regression model to obtain the predicted microecological age;

[0009] According to the predicted microecological age and in comparison with the microecological age development curve, a risk assessment is conducted on whether the infant's development is ahead of schedule or delayed; the microecological age development curve is predicted on a large-scale data set based on a machine learning regression model and a reference range for developmental judgment standards is set.

[0010] Furthermore, the method for constructing the infant microecological maturity judgment model comprises the following steps:

[0011] Collect and organize the original sequencing data of infant intestinal microorganisms and their clinical information, analyze and integrate the sequencing data, and obtain the intestinal microbial relative abundance information table for constructing the microecological development maturity classification model;

[0012] All samples were divided into two categories: immature and mature, with the actual age of the samples ranging from 18 to 36 months as the classification age;

[0013] For multiple different classification result data, feature selection is performed, and a microecological development maturity classification model is constructed, and the accuracy of classification prediction results is statistically analyzed;

[0014] Compare the accuracy of different classification moon age prediction results and determine the best classification moon age;

[0015] Using the optimal classification age as the dividing line, a machine learning method was used to construct a model for judging the maturity of infant microecology.

[0016] Furthermore, the optimal classification age is 27 months.

[0017] Furthermore, the method for predicting the microecological age of infants and young children using a machine learning regression model comprises the following steps:

[0018] The intestinal microbial composition information of each individual is obtained through sequencing, and the relative abundance of each bacterial community is calculated;

[0019] The Boruta method was used to screen key biomarkers for evaluating the developmental maturity of an individual's intestinal microecology;

[0020] Based on the key biomarker set, an individual intestinal microecological age prediction model is iteratively optimized and verified, and the optimal preset number of multiple models are selected for subsequent use;

[0021] The optimal multiple models were used to predict the intestinal microecological age of the test samples, and the average value after removing the highest and lowest values ​​was used as the final individual evaluation result.

[0022] Furthermore, the method for generating the microecological age development curve and calculating the reasonable reference interval of microecological development age at different months of age comprises the following steps:

[0023] Collect infant microecological flora data, including raw sequencing data and phenotypic information, and construct a reference data set;

[0024] Analyze the collected raw data to obtain the bacterial community composition and structure data of each sample;

[0025] The microecological age of each sample was calculated using a machine learning regression model;

[0026] A scatter plot was constructed based on the actual age and microecological age of each sample in the reference data set. The microecological age development curve was fitted based on the median of the microecological age corresponding to each month of age. The lower limit of the microecological development age was fitted according to the preset upper limit quantile, and the upper limit of the microecological development age was fitted according to the preset lower limit quantile.

[0027] Furthermore, the method for evaluating the development of infant microecological flora also includes an analysis of the association between allergies and whether the microecological age development is abnormal, including the following steps:

[0028] Collect fecal samples from infants and young children to obtain microecological flora data, and collect phenotypic information on whether each sample corresponds to an individual with allergies;

[0029] The infant microecological maturity judgment model is used to determine whether the microecological age is mature. For immature samples, the machine learning regression model is used to predict the microecological age, and the microecological age development curve is combined to determine whether the microecological age development is abnormal;

[0030] An association analysis was conducted between the occurrence of allergic diseases and abnormal microecological age development.

[0031] Based on the same inventive concept, the second aspect of the present invention provides an infant microecological flora development assessment system, comprising:

[0032] A microecological maturity judgment module is used to extract and sequence fecal microorganisms of infants and young children to obtain microecological flora composition data, and classify them using an infant microecological maturity judgment model to judge whether the microecological flora is mature; the infant microecological maturity judgment model is constructed based on the best classification age obtained by statistics;

[0033] The microecological age prediction module is used to predict the microecological age using a machine learning regression model when the microecological system is not mature, and obtain the predicted microecological age;

[0034] And a risk assessment module, which is used to conduct risk assessment on whether the development of infants and young children is ahead of schedule or delayed based on the predicted microecological age and the microecological age development curve; the microecological age development curve is predicted on a large-scale data set based on a machine learning regression model and a reference range of development judgment standards is set.

[0035] Furthermore, the infant microecological flora development assessment system also includes an allergy association analysis module for analyzing the association between allergies and whether the microecological age development is abnormal.

[0036] The third aspect of the present invention provides a computer system, comprising a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor, wherein the computer program / instructions, when executed by the processor, implement the steps of the method for evaluating the development of the microecological flora of infants and young children.

[0037] A fourth aspect of the present invention provides a computer program product, comprising a computer program / instructions, which, when executed by a processor, implement the steps of the method for evaluating the development of infant microecological flora.

[0038] Beneficial effects: The present invention provides a method for evaluating the development of infant microecological flora, which combines machine learning classification model and regression model, and can accurately predict and evaluate the microecological age development of infants of different ages according to the composition structure of fecal microorganisms. First, the classification model is used to judge whether the infant microecological development is mature or not, and the machine learning regression model is used to accurately predict the microecological development age for immature samples, and the reference curve is compared to accurately evaluate whether the development is normal. Compared with the prior art, the present invention has the following advantages: 1. The present invention provides a model for judging the maturity of infant microecological flora. Through the study of the microecological development characteristics of Chinese infants and young children, it is determined that 27 months of age is the key node for the maturity of infant microecological flora. 2. The present invention constructs a set of methods for judging the maturity of infant microecological flora, and the combination of microecological maturity judgment model and microecological age prediction model can effectively improve the accuracy of model construction and prediction. 3. Based on a large-scale data set (microecological samples of infants of different ages around the world), the present invention constructs the infant microecological age development curve and upper and lower reference intervals for the first time, which can scientifically and reasonably evaluate whether the infant microecological development is normal. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is an overall flow chart of the method for evaluating the development of infant microecological flora according to an embodiment of the present invention.

[0040] Figure 2 This is a comparison diagram of the effects of the microecological classification model with different age groups as the dividing line in the embodiment of the present invention.

[0041] Figure 3 This is a rendering of the microecological maturity judgment model in an embodiment of the present invention.

[0042] Figure 41 is a diagram showing the effect of predicting the microecological development age of Chinese children in an embodiment of the present invention, wherein (a) shows the prediction effect on the training set, and (b) shows the prediction effect on the test.

[0043] Figure 5 It is the infant microecological age development curve and reference interval diagram in the embodiment of the present invention.

[0044] Figure 6 This is a flowchart of the operation of the infant microecological flora development assessment system according to an embodiment of the present invention.

[0045] Figure 7 Graph showing the relationship between intestinal age development and allergy incidence in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0047] Example 1

[0048] like Figure 1 As shown, the embodiment of the present invention discloses a method for evaluating the development of microecological flora in infants and young children, which mainly includes the following steps:

[0049] Extract and sequence fecal microorganisms of infants and young children to obtain data on the composition of microecological flora, and classify them using the infant microecological maturity judgment model to determine whether the microecological flora is mature; the infant microecological maturity judgment model is constructed based on the best classification age obtained by statistics;

[0050] If it is not mature, the microecological age is predicted using a machine learning regression model to obtain the predicted microecological age;

[0051] According to the predicted microecological age and in comparison with the microecological age development curve, a risk assessment is conducted on whether the infant's development is ahead of schedule or delayed; the microecological age development curve is predicted on a large-scale data set based on a machine learning regression model and a reference range for developmental judgment standards is set.

[0052] Based on more than 800 cases of Chinese infant microbial flora data, the embodiment of the present invention calculates the optimal segmentation point of infant flora maturity by statistical methods, and constructs a classification model based on this to predict whether the microecology is mature. In addition, in children under 3 years old, a regression model is constructed to predict the age of microecological development. And based on the prediction results of more than 20,000 cases of data worldwide, a microecological age development curve and reference interval are constructed.

[0053] The following is a detailed introduction to the infant microecological maturity judgment model based on the machine learning classification model, the construction of the microecological age prediction model based on the machine learning regression model, and the generation of the microecological age development curve and the calculation of the reference interval involved in the embodiments of the present invention.

[0054] The infant microecological maturity judgment model based on machine learning classification model can be used to evaluate the development and maturity of infant microecological flora, mainly involving the optimal segmentation point of microecological flora and classification model and markers. The main steps of constructing the microecological maturity judgment model include:

[0055] Step S101: Arrange and merge the fecal 16S rRNA sequencing data of 872 Chinese infants and young children from different clinical projects. Process the sequencing data based on the 16S amplicon analysis process of DADA2 (https: / / benjjneb.github.io / dada2 / tutorial.html). The operation steps are as follows:

[0056] (1) First, the 16S rRNA sequence of each sample was obtained by removing the linker and splitting, and then quality filtering, noise removal, merging and chimera removal were performed to obtain an ASV (Amplicon Sequence Variant) sequence with 100% consistency.

[0057] (2) The ASV sequence was aligned to the Silva database (v138) to obtain the species name of the sequence.

[0058] (3) Integrate the naming of different reference sequences, merge the abundance of sequences annotated to the same species into the abundance of the same species, and obtain the bacterial community composition data of the samples used;

[0059] Step S102: According to the actual age of each child, all samples are divided into two categories: immature and mature, with 18 to 36 months as the classification age, and microecological development maturity classification models are constructed respectively, and the accuracy of classification prediction results is statistically analyzed. The operation steps are as follows:

[0060] (1) After standardizing the original species abundance table, age information and species abundance table were merged.

[0061] (2) The samples were divided into immature and mature categories according to age nodes to obtain the initial data for model construction. Intestinal microbial feature selection was performed, and the top 30 contributing microorganisms were selected to construct the model. The AUC value representing the prediction accuracy of the model was recorded and repeated 20 times.

[0062] In this embodiment, the selected microbial marker set for the mature development of infant microecology is: Enterococcus, UBA1819, Faecalibacterium, Dialister, Bacteroides, Incertae Sedis, Ruminococcus torques group, Staphylococcus, Butyricicoccus, Eubacterium nodatumgroup, Eubacterium eligens group, Lachnoclostridium, Methyloversatilis, Pseudomonas, Monoglobus, Abiotrophia, Oscillibacter, Blautia, Eubacteriumventriosum group, Sellimonas, Eisenbergiella, Saccharimonadaceae, TM7x, Clostridiuminnocuum group, Lachnospiraceae NK4A136 group, Ruminococcus gnavusgroup, Flavonifractor, Eubacterium fissicatena group, Holdemania and CandidatusSoleaferrea.

[0063] Step S103, 18 to 36, repeat the steps in (2) for each age category, and finally compare the differences in AUC values ​​of adjacent category prediction results through t-test. Figure 2 As shown in Figure 2, there is a significant difference in classification between 27 and 28 months, that is, the improvement in classification effect gradually reaches saturation after 28 months. Figure 2 Therefore, 27 months is the best age to determine developmental maturity.

[0064] Step S104: Based on the results of the previous step, the samples were divided into mature and immature categories with 27 months as the boundary, markers were screened in the intestinal microecological data, and a machine learning binary classification model was constructed. The AUC of the constructed infant microecological maturity judgment model can reach 97.3%. Figure 3 .

[0065] The construction of the infant microecological age prediction model based on machine learning regression model mainly involves data collection and collation, key biomarker determination, and regression model training. The specific steps include:

[0066] Step S201, data collection and collation: Fecal samples of 872 infants were collected for 16S rRNA sequencing to obtain raw sequencing files. The distribution of samples in each age group is shown in Table 1. The raw sequencing data was then processed based on the 16S amplicon analysis process of DADA2 to obtain a species abundance composition table for all samples. The operation steps are as follows:

[0067] (1) First, the original sequence was removed from the linker and split to obtain the 16S rRNA sequence of each sample, and then it was quality filtered, denoised, merged, and chimera removed to obtain an ASV (Amplicon Sequence Variant) sequence with 100% consistency.

[0068] Table 1. Distribution of samples in different age groups

[0069]

[0070]

[0071] (2) The ASV sequence was aligned to the Silva database (v138) to obtain the species name of the sequence.

[0072] (3) Integrate the naming of different reference sequences, merge the abundance of sequences annotated to the same species into the abundance of the same species, and obtain the microbial composition data of the sample.

[0073] Step S202, key biomarkers: The data in step S201 are split into a training set and a test set at a ratio of 4:1. After Boruta feature selection and fitting with the actual developmental age, 34 key core bacteria for intestinal microecological developmental maturity are determined.

[0074] In this embodiment, the selected microbial marker set for predicting the microecological development age of infants and young children is: Bifidobacteriaceae, Incertae Sedis, Saccharimonadaceae, Abiotrophia, Actinomyces, Bacteroides, Bifidobacterium, Blautia, Butyricicoccus, Clostridium sensu stricto1, Dialister, Enterococcus, Erysipelatoclostridium, Escherichia-Shigella, Faecalibacterium, Flavonifractor, Intestinibacter, Methyloversatilis, Monoglobus, Parabacteroides, Phascolarctobacterium, Pseudomonas, Romboutsia, Rothia, Staphylococcus, Streptococcus, TM7x, UBA1819, [Clostridium]innocuum group, [Eubacterium]hallii group, [Eubacterium]nodatum group, [Eubacterium]ventriosum group, [Ruminococcus]gnavus group and [Ruminococcus]torques group.

[0075] Step S203, model training: The data in step S201 are randomly split multiple times in a ratio of 4:1 into a training set and a test set. The 34 core bacteria in the training set in the previous step are used as features to train a regression model for predicting the intestinal microecological age multiple times. The random forest is selected for model training; the optimal ten training models are selected, and the intestinal microecological development age in months is output according to the model.

[0076] Step S204: Based on the prediction results of the ten optimal models, the highest and lowest values ​​are removed and the average value is taken to determine the developmental age in months of each sample.

[0077] Figure 4 The performance of the microecological age regression prediction model on the training set and the test set is given. In particular, in the test set, there is a correlation of 0.83 between the predicted age and the actual age.

[0078] In this embodiment, the intestinal microecological age development curve is fitted and drawn, and the Q10 and Q90 intervals are calculated, which are defined as the reasonable development interval range. The specific steps include:

[0079] Step S301: Collect global infant fecal microecological data. Screen infant microecological flora data from public databases such as NCBI and SRA, including raw sequencing data and phenotypic information (such as age, gender, country, etc.), and construct a reference data set.

[0080] Step S302: Use a unified data analysis process to process and analyze the raw data collected in step S301 to obtain the bacterial flora composition structure data of each sample.

[0081] Step S303, using the constructed microecological age prediction model to calculate the microecological age of each sample in the public data set;

[0082] Step S304: construct a scatter plot based on the actual age and microecological age of each sample in the reference database, and fit the microecological development age reference curve according to the median (Q50) of the microecological age corresponding to each month of age, fit the lower limit of the microecological development age according to the 10% quantile (Q10), and fit the upper limit of the microecological development age according to the 90% quantile (Q90).

[0083] The infant microecological age development curve and reference interval fitted in this embodiment are as follows Figure 5 .

[0084] Example 2

[0085] The embodiment of the present invention discloses a system for evaluating the development of infant microecological flora, including: a microecological maturity judgment module for extracting and sequencing infant fecal microorganisms, obtaining microecological flora composition data, and classifying using an infant microecological maturity judgment model to determine whether the microecological flora is mature; a microecological age prediction module for predicting microecological age using a machine learning regression model when the microecology is not mature, and obtaining a predicted microecological age; and a risk assessment module for performing a risk assessment on whether the development of infants is advanced or delayed according to the predicted microecological age and the microecological age development curve. The infant microecological flora development assessment system of this embodiment and the infant microecological flora development assessment method of the aforementioned embodiment are derived from the same inventive concept, and the specific implementation of each module is referred to the above method embodiment.

[0086] Figure 6The evaluation process of the infant microecological flora development evaluation system is illustrated. First, based on the sequencing information of infant feces, the microecological maturity judgment model is used to judge whether the microecological development is mature. Combined with the actual age of the infant and the developmental maturity judgment age (27 months old) determined in the present invention, there are the following 4 situations: 1. The actual age is ≥ 27 months, and the microecological development is mature; the microecological age is considered to be normally developed and mature; 2. The actual age is ≥ 27 months, and the microecological development is immature; the microecological age is considered to be normally developed and delayed; 3. The actual age is < 27 months, and the microecological development is mature; the microecological age is considered to be normally developed too early, combined with the questionnaire judgment; 4. The actual age is < 27 months, and the microecological development is immature; this situation is the situation focused on by the present invention. The microecological age is predicted by the regression model, and it is judged whether the microecological age is in the reference interval. If so, the microecological age is considered to be normally developed, otherwise, the microecological age is considered to be abnormal, and measures such as dietary nutrition intervention are required.

[0087] Example 3

[0088] The embodiment of the present invention discloses a method for evaluating the development of infant microecological flora, which, based on embodiment 1, further includes: performing correlation analysis between allergies and whether the microecological age development is abnormal, involving data collection, data processing, microecological age prediction, and correlation between abnormal development and allergies. The specific steps include:

[0089] Step S401, data collection: 234 fresh feces of infants and young children (mostly under 3 years old) were collected, DNA was extracted and amplified, and then the DNA was sequenced on a machine, and the phenotypic information of whether each sample corresponds to the individual's allergy was collected.

[0090] Step S402, data processing: Process the data from the sequencing machine based on the 16S amplicon analysis process of DADA2 (https: / / benjjneb.github.io / dada2 / tutorial.html). The operation steps are as follows:

[0091] (a) First, the downloaded sequence is removed from the machine and split to obtain the 16S rRNA sequence of the sample. Then, after quality filtering, noise removal, merging, and chimera removal, a 100% consistent ASV (Amplicon Sequence Variant) sequence is obtained.

[0092] (b) The ASV sequence was aligned to the Silva database (v138) to obtain the species name of the sequence.

[0093] (c) Integrate the names of different reference sequences, merge the abundance of sequences annotated to the same species into the abundance of the same species, and obtain the bacterial community composition data of the sample;

[0094] Step S403, judging whether the microecological age is mature: using the microecological maturity judgment model to predict whether the sample is mature.

[0095] Step S404, microecological age prediction: Because the sample ages are still relatively young, most of them are judged as immature by the maturity judgment model. For the samples judged as immature in the previous step, the microecological age is predicted using a machine learning regression model to obtain specific microecological development age numbers.

[0096] Step S405: correlation between allergy and abnormal microecological age development: statistics on the correlation between the occurrence of allergic diseases and abnormal microecological age (normal between Q10 and Q90, abnormal outside the range) in these 234 samples. The results are shown in Table 2. Among normal children, the rate of abnormal microecological age development is 14.29%, while the rate of abnormal microecological age development in allergic children is as high as 42.14%. Figure 7 shown.

[0097] Table 2. Correlation between microecological age assessment results and allergic phenotype

[0098]

[0099] This embodiment can prove that after microecological age assessment, the microecological age of allergic children is more likely to deviate from the reference range compared with normal children, providing a new idea for early prevention and treatment of allergic diseases that are highly prevalent in infants and young children in the future.

[0100] Example 4

[0101] The embodiment of the present invention discloses an infant microecological flora development assessment system, which, based on Example 2, further includes an allergy association analysis module for associating allergies with whether the microecological age development is abnormal, and for associating allergies with whether the microecological age development is abnormal. The detailed implementation steps of this module can be found in Example 3.

[0102] Example 5

[0103] An embodiment of the present invention discloses a computer system, including a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor, wherein the computer program / instruction implements the steps of the method in the aforementioned method embodiment when executed by the processor.

[0104] Example 6

[0105] An embodiment of the present invention discloses a computer program product, including a computer program / instruction, which implements the steps of the method in the aforementioned method embodiment when the computer program / instruction is executed by a processor.

Claims

1. A method for evaluating the development of microecological flora in infants and young children, characterized in that: The following steps are involved: Extract and sequence fecal microorganisms of infants and young children to obtain microecological flora composition data, and classify them using a microecological maturity judgment model for infants and young children to determine whether the microecological flora is mature; the microecological maturity judgment model for infants and young children is constructed based on the best classification age obtained by statistics; If it is not mature, the microecological age is predicted using a machine learning regression model to obtain the predicted microecological age; According to the predicted microecological age and the microecological age development curve, the risk assessment of whether the infants and young children are ahead of or behind in development is carried out; The microecological age development curve is predicted on a large-scale data set according to a machine learning regression model and a reference range for development judgment standards is set.

2. A method for evaluating the development of infant microecological flora according to claim 1, characterized in that: The method for constructing the infant microecological maturity judgment model comprises the following steps: Collect and organize the original sequencing data of infant intestinal microorganisms and their clinical information, analyze and integrate the sequencing data, and obtain the intestinal microbial relative abundance information table for constructing the microecological development maturity classification model; All samples were divided into two categories: immature and mature, with the actual age of the samples ranging from 18 to 36 months as the classification age; For multiple different classification result data, feature selection is performed, and a microecological development maturity classification model is constructed, and the accuracy of classification prediction results is statistically analyzed; Compare the accuracy of different classification moon age prediction results and determine the best classification moon age; Using the optimal classification age as the dividing line, a machine learning method was used to construct a model for judging the maturity of infant microecology.

3. A method for evaluating the development of infant microecological flora according to claim 2, characterized in that: The optimal classification age is 27 months.

4. A method for evaluating the development of infant microecological flora according to claim 1, characterized in that: The method for predicting the microecological age of infants and young children using a machine learning regression model includes the following steps: The intestinal microbial composition information of each individual is obtained through sequencing, and the relative abundance of each bacterial community is calculated; The Boruta method was used to screen key biomarkers for evaluating the developmental maturity of an individual's intestinal microecology; Based on the key biomarker set, an individual intestinal microecological age prediction model is iteratively optimized and verified, and the optimal preset number of multiple models are selected for subsequent use; The optimal multiple models were used to predict the intestinal microecological age of the test samples, and the average value after removing the highest and lowest values ​​was used as the final individual evaluation result.

5. A method for evaluating the development of infant microecological flora according to claim 1, characterized in that: The method for generating the microecological age development curve and calculating the reasonable reference interval of microecological development age at different months of age comprises the following steps: Collect infant microecological flora data, including raw sequencing data and phenotypic information, and construct a reference data set; Analyze the collected raw data to obtain the bacterial community composition and structure data of each sample; The microecological age of each sample was calculated using a machine learning regression model; A scatter plot was constructed based on the actual age and microecological age of each sample in the reference data set. The microecological age development curve was fitted based on the median of the microecological age corresponding to each month of age. The lower limit of the microecological development age was fitted according to the preset upper limit quantile, and the upper limit of the microecological development age was fitted according to the preset lower limit quantile.

6. A method for evaluating the development of infant microecological flora according to claim 1, characterized in that: It also includes an analysis of the association between allergies and abnormal microecological age development, including the following steps: Collect fecal samples from infants and young children to obtain microecological flora data, and collect phenotypic information on whether each sample corresponds to an individual with allergies; The infant microecological maturity judgment model is used to determine whether the microecological age is mature. For immature samples, the machine learning regression model is used to predict the microecological age, and the microecological age development curve is combined to determine whether the microecological age development is abnormal; An association analysis was conducted between the occurrence of allergic diseases and abnormal microecological age development.

7. A system for evaluating the development of microecological flora in infants and young children, characterized in that: include: A microecological maturity judgment module is used to extract and sequence fecal microorganisms of infants and young children to obtain microecological flora composition data, and classify them using an infant microecological maturity judgment model to judge whether the microecological flora is mature; the infant microecological maturity judgment model is constructed based on the best classification age obtained by statistics; The microecological age prediction module is used to predict the microecological age using a machine learning regression model when the microecological system is not mature, and obtain the predicted microecological age; And a risk assessment module, which is used to conduct risk assessment on whether the development of infants and young children is ahead of schedule or behind schedule based on the predicted microecological age and the microecological age development curve; The microecological age development curve is predicted on a large-scale data set according to a machine learning regression model and a reference range for development judgment standards is set.

8. The infant microecological flora development assessment system according to claim 7, characterized in that: It also includes an allergy association analysis module, which is used to conduct association analysis between allergies and whether the microecological age development is abnormal.

9. A computer system comprising a memory, a processor, and a computer program / instruction stored in the memory and executable on the processor, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images