Information processing device, information processing method, and information processing program

Abstract

To enable the evaluation of individual differences in intestinal bacterial flora.SOLUTION: An information processing device includes: a memory unit that stores specimen data on intestinal bacterial flora of a population; a clustering unit that classifies the specimen data into multiple clusters based on information on the intestinal bacterial flora in the specimen data; an evaluation unit that applies a predetermined dimensionality reduction method to clustering results obtained from the clustering unit and evaluates vectors between clusters that may shift due to intrinsic factors; and an output unit that outputs cluster evaluation results obtained from the evaluation unit in a predetermined format.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] There are technologies related to the analysis and proposal of intestinal flora. For example, as an example of a technology for understanding a person's health condition, there is a technology for generating clusters based on data on human resident bacteria (see Patent Document 1).

[0003] There is also a technology for calculating an intestinal flora score using multiple indices related to intestinal bacteria (see Patent Document 2). This technology shows a method for assessing and scoring intestinal dysbiosis, an abnormal state of the intestinal environment, using at least one index selected from a population group for setting judgment reference values ​​consisting of an index related to intestinal bacterial diversity, an index related to short-chain fatty acid production, an index related to intestinal immunity, an index related to oral bacteria, and an index related to diarrhea and constipation.

[0004] There is also a technology relating to a method for estimating disease risk from information on the intestinal microbiota (see Patent Document 3). This technology includes a selection process and a classification process for selecting an evaluation group from evaluation group groups consisting of 5 types, 10 types, and 15 types. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2019-041625 [Patent Document 2] Japanese Patent Publication No. 2020-078273 [Patent Document 3] Japanese Patent Application Publication No. 2023-004411 Summary of the Invention [Problem to be solved by the invention]

[0006] Subjects who undergo an intestinal microbiota test can receive suggestions on how to increase specific bacteria (such as lactic acid bacteria, butyric acid bacteria, and acetic acid bacteria). However, it is not possible to receive optimal suggestions tailored to individual differences in the intestinal environment. Meanwhile, conventional techniques have been put into practical use to qualitatively classify individual differences in intestinal microbiota (Patent Document 3) and quantitatively score and evaluate them (Patent Document 2). However, no suggestions have been made that capture individual differences in intestinal microbiota or changes in the continuity of intestinal microbiota that may change over time, such as on a yearly basis.

[0007] An object of the present disclosure is to provide an information processing device, an information processing method, and an information processing program that can evaluate changes in individual differences in intestinal bacterial flora. [Means for solving the problem]

[0008] The information processing device of the present disclosure includes a memory unit that stores specimen data of the intestinal microbiota of a population; a clustering unit that classifies the specimen data into multiple clusters based on the intestinal microbiota information of the specimen data; an evaluation unit that applies a predetermined dimensionality reduction method to the clustering results obtained from the clustering unit and evaluates vectors between clusters that may move due to intrinsic factors; and an output unit that outputs the evaluation results of the clusters obtained from the evaluation unit in a predetermined format. [Effects of the Invention]

[0009] The information processing device, information processing method, and information processing program disclosed herein have the effect of enabling evaluation of changes in individual differences in the intestinal bacterial flora. [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a block diagram showing the hardware configuration of an information processing device. [Figure 2] FIG. 2 is a block diagram showing the configuration of the information processing apparatus according to the first embodiment. [Figure 3] FIG. 3 is a diagram showing an example of enterotypes of intestinal microbiota by clustering. [Figure 4] FIG. 4 shows an example of the bacterial composition of type B. [Figure 5] Figure 5 shows an example of the bacterial composition of the R type. [Figure 6] FIG. 6 shows an example of the bacterial composition of P type. [Figure 7] FIG. 7 shows an example of a three-dimensional spatial distribution in which specimen data for each cluster is plotted. [Figure 8] FIG. 8 shows an example of a three-dimensional spatial distribution in which specimen data for each cluster is plotted. [Figure 9] FIG. 9 is an example of a comparison diagram showing an example of evaluation of continuity to subgroups based on the evaluation results. [Figure 10] FIG. 10 is a flowchart showing the flow of information processing by the information processing apparatus of the first embodiment. [Figure 11] FIG. 11 is a diagram showing the configuration of the second embodiment. [Figure 12] FIG. 12 is a verification example showing transition probabilities between subgroups. [Figure 13] FIG. 13 is a verification example showing transition probabilities between subgroups. [Figure 14] FIG. 14 is a verification example showing transition probabilities between subgroups. [Figure 15] FIG. 15 is a flowchart showing the flow of information processing by the information processing apparatus of the second embodiment. [Figure 16] FIG. 16 is a block diagram showing the configuration of an information processing apparatus according to the third embodiment. [Figure 17] FIG. 17 shows an example of comparing the presence or absence of disease among subgroups for each main group. [Figure 18] Figure 18 shows an example of a comparison of dietary habits between subgroups for each main group. [Figure 19] Figure 19 shows an example of a comparison of dietary habits between subgroups for each main group. [Figure 20] FIG. 20 is a flowchart showing the flow of information processing by the information processing device of the third embodiment. [Figure 21] FIG. 21 shows an example of outputting a space in which the positional structure between groups is expressed in coordinate format. DETAILED DESCRIPTION OF THE INVENTION

[0011] An example of an embodiment of the disclosed technology will be described below with reference to the drawings. Note that the same or equivalent components and parts in each drawing are given the same reference numerals. Also, the dimensional proportions in the drawings are exaggerated for the convenience of explanation and may differ from the actual proportions.

[0012] An overview of an embodiment of the present disclosure will be described. Regarding intestinal flora groups (also called enterotypes or clusters), evaluation of the continuity / discontinuity between groups and the transition probability have not been performed to date. Furthermore, unless both the transition probability between groups and the association with lifestyle habits for each subtype can be shown, it is difficult to propose improvement measures appropriate for the subject (user). It is possible to understand where the subject currently belongs among a group of closely related subtype structures and which subtypes are likely or unlikely to change to as a result of improvement or deterioration. Furthermore, it is possible to not only compare and evaluate the degree of dysbiosis and disease risk between types, but also to evaluate the continuity and transition probability of types.

[0013] In the following, a first embodiment will describe a method for evaluating the continuity of individual differences in intestinal microbiota. A second embodiment will describe a method for calculating the transition probability of individual differences in intestinal microbiota. A third embodiment will describe a method for outputting health management information from the results of cluster comparison.

[0014] The continuity of individual differences in gut microbiota is a property that can be obtained by evaluating vectors between clusters that can shift due to endogenous factors. In evaluating vectors between clusters, the gut microbiota of the same individual has the property of changing over time. In addition, the gut microbiota of different individuals has the property of being continuous in terms of series information (time, weight, etc.). The transition probability of individual differences in gut microbiota indicates the probability that the characteristics of the gut microbiota will change with changes in series information, and that this probability has different properties depending on the characteristics of the gut microbiota. The output of health management information contributes to realizing health management and diet support tailored to the characteristics of the user's gut microbiota.

[0015] In addition, a database that electromagnetically stores intestinal flora data is constructed in the memory unit. The intestinal flora specimen data is linked to a database of user attributes (member ID, test date and time, height, weight, sex, age, etc.) and a database of user lifestyle information.

[0016] [First embodiment] 1 is a block diagram showing the hardware configuration of an information processing device 100. Note that the information processing device 200 of the second embodiment and the information processing device 300 of the third embodiment may also have a similar hardware configuration.

[0017] 1, the information processing device 100 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, a storage 14, an input unit 15, a display unit 16, and a communication interface (I / F) 17. Each component is connected to each other via a bus 19 so as to be able to communicate with each other.

[0018] The CPU 11 is a central processing unit that executes various programs and controls each part. That is, the CPU 11 reads a program from the ROM 12 or the storage 14 and executes the program using the RAM 13 as a work area. The CPU 11 controls each of the above components and performs various arithmetic processing in accordance with the program stored in the ROM 12 or the storage 14. In this embodiment, a prediction program is stored in the ROM 12 or the storage 14.

[0019] The ROM 12 stores various programs and various data. The RAM 13 temporarily stores programs or data as a working area. The storage 14 is configured with a storage device such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) and stores various programs including the operating system and various data.

[0020] The input unit 15 includes a pointing device such as a mouse and a keyboard, and is used to perform various inputs. The input unit 15 may also include a device capable of inputting audio, such as a microphone. For example, input information, which will be described later, or a part of the information may be input as audio.

[0021] Display unit 16 is, for example, a liquid crystal display, and displays various types of information. Display unit 16 may employ a touch panel system and function as input unit 15. Display unit 16 may also be capable of displaying text input by voice on the liquid crystal display, or may include a device capable of outputting voice.

[0022] The communication interface 17 is an interface for communicating with other devices such as terminals, etc. For this communication, for example, a wired communication standard such as Ethernet (registered trademark) or FDDI, or a wireless communication standard such as 4G, 5G, or Wi-Fi (registered trademark) is used.

[0023] The following describes each functional configuration of the information processing device 100. Fig. 2 is a block diagram showing the configuration of the information processing device 100 according to the first embodiment. Each functional configuration is realized by the CPU 11 reading out an information processing program stored in the ROM 12 or storage 14, expanding the program in the RAM 13, and executing the program.

[0024] As shown in FIG. 2, the information processing device 100 functionally comprises a storage unit 102, a clustering unit 110, an evaluation unit 112, and an output unit 114.

[0025] The storage unit 102 stores specimen data of the intestinal flora of the population. The storage unit 102 also stores information processed by each unit of the information processing device 100.

[0026] The clustering unit 110 classifies the specimen data stored in the memory unit 102 into multiple clusters based on the information on the intestinal microflora, and stores the clustering results in the memory unit 102. For example, the clustering method described in Patent Document 1 can be used, which can classify enterotypes into multiple clusters based on the bacterial carriage pattern. The number of classifications can also be specified and output for the clustering. For the clustering method, non-hierarchical clustering, k-means, etc. may also be used.

[0027] As an example of the processing of the clustering unit 110, a clustering result when approximately 50,000 specimen data of people aged 16 or over are used as training data will be described.

[0028] Figure 3 shows an example of enterotypes of gut microbiota determined by clustering. Dirichlet mixture clustering was used to perform primary classification into three main enterotypes: B, R, and P. The enterotypes were B: predominantly Bacteroides, R: predominantly Ruminococcus, and P: predominantly Prevotella. Secondary classification was then performed to obtain sub-subgroups from each main group. Subgroups were identified: B1 and B2 as subgroups of B, R1 and R2 as subgroups of R, and P1 and P2 as sub-subgroups of P. Comparison of the alpha diversity (Shannon index) of these six subtypes revealed differences among all main types, with particularly significant differences between B and P. The enterotypes of the clusters are not limited to these six types, and four types (for example, R, P, B1, and B2) or seven types (for example, R1, R2, P1, P2, B1-1, B1-2, and B2) may be output. As described above, the clustering results of the intestinal microbiota disclosed herein are classified into main groups, which are primary classifications representing enterotypes, and subgroup clusters, which are secondary classifications for the enterotypes of the main groups.

[0029] Subgroups are characterized by the abundance of bacterial species with a moderate or lower detection frequency. Figure 4 shows an example of the bacterial composition of type B. Figure 5 shows an example of the bacterial composition of type R. Figure 6 shows an example of the bacterial composition of type P. Regarding type B, the classification of types B1 and B2 is determined by the difference in the frequency distribution of bacterial genera such as Faecalibacterium, Fusobacterium, Veillonella, or Escherichia-Shigella. Regarding type R, the classification of types R1 and R2 is determined by the difference in the frequency distribution of bacterial genera such as Ruminococcus and Bifidobacterium. Regarding type P, the classification of types P1 and P2 is determined by the difference in the frequency distribution of bacterial genera such as Ruminococcus, Prevotella, or Megamonas.

[0030] The evaluation unit 112 applies a dimensionality reduction method to the clustering results obtained from the clustering unit 110 to evaluate vectors between clusters that may move due to intrinsic factors. In evaluating the vectors, calculations are performed while distinguishing between the main group and the subgroups. In the case of performing calculations while distinguishing between the main group and the subgroups, the evaluation unit 112 generates a distribution by plotting the sample data belonging to the main group and the subgroups in a three-dimensional space for the evaluation results obtained by applying a dimensionality reduction method to the sample data of the intestinal microbiota of the group.

[0031] The output unit 114 outputs the cluster evaluation results for each user in a predetermined format. The output format can be a space in which the positional structure between groups is expressed in coordinate format. Alternatively, the groups can be divided into groups in a map format, with the location of the cluster and an arrow pointing to a healthier area displayed. Figure 21 shows an example of outputting a space in which the positional structure between groups is expressed in coordinate format. According to Figure 21, the map area is divided into a risky area, an average area, and a good area based on the cluster, etc. The current location (B1) indicates the user's current group, which belongs to a group classified as a risky area. By adjusting the user's diet, etc., the user can move from the current location (B1) group. By improving their intestinal environment, they can move to a group belonging to, for example, the average area group. In this example, the arrow indicates the group (202) belonging to the highly recommended average area, which has the effect of encouraging the user's awareness of improving their intestinal environment and the intestinal bacteria they need. On the other hand, even if a person is unable to move to the recommended group indicated by the arrow, he or she may be able to move to the group (201, 203) in the average area by improving his or her diet and lifestyle. For example, as shown in FIG. 21, when the groups are connected by a line, this indicates that a move is possible. In this way, the output unit 114 can output, in a coordinate-format representation of a space (map), the groups connected to the current group and belonging to a good area as a destination. Furthermore, the output unit 114 can output a graph representing the relationship between the recommended group and the group in a good area to which a person may move, among the groups belonging to the good area. The representation format of this map may be a two-dimensional map, a three-dimensional map, or a roadmap. The cluster evaluation map can be divided into multiple areas, for example, into B2 type, B1-2 type, and B1-1 type, and each division can be further subdivided into multiple levels.

[0032] For example, the method described in Reference 1 can be used as a dimensionality reduction method. In Reference 1, samples are evaluated for each diversity index using a PHATE scatter plot. The vector components of the diversity index are obtained as evaluations and can be used as scales for the axes of a three-dimensional space (Figures 7 and 8). [Reference 1] Tap, J., Lejzerowicz, F., Cotillard, A. et al. Global branches and local states of the human gut microbiome define associations with environmental and intrinsic factors. Nat Commun 14, 3310 (2023).

[0033] The axis of the diversity index that can be calculated with reference to the above-mentioned Reference 1 can be calculated by distinguishing between main groups and subgroups. Bacteria that contribute to the main group axis include the genera Bacteroides, Prevotella, and Ruminococcus. Bacteria that contribute to the subgroup axis include Fusobacterium, Megamonas, Esherichia, Shigella, Alistipes, Faecalibacterium, Bifidobacterium, and Blautia.

[0034] Figures 7 and 8 show examples of three-dimensional spatial distributions in which specimen data for each cluster is plotted. Although each main group is separated along the first axis, they are not completely independent, and some specimens were observed moving between main groups along the first axis. For type B, types B1 and B2 were separated along the second and third axes, with type B1 in an intermediate position on each axis and type B2 in a biased position. For type R, no significant separation structure was observed, but a structure in which type R2 was concentrated in the center and lined up on the periphery was observed. For type P, separation was observed only along the second axis.

[0035] Figure 9 is an example of a comparison diagram showing an evaluation example of continuity to subgroups based on the evaluation results. The comparison diagram shows an example where continuity between 4 types and 7 types has been manually identified and grouped based on the overlap of each group in the evaluation results. In some cases, there is a transition to the same type, but in other cases, different continuity is shown due to individual differences, such as B1 overlapping with R2, or B2 overlapping with B1-2.

[0036] By comparing the evaluations between these clusters with user information, it becomes possible to explain the continuity of individual differences, such as the nature of changes over time and the continuous nature of sequential information, and to make suggestions to users that take into account the enterotypes that correspond to the continuity.

[0037] (Processing flow) Next, the operation of the information processing device 100 will be described. Fig. 10 is a flowchart showing the flow of information processing by the information processing device 100 of the first embodiment. The CPU 11 reads out an information processing program from the ROM 12 or storage 14, expands it in the RAM 13, and executes it, thereby performing information processing as an information processing method. The CPU 11 functions as each part of the information processing device 100, causing the following processing to be executed.

[0038] In step S100, the CPU 11 acquires specimen data of the intestinal flora of the population stored in the storage unit 102.

[0039] In step S102, the CPU 11 classifies the specimen data into a plurality of clusters based on the information on the intestinal flora of the specimen data, and stores the clustering results in the storage unit 102.

[0040] In step S104, the CPU 11 applies a dimension reduction technique to the clustering result to evaluate vectors between clusters that may move due to intrinsic factors.

[0041] In step S106, the CPU 11 generates a distribution of the evaluation results by plotting the specimen data belonging to the main group and the subgroups in a three-dimensional space.

[0042] In step S108, the CPU 11 outputs a three-dimensional spatial distribution that visualizes the positional structure between groups as a result of the evaluation.

[0043] As described above, the information processing device 100 according to this embodiment can evaluate the continuity between groups regarding changes in individual differences in the intestinal microflora.

[0044] [Second embodiment] The second embodiment is an aspect that includes the calculation of transition probabilities of individual differences in intestinal bacterial flora. Fig. 11 is a diagram showing the configuration of the second embodiment. An information processing device 200 of the second embodiment has a configuration in which a transition calculation unit 210 is added to the configuration of the information processing device 100 of the first embodiment. Note that similar parts are given the same reference numerals and descriptions thereof will be omitted.

[0045] The transition calculation unit 210 calculates the transition probability between clusters using the evaluation results and the attribute information of the specimen data. Here, the transition probability of a subgroup is calculated, and the calculation of the transition probability involves calculating, for each subgroup, the transition probability of a subgroup transitioning to another subgroup using specimen data collected multiple times in a time series. It is assumed that clustering results are created for each time series of specimen data. For each subgroup, the transition probability of the subgroup may be calculated based on whether or not each user of the specimen data transitioned to another subgroup between the previous and next examination dates and times using at least the user's membership ID, examination date and time, height, and weight information from the specimen data.

[0046] The output unit 114 of the second embodiment outputs the transition probability for each subgroup calculated by the transition calculation unit 210 along with the evaluation result (a space in which the positional structure between groups is expressed in coordinate format). The output unit 114 has a function of outputting a notification of advice regarding the direction of movement according to the transition probability. The output unit 114 also has a function of changing the thickness of the arrow according to the transition probability (see FIG. 21). For example, when the belonging groups are connected by a line as shown in FIG. 21, the thickness of the line may indicate the transition probability. According to FIG. 21, the current location (B1) and the belonging group (202) are connected by a thick arrow, but the belonging group (201) is connected by a line. Therefore, it can be indicated that the user can move to a belonging group in an area with a low health risk not only by transitioning to the belonging group (202) connected by a thick arrow, but also by transitioning to the belonging group (201) connected by a line.

[0047] Figures 12–14 show examples of transition probabilities between subgroups. Using the sample dataset of multiple test subjects (n=4210) as the population, we examined the migration rate for each enterotype. First, 60–70% of individuals remained in each subgroup. Furthermore, within each main group, individuals in better states (B1, R1, P1) remained in the same state more often than individuals in worse states (B1, R1, P1). Within the main group, migration from B2 to B1 was the most prominent (30%), followed by migration from P2 to P1 (20%). However, migration between subgroups does not necessarily remain within the main group. As can be seen from Figures 7 and 8 (three-dimensional space), significant migration was also observed between B / P and R types (mainly between B1 and R2 and between P1 and R1).

[0048] The operation of the information processing device 200 of the second embodiment will be described below. Fig. 15 is a flowchart showing the flow of information processing by the information processing device 200 of the second embodiment. In the second embodiment, step S200 is executed after step S106.

[0049] In step S200, the CPU 11 calculates, for each subgroup, a transition probability of the subgroup transitioning to another subgroup, using sample data collected multiple times in time series.

[0050] In step S202, the CPU 11 outputs the transition probability for each subgroup calculated by the transition calculation unit 210 together with the evaluation result (distribution in three-dimensional space) of the evaluation unit 112.

[0051] As described above, the information processing device 200 of the second embodiment can output transition probabilities between groups regarding changes in individual differences in intestinal bacterial flora.

[0052] [Third embodiment] The third embodiment is an aspect in which health management information is output from the cluster comparison results. FIG. 16 is a block diagram showing the configuration of an information processing device of the third embodiment. The information processing device 300 of the third embodiment has a configuration in which a discrimination unit 310 is added to the configuration of the information processing device 200 of the second embodiment. It may also be an aspect in which a discrimination unit 310 is added to the information processing device 100 of the first embodiment (an aspect in which the transition calculation unit 210 is not included). Note that similar parts are given the same reference numerals and descriptions thereof will be omitted.

[0053] The storage unit 102 of the third embodiment stores disease information and health habit information corresponding to specimen data, and the degree of association between the clusters. The disease information includes, for example, hyperuricemia, diabetes, heart disease, liver disease, chronic kidney disease, IBS, ulcerative colitis, colon polyps, etc. The health habit information includes questionnaire information on diet, smoking, and sleep. It also includes physical information such as the user's height and weight (BMI). The storage unit 102 also stores health management information for each cluster obtained by processing in the discrimination unit 310. The health management information is information that serves as support advice linked to the user's enterotype and other detailed information (intestinal flora, habit data). The linking is performed by associating information of similar users stored in the original database, such as the success results of other users with typical enterotypes.

[0054] The discrimination unit 310 discriminates the comparison result of the health management of the sub-group against the main group using the degree of association, and stores health management information according to the comparison result for each cluster in association with the cluster (group) in the storage unit 102. An example of the health management comparison is the odds ratio.

[0055] In the third embodiment, the output unit 114 accepts the designation of a user whose sample data is stored in the storage unit 102, acquires health management information for the cluster to which the user's sample data belongs, and outputs the health management information. The output unit 114 has a function for outputting recommended intestinal health recipes and advice based on the user's cluster and the clusters to which the user can move. The output unit 114 also has a function for changing the display of the health management information based on the input of the user's attribute information and lifestyle information (body type, age, gender, medical history, etc.). An example of the display of the health management information is the display in the display area C1 in FIG. 21. Furthermore, when the input information is changed, the display of the bacterial flora map at the top, the current location, and the health management information change to content specific to the selected input information. FIG. 21 shows that the user's attribute information, such as "No Setting," "Diet," "Immunity," "Health," and "Beauty," can be selected as items for changing the input information. Similarly, lifestyle information can be input by selecting it.

[0056] The comparison of health management by the discriminator 310 will be exemplified below. The odds ratio of a disease and the odds ratio of a habit will be given as examples.

[0057] Figure 17 shows an example of a comparison of the presence or absence of disease between subgroups for each main group. Using types B1, R1, and P1 as the reference, odds ratios were calculated after excluding invalid responses for each disease. For type B, the odds ratios were particularly high for diabetes (odds ratio 1.73, confidence interval 1.45-2.06) and ulcerative colitis (odds ratio 1.6, confidence interval 1.10-2.31). For type R, the odds ratios were particularly high for chronic kidney disease (odds ratio 1.91, confidence interval 0.98-3.88) and ulcerative colitis (odds ratio 1.7, confidence interval 0.73-4.42). For P type, the odds ratios for liver disease (odds ratio 2.09, confidence interval 1.36-3.25), hyperuricemia (odds ratio 2.45, confidence interval 1.88-3.22), diabetes (odds ratio 1.78, confidence interval 1.39-2.29), chronic kidney disease (odds ratio 1.61, confidence interval 0.64-4.05), IBS (odds ratio 2.4, confidence interval 1.63-3.58), colon polyps (odds ratio 1.71, confidence interval 1.06-2.78), and ulcerative colitis (odds ratio 6.86, confidence interval 2.24-29.80) were higher.

[0058] Figures 18 and 19 show examples of dietary habits compared between subgroups for each main group. Using Type B1, Type R1, and Type P1 as the reference, we calculated the odds ratio for the other subgroup for each lifestyle habit item, which was binary-labeled as applicable / not applicable. Dietary habits were defined based on the weekly intake frequency reported in the questionnaire, with two patterns: (1) intake four or more times per week, and (2) intake less than once per week. While no significant differences were observed for Type B, Type B2 tended to consume less green and yellow vegetables (odds ratio 1.15, confidence interval 1.01-1.30) and mushrooms (odds ratio 1.14, confidence interval 1.04-1.23) compared to Type B1. Regarding R-type, compared with R1-type, R2-type individuals were more likely to consume sweetened beverages (odds ratio 1.25, confidence interval 1.17-1.33) frequently, but less likely to consume root vegetables (odds ratio 1.32, confidence interval 1.16-1.50), green and yellow vegetables (odds ratio 1.46, confidence interval 1.27-1.67), or fruits (odds ratio 1.49, confidence interval 1.37-1.62). With regard to P types, compared to P1 types, P2 types were particularly more likely to consume sweetened beverages (odds ratio 1.21, confidence interval 1.09-1.34), but less likely to consume light-colored vegetables (odds ratio 1.36, confidence interval 1.02-1.82), green and yellow vegetables (odds ratio 1.4, confidence interval 1.15-1.70), natto (odds ratio 1.3, confidence interval 1.19-1.43), or fruit (odds ratio 1.44, confidence interval 1.28-1.62). The above-mentioned information on insufficient intake was associated with each cluster as insufficient health management information.

[0059] The operation of the information processing device 300 of the third embodiment will be described below. Fig. 20 is a flowchart showing the flow of information processing by the information processing device 300 of the third embodiment. In the third embodiment, step S300 is executed after step S202.

[0060] In step S300, the CPU 11 uses the degree of association to determine the comparison result of the health management of the sub-group with the main group.

[0061] In step S302, CPU 11 stores health management information corresponding to the comparison result for each cluster in association with the cluster. Note that the processing up to step S302 may be performed in advance, and the processing of step S304 may be executed at any appropriate timing.

[0062] In step S304, CPU 11 accepts the designation of a user whose specimen data is available in storage unit 102, acquires health management information of the cluster to which the specimen data of that user belongs, and outputs the health management information.

[0063] As described above, the information processing device 300 of the third embodiment can propose optimal health management according to the cluster to which the user's intestinal flora belongs.

[0064] The technology of this embodiment is not limited to the above-described embodiment, and various modifications and applications are possible within the scope of the gist of this embodiment.

[0065] In the above embodiments, the information processing performed by the CPU after reading the software (program) may be performed by various processors other than the CPU. Examples of such processors include programmable logic devices (PLDs) (such as field-programmable gate arrays (FPGAs)) whose circuit configuration can be changed after fabrication, graphics processing units (GPUs), and application-specific integrated circuits (ASICs) that are dedicated electrical circuits that are processors with circuit configurations specifically designed to perform specific processes. Information processing may be performed by one of these various processors, or by a combination of two or more processors of the same or different types (e.g., multiple FPGAs, multiple GPUs, or a combination of a CPU and an FPGA). The hardware structure of these various processors is, more specifically, an electrical circuit that combines circuit elements such as semiconductor devices.

[0066] In the above embodiment, the information processing program is pre-stored (installed) in storage, but the present invention is not limited to this. The program may be provided in a form stored in a non-transitory storage medium such as a CD-ROM (Compact Disk Read Only Memory), a DVD-ROM (Digital Versatile Disk Read Only Memory), or a USB (Universal Serial Bus) memory. The program may also be downloaded from an external device via a network. [Explanation of symbols]

[0067] 100, 200, 300 Information processing equipment 102 Storage section 110 Clustering Department 112 Evaluation Department 114 Output section 210 Transition calculation section 310 Discrimination part

Claims

1. a memory unit storing sample data of the group's gut microbiota; a clustering unit that classifies the specimen data into a plurality of clusters based on information on the intestinal flora of the specimen data; an evaluation unit that applies a predetermined dimensionality reduction method to the clustering result obtained from the clustering unit and evaluates vectors between clusters that may move due to intrinsic factors; an output unit that outputs the cluster evaluation results obtained from the evaluation unit in a predetermined format; An information processing device comprising:

2. The clustering results of the intestinal microbiota are classified into main groups, which are primary classifications representing enterotypes, and subgroup clusters, which are secondary classifications of the enterotypes of the main groups; the evaluation unit performs calculations by distinguishing between the main group and the subgroup in evaluating the vectors. The information processing device according to claim 1 .

3. When performing the distinguishing calculation, the evaluation unit generates a distribution in which the specimen data belonging to the main group and the subgroup are plotted in a three-dimensional space for the evaluation results obtained by applying a dimensionality reduction method to the specimen data of the intestinal bacterial flora of the population; the output unit outputs a three-dimensional spatial distribution that visualizes the positional structure between the groups. The information processing device according to claim 2 .

4. further comprising a transition calculation unit; The attribute information of the specimen data includes a member ID, a test date and time, a height, and a weight, the transition calculation unit calculates a transition probability between clusters using the evaluation result and the attribute information of the specimen data. The information processing device according to claim 1 .

5. The clustering results of the intestinal microbiota are classified into main groups, which are primary classifications representing enterotypes, and subgroup clusters, which are secondary classifications of the enterotypes of the main groups; the transition calculation unit calculates the transition probability of the subgroup in calculating the transition probability; The information processing device according to claim 4 .

6. the transition calculation unit calculates, for each of the subgroups, a transition probability of a subgroup transitioning to another subgroup, using specimen data collected multiple times in a time series as the specimen data, in calculating the transition probability; the output unit outputs the transition probability for each of the subgroups. The information processing device according to claim 5 .

7. Further including a discrimination unit, the storage unit stores disease information and health habit information corresponding to the specimen data and the degrees of association of the clusters; the determining unit determines a comparison result of the health management of the sub-group with the main group using the degree of association, and stores health management information corresponding to the comparison result for each cluster in the storage unit; the output unit receives designation of a user from whom the specimen data has been obtained, acquires the health management information of a cluster to which the specimen data of the user belongs from the storage unit, and outputs the health management information. The information processing device according to claim 2 .

8. Classifying the sample data of the intestinal flora of the population into a plurality of clusters based on the intestinal flora information of the sample data; A predetermined dimension reduction method is applied to the clustering results obtained by the clustering, and vectors between clusters that may move due to endogenous factors are evaluated; outputting the cluster evaluation results obtained by the evaluation in a predetermined format; An information processing method in which processing is performed by a computer.

9. Classifying the sample data of the intestinal flora of the population into a plurality of clusters based on the intestinal flora information of the sample data; A predetermined dimension reduction method is applied to the clustering results obtained by the clustering, and vectors between clusters that may move due to endogenous factors are evaluated; outputting the cluster evaluation results obtained by the evaluation in a predetermined format; An information processing program that causes a computer to execute a process.

Images