Systems and methods for estimating the relative abundance of Faecalibacterium prausnitzii (FPRAU) in the gut microbiome ecosystem from food frequency survey-based nutrient intake data and making associated recommendations for improving Faecalibacterium prausnitzii

Abstract

The present invention relates to a system and method for estimating an individual's amount of Faecalibacterium prausnitzii (Fprau) and providing personalized recommendations for maintaining or improving Fprau. In some embodiments of the present invention, the individual's Fprau amount is estimated based on a food frequency questionnaire (FFQ) record. In some embodiments, the method is performed by a computer system. In some embodiments of the present invention, personalized recommendations and dietary advice are given to the individual to maintain or improve the individual's Fprau.

Description

[Technical field]

[0001] The present invention relates to a system and method for estimating Faecalibacterium prausnitzii (Fprau) load in an individual. In some embodiments of the invention, the Fprau load of an individual is estimated based on nutrient data derived from a food frequency questionnaire (FFQ) record of the individual. In some embodiments, the method is implemented by a computer system. In some embodiments of the invention, personalized recommendations and dietary and nutritional advice are provided to an individual to maintain or improve the Fprau load of the individual. [Background technology]

[0002] Faecalibacterium prausnitzii (Fprau) is an important bacterium in the human gut microbiome ecosystem, which is associated with the causation of various symptoms (e.g., its importance in human health (Miquel, S et al. Current opinion in microbiology, 2013, Ferreira-Halder, CV et al. Clinical gastroenterology, 2017), anti-inflammatory (Quevrain, E et al. Gut, 2016; Sokol, H et al. PNAS, 2008), ulcerative colitis (Machiels, K et al. Gut, 2014), Crohn's disease (Takahashi, K et al. Digestion, 2016), pediatric allergy (e.g. asthma) (Demirci, M et al. Allergologia et immunopathologia, 2019), IBD (Zhao H, Xu H, Chen S, He J, Zhou Y, Nie Y., 2020.J Gastroenterol Hepatol.;Machiels K et al., Gut.2014), frailty (Jackson MA et al.Genome Med.2016), etc.).

[0003] Furthermore, Fprau is affected under conditions of combined stress to the gut microbiome ecosystem, such as drastic dietary changes or antibiotic use. For example, Mardinoglu et al. (Cell Metabolism 2018) show a decrease in Fprau under ketogenic diet loading. Similarly, Palleja et al. (Nature Microbiology 2018) show a decrease in Faecalibacterium prausnitzii (Fprau) under antibiotic administration. Furthermore, David et al. (Nature 2014) provide evidence of a decrease in Fprau abundance under high-fat diet loading.

[0004] Typically, evaluation of bacteria in the gut microbiome ecosystem requires collection of fecal samples, storage and processing of samples, laboratory steps such as DNA extraction and sequencing, complex bioinformatics analysis, and scientific evaluation. This is costly, time-consuming, and laborious, and requires specialized skills and expertise that are not always available or easily accessible to everyone. Furthermore, many adults are reluctant to provide fecal samples.

[0005] Therefore, there is a need for non-invasive and simpler methods to estimate the relative abundance of F. prausnitzii (Fprau) and to promote Fprau in the human gut microbiome ecosystem.

[0006] [Summary of the Invention] We have found a simpler method to estimate the relative abundance of Faecalibacterium prausnitzii (Fprau) from nutritional intake data. Here, the key steps of the present invention are: (i) an individual's response to certain dietary questions, (ii) an estimation of nutrient intake for the individual, (iii) the use of a machine learning-based model, and (iv) prediction of the estimated relative abundance of F. prausnitzii (Fprau).

[0007] Thus, the present invention broadly relates to a method for determining the status of intestinal Faecalibacterium prausnitzii (Fprau), comprising: The present invention relates to a method comprising the steps of: (i) assessing the relative amount of Fprau in an individual's gut microbiome ecosystem; and (ii) accordingly providing recommendations for maintaining or improving the relative amount of Fprau.

[0008] In another aspect, the present invention provides a method for optimizing one or more dietary interventions for a subject, comprising: (i) determining the Fprau status of a subject according to the method of any one of claims 1 to 5; (ii) administering said dietary intervention to said subject.

[0009] The methods and systems of the present invention advantageously implement artificial intelligence-based machine learning methods to estimate the Fprau abundance of an individual's gut microbiome from nutritional data derived from a food frequency survey (FFQ).

[0010] One advantage of the present invention is that an individual does not need to provide a biological sample to obtain an estimate of their Fprau amount. Instead, the present invention works by using a predictive model based on data provided by the user regarding their responses to a set of food frequency surveys to identify nutrient intakes as predictive features.

[0011] In another embodiment, the invention relates to a kit comprising a food frequency survey to determine nutrient intake to predict Fprau status of said subject, and a computer-implemented tool for dietary recommendations to maintain or improve Fprau relative amounts.

[0012] One advantage of some embodiments of the present invention is that for Fprau's status assessment, the questionnaire responses of individual users are evaluated to personalize recommendations and advise to maintain or improve an individual's Fprau's status.

[0013] Various embodiments of the disclosed system display to the user a dashboard or other suitable user interface customized based on the user's input into the questionnaire, the predicted Fprau volume, and personalized recommendations for maintaining or improving Fprau.

[0014] In some embodiments, the disclosed system may be linked to automatically collect the required input data from food records captured by a user in various forms, such as a food diary or an app that logs food and drink records.

[0015] In some embodiments, the systems and methods disclosed herein may also be used by nutritionists, health care professionals, and others other than individual users.

[0016] Further advantages of the present disclosure will become apparent from the following detailed description and the associated drawings. [Brief description of the drawings]

[0017] [Figure 1A] ROC performance of low vs. non-low models (I). ROC performance of low vs. non-low models for Fprau quantile with bin definition based on (mean-1*std) vs. residuals. ROC of (A) training in cross-validation mode (B) holdout / test set. [Figure 1B] ROC performance of low vs. non-low models (I). ROC performance of low vs. non-low models for Fprau quantile with bin definition based on (mean-1*std) vs. residuals. ROC of (A) training in cross-validation mode (B) holdout / test set. [Figure 2A] ROC performance of low vs. non-low models (II). ROC performance of low vs. non-low models for Fprau volume using bin definition based on 1st / lowest quartile vs. remaining. ROC of (A) training in cross-validation mode (B) holdout / test set. [Figure 2B]ROC performance of low vs. non-low models (II). ROC performance of low vs. non-low models for Fprau volume using bin definition based on 1st / lowest quartile vs. remaining. ROC of (A) training in cross-validation mode (B) holdout / test set. [Figure 3A] Important features for the low vs. non-low model (I). Important features and their association with Fprau are shown. [Figure 3B] Important features for the low vs. non-low model (I). Important features and their association with Fprau are shown. [Figure 4A] Important features for the low vs. non-low model (II). Important features and their association with Fprau are shown. [Figure 4B] Important features for the low vs. non-low model (II). Important features and their association with Fprau are shown. [Figure 5A] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5B] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5C]SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5D] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5E] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5F] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5G]SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5H] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5I] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 5J] SHAP dependency plots of key features in low vs. non-low models. SHAPE dependency plots for low vs. non-low models in Fprau volume using quartile-based tier definitions are shown for key example features. The reference class here was "low," so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "low" class. [Figure 6] Results of F. prau abundance determined by quantitative PCR technique in samples collected after A) 24 hours and B) 48 hours. [Figure 7] F.prau ASV6 responded to inulin, Pump_full and B vitamins + inositol. A) Samples collected after 24 hours and B) 48 hours. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] definition Some definitions are provided below. However, definitions may be found in the "Embodiments" section below, and the heading "Definitions" above does not imply that such disclosure in the "Embodiments" section is not a definition.

[0019] All percentages stated herein are by total weight of the composition unless otherwise stated. As used herein, "about", "approximately" and "substantially" are understood to refer to numbers within a certain range of numerical values, for example, within the range of -10% to +10% of the referenced number, preferably within the range of -5% to +5% of the referenced number, more preferably within the range of -1% to +1% of the referenced number, and most preferably within the range of -0.1% to +0.1% of the referenced number. All numerical ranges herein should be understood to include all integers or fractions within the range. Furthermore, these numerical ranges should be interpreted to support claims directed to any number or subset of numbers within the range.

[0020] The terms "comprise", "comprises", and "comprising" should be interpreted as inclusive rather than exclusive. Similarly, the terms "include", "including", and "or" should all be interpreted as inclusive unless such interpretation is clearly prevented by the context. However, the compositions disclosed herein may not include elements not specifically disclosed herein. Thus, disclosure of an embodiment using the term "comprising" includes disclosure of an embodiment "consisting essentially of" the specified components, and an embodiment "consisting of" the specified components.

[0021] The terms "at least one of" and "and / or" used in the context of "at least one of X or Y" and "X and / or", respectively, should be interpreted as "X" or "Y" or "X and Y". For example, "at least one inositol or sorbitol" and "inositol and / or sorbitol" should be interpreted as "inositol without sorbitol", or "sorbitol without inositol", or "inositol without sorbitol".

[0022] As used herein, the terms "examples" and "such as," especially when followed by a list of terms, are merely exemplary and illustrative and should not be considered as exclusive or inclusive. As used herein, a condition "associated with" or "linked with" another condition means that the conditions occur simultaneously, preferably that the conditions are caused by the same underlying condition, and most preferably that one of the specified conditions is caused by the other specified condition.

[0023] The relative terms "promote," "improve," "increase," "enhance," and the like refer to an enhanced F. prausnitzii status in the subject's microbiome following administration of a composition disclosed herein (including sorbitol and / or inositol) as obtained by the recommended dosing herein, as compared to the F. prausnitzii status in the subject's microbiome, which can be assessed by at least one or more of: (i) a higher total amount of F. prausnitzii in the subject's microbiome (i.e., total cfu of Faecalibacterium prausnitzii), or (ii) a higher relative percentage of Faecalibacterium prausnitzii compared to other bacteria in the subject's microbiome (i.e., cfu of Faecalibacterium prausnitzii / cfu of other bacteria).

[0024] As used herein, the terms "food," "food product," and "food composition" refer to a product or composition intended for oral ingestion by humans or other mammals and containing at least one nutrient for humans or other mammals.

[0025] As used herein, "nutritional composition" and "nutritional product" include any number of food ingredients and possibly optional additional ingredients based on the functional needs of the product and in full compliance with all applicable regulations. Optional ingredients may include, but are not limited to, conventional food additives such as one or more acidulants, additional thickeners, pH adjusting buffers or pH adjusting agents, chelating agents, colorants, emulsifiers, excipients, flavorings, minerals, osmotic agents, pharma-ceutically acceptable carriers, preservatives, stabilizers, sugars, sweeteners, modifiers and / or vitamins. Optional ingredients may be added in any suitable amount.

[0026] As used herein, a "lifestyle characteristic" is any lifestyle choice made by a subject, including any dietary intake data, activity measures or data obtained from lifestyle, motivation, or preference questionnaires. In one embodiment, the lifestyle characteristic is whether the subject is an alcoholic or non-alcoholic. In another embodiment, the lifestyle characteristic is whether the subject is a vegetarian or omnivore.

[0027] In some embodiments, the term "nutrient" as used herein refers to a compound that has a beneficial effect on the body, for example, providing energy, growth or health. This term includes organic and inorganic compounds. As used herein, the term "nutrient" can include, for example, macronutrients, micronutrients, essential nutrients, conditionally essential nutrients and phytonutrients. These terms are not necessarily mutually exclusive. For example, certain nutrients can be defined as either macronutrients or micronutrients depending on a particular classification system or list. The phrase "at least one nutrient" or "one or more nutrients" means, for example, 1, 2, 3, 4, 5, 10, 20 or more nutrients.

[0028] In various embodiments, the term "macronutrient" is used herein consistent with usage well understood in the art to generally include nutrients required in large amounts for normal growth and development of an organism. Macronutrients in these embodiments may include, but are not limited to, carbohydrates, fats, proteins, amino acids, and water. Certain minerals, such as calcium, chloride, sodium, or potassium, may also be classified as macronutrients.

[0029] In various embodiments, the term "micronutrients" is used herein consistent with its use well understood in the art to generally include compounds that have a beneficial effect on the body (e.g., providing energy, growth, or health) but that are required only in minor or trace amounts. In such embodiments, the term includes organic and inorganic compounds, such as amino acids, nucleotides, and fatty acids; vitamins, antioxidants, minerals, trace elements (e.g., iodine), and electrolytes (e.g., sodium chloride and salts thereof).

[0030] In various embodiments, the term "essential nutrient" is used herein consistent with its well-understood usage in the art. Essential nutrients cannot be synthesized in the body or in sufficient quantities and must be obtained by an organism from its environment. Essential nutrients include essential fatty acids, essential amino acids, vitamins, and certain dietary minerals. For example, there are two essential fatty acids in humans: alpha-linolenic acid (an omega-3 fatty acid) and linoleic acid (an omega-6 fatty acid). Of the 20 amino acids, nine cannot be endogenously synthesized by humans (phenylalanine, valine, threonine, tryptophan, methionine, leucine, isoleucine, lysine, and histidine), and are considered essential amino acids.

[0031] In various embodiments, the term "conditionally essential nutrients" is used herein in accordance with well-understood usage in the art. Conditionally essential nutrients are specific organic molecules that can normally be synthesized by an organism, but under certain conditions, such biosynthesis is not sufficient to prevent shedding syndrome. For example, choline, inositol, taurine, arginine, glutamine and nucleotides are classified as conditionally essential, particularly for the diet and metabolism of newborns.

[0032] In various embodiments, the term "non-essential nutrient" is used herein in accordance with its well-understood usage in the art. Non-essential nutrients are nutrients that can be synthesized by the body. Non-essential nutrients can often be absorbed from ingested food. Non-essential nutrients are substances found in food that can still have significant health effects, whether beneficial or toxic. For example, it has recently been found that most dietary fiber is not absorbed by the human digestive tract, but is important for maintaining a large proportion of intestinal motility to avoid constipation, or has a beneficial effect on the gut microbiome, with various bacteria having different abilities or preferences for fiber utilization.

[0033] In various embodiments, the term "deficiency" is used herein consistent with its well understood usage in the art. Deficiency can result from many causes, including inadequate nutrient intake, referred to as malnutrition, or conditions that prevent the utilization of nutrients in the body. Conditions that can prevent the utilization of nutrients include problems with nutrient adsorption, substances that require more of a nutrient than normal, conditions that cause the breakdown of a nutrient, and conditions that cause the excretion of greater amounts of a nutrient.

[0034] In various embodiments, the term "toxicity" is used herein consistent with its well-understood usage in the art. Nutrient toxicity occurs when an excess of a nutrient is harmful to an organism.

[0035] A "subject" or "individual" is a mammal, preferably a human, but may also be a companion animal such as a dog or cat.

[0036] In some embodiments, the low, non-low Fprau bins are defined as "low" as below the first or lower quartile of the population Fprau distribution, and "non-low" as the remainder of the distribution.

[0037] In some embodiments, the high and non-high Fprau classes are defined as "high" as above the third or upper quartile of the population Fprau distribution, and "non-high" as the remainder of the distribution.

[0038] In some embodiments, the low and high Fprau classes are defined as "low" being below the first or lower quartile of the Fprau distribution and "high" being above the third or upper quartile of the Fprau distribution.

[0039] In some embodiments, the low Fprau bin is defined as data that is less than the mean minus the standard deviation of the Fprau distribution, and the non-low Fprau bin is defined as the remainder of the data.

[0040] In some embodiments, high Fprau bins are defined as data that are above the mean plus the standard deviation of the Fprau distribution, and non-high Fprau bins are defined as the remainder of the data.

[0041] In some embodiments, the low and high Fprau bins are defined as "low" as data less than the mean minus the standard deviation in the Fprau distribution and "high" as data greater than the mean plus the standard deviation in the Fprau distribution.

[0042] In some embodiments, low, non-low, high, and non-high Fprau tiers are defined based on data distribution, such as in different population data sets with different numerical cutoffs that would be apparent to one of skill in the art.

[0043] Although variations on the above, it should be understood that such groups can be defined in somewhat different ways, such as median / mean + / - 1 standard deviation, or median / mean + / - 1 / 2 standard deviation, or median / mean + / - 1 / 2 interquartile range, or many other possible ways, such as a different percentage of the data points falling into the brackets than those set forth above, that would be apparent to one skilled in the art of data analysis.

[0044] The Receiver Operating Characteristic (ROC) curve is one of the best developed statistical tools for describing the performance of diagnostic tests measured on a continuous scale. The use of ROC is based on having two outcomes from the prediction. Numerical indices of the ROC curves were used to summarize the curves. These summary measures were also used to compare ROC curves.

[0045] "Area under the ROC curve" (AUC) is the most widely used summary measure. A perfect predictive model with an ideal ROC curve has an AUC value = 1.0, while a random predictive model has an AUC = 0.5. ROC curve AUC values ​​moving from 0.5 towards 1.0 indicate an improvement and better performance of the predictive model.

[0046] Many other measures of model performance, such as True Positives (TP), False Positives (FP), True Negatives (TN), False Negatives (FN), Total Predicted Positives, Total Predicted Negatives, Total Real Positives, Total Real Negatives, Sensitivity / Hit Rate / Recall / True Positive Rate (TPR), Specificity / Selectivity / True Negative Rate (TNR), Prevalence, Precision / Positive Predictive Value (PPV), Negative Predictive Value (NPV), Miss Rate / False Negative Rate (FNR), Fallout / False Positive Rate (FPR), False Discovery Rate (FDR), False Omission Rate (FOR), Prevalence Threshold (PT), Threat Score (TS) / Critical Success Index (CSI), Accuracy (ACC), Balanced Accuracy (BA), Random Accuracy, Total Accuracy, F1-Score, Matthews Correlation Coefficient (MCC), Fowlkes Mallows Index (FM), Informedness / Bookmaker Indices such as informedness (BM), markedness (MK) / delta P, positive likelihood ratio (LR+), negative likelihood ratio (LR-), diagnostic odds ratio (DOR), and kappa can be calculated on the confusion matrix.

[0047] AUC-ROC is the area under the curve, constructed by plotting the true positive rate against the false positive rate at various probabilities. AUC-PR is the area under the precision-recall curve.

[0048] The term "feature" is used repeatedly herein. In some embodiments, the term "feature" as used herein refers to an input parameter to a model. This term includes answers obtained from a set of questionnaires, e.g., nutritional intake derived from a food frequency survey. These features are not necessarily mutually exclusive.

[0049] In various embodiments, user-specific (or population-specific) inputs into the system of the present disclosure are programmable and configurable, and include gender, age, weight, height, physical activity level, whether or not obese, and the like.

[0050] Embodiment The inventors have shown that it is possible to create a predictive tool based on features obtained from questionnaires, e.g. food frequency surveys converted into nutrient intakes, and that this makes it possible to predict the intestinal Fprau status (e.g. low or not low).

[0051] In a first embodiment, the present invention provides a method for determining the status of Faecalibacterium prausnitzii (Fprau) in the intestine, comprising: (i) determining intestinal Fprau status in a subject; (ii) providing recommendations for improving or maintaining the Fprau condition in said subject; and The present invention provides a method comprising:

[0052] In one embodiment, the methods and systems of the present invention implement artificial intelligence-based machine learning methods to estimate an individual's gut microbiome Fprau abundance from nutrient data derived from a food frequency questionnaire (FFQ).

[0053] In another embodiment, this is done by using a predictive model based on data provided by the user regarding their responses to a set of food frequency surveys to identify nutrient intakes as predictive features.

[0054] In a further embodiment, the determination of the status of the gastrointestinal Fprau may further be provided by a biological sample to quantify the diversity of the subject's microbiome.

[0055] In a preferred embodiment, the present invention determines the Fprau status of an individual in relation to its location in the distribution in a larger population. For example, low, high or low, non-low or high, non-high, in terms of having low or not low (notLow), high or not high (notHigh), or combined together to determine low, medium, high, and optionally cross-validated by another low rating vs. high rating, are defined in various ways based on the distribution found in large-sized general populations, such as the American Gut Project (AGP) (McDonald D, et al. mSystems. 2018).

[0056] Once the relative amount of Fprau is determined, the systems and methods of the present invention provide recommendations such as nutritional supplements, dietary recommendations, menu recommendations and recipe recommendations to improve or maintain the amount of Fprau in the intestinal ecosystem, thereby contributing to maintaining and improving the condition of Fprau or enhancing the proliferation of Fprau.

[0057] In preferred embodiments, some interventions to maintain or improve their abundance and function are provided as Example 5.

[0058] In addition, other methods are known in the art, which may include: (i) Dietary fiber intake (Lin D et al.Br J Nutr.2018;Benus RF et al.Br J Nutr.2010); (ii) Consuming a Mediterranean diet (Gutierrez-Diaz I et al. J Agric Food Chem. 2017; Meslier V et al. Gut. 2020; Haro C et al. J Clin Endocrinol Metab. 2016; (iii) consuming other foods (Verhoog, S et al. Nutrients, 2019; Fritsch J et al. 2020; Kahleova H et al. Nutrients. 2020; Medina-Vera I et al. Diabetes Metab. 2019; (iv) consuming foods containing pectin, such as fruits (Lopez-Siles, M et al. Applied and environmental microbiology, 2012); (v) drinking red wine (Moreno-Indias I et al. Food Funct. 2016); (vi) consuming raisins (Wijayabahu AT et al. Nutr J. 2019), etc.

[0059] Further interventions to benefit Fprau could be through vitamins or probiotics, but current human clinical trial data appear to be lacking in this regard.

[0060] In another embodiment, the method comprises assessing a characteristic parameter associated with intestinal Fprau's status as low, medium or high.

[0061] In one embodiment of the present invention, the improvement or maintenance of Fprau levels can be determined from biological samples taken from a subject before and after the recommendations of the present invention by measuring parameters of microbial species in the gut. Thus, the maintenance or improvement of Fprau can be determined over time after an individual follows, for example, the nutrition, diet, menu and recipe recommendations of the present invention.

[0062] In various embodiments, the systems disclosed herein provide supplement, food item, menu, or recipe recommendations that indicate a nutritional impact on Fprau. In these embodiments, the system determines and stores one or more indicators of the needs of an individual for whom a recommendation is being calculated, over a given period of time, such as a meal, a full day, a week, or a month.

[0063] In further embodiments, individuals can provide their own weightings that are tailored to their personal preferences and health status. Using these personalized ranges and / or weightings, the disclosed system can then calculate fully personalized advice for maintaining or improving the individual's Fprau status.

[0064] In one embodiment, the system of the present disclosure includes or is connected to a database that includes food items, menus or recipes, and their respective nutrient contents. In this embodiment, the system of the present disclosure includes a fuzzy search function that allows a user to input a food item that has been consumed (or is to be consumed) and then search the database to find items that are closest to the user-provided item. In this embodiment, the system of the present disclosure uses stored nutritional information about the matched food items to determine whether the item is specifically Fprau's Microbiome Friendly, as described below.

[0065] In various embodiments, the systems of the present disclosure further include an interface (e.g., a graphical user interface) for displaying the amount of each nutrient available in each food that makes up the meal and for displaying the amount of available energy to be ingested. In some embodiments, the interface allows the user to modify the amounts of various foods or energy to be ingested. In other embodiments, the system is configured to determine the amount of food to be ingested or energy to be consumed using data not entered by the user, such as by scanning one or more bar codes, QR codes, or RFID tags, an image recognition system, or by tracking items ordered from a menu or purchased at a grocery store.

[0066] Various embodiments of the disclosed system display a dashboard or other suitable user interface to the user, customized based on the user's needs. In embodiments of the system disclosed herein, a graphical user interface is advantageously provided that for the first time allows a user to input data regarding the user's responses to a set of questionnaires and see a display of a score that reflects the overall placement of the user's condition in a commonly observed distribution of Fprau volumes, based appropriately on a prediction.

[0067] In some embodiments, the disclosed system may be linked to automatically collect the required input data from food records captured by a user in various forms, such as a food diary or an app that logs food and drink records.

[0068] All of the disclosed methods and procedures described in this disclosure can be implemented using one or more computer programs or components. These components may be provided as a series of computer instructions on any conventional computer-readable or machine-readable medium, including volatile and non-volatile memories such as RAM, ROM, flash memory, magnetic or optical disks, optical memory, or other storage media. The instructions may be provided as software or firmware, and may also be implemented in whole or in part in hardware components such as ASICs, FPGAs, DSPs, or any other similar devices. The instructions may be configured to be executed by one or more processors that, when executing the series of computer instructions, perform or facilitate the performance of all or part of the disclosed methods and procedures.

[0069] As mentioned above, the disclosed system, in some embodiments, relies on one or more modules (hardware, software, firmware, or a combination thereof) to perform various functions described above.

[0070] Those skilled in the art will understand that they can freely combine all aspects of the invention disclosed herein without departing from the scope of the invention disclosed. Furthermore, aspects described for different embodiments of the invention may be combined. Although the invention has been described by way of examples, it should be understood that changes and modifications can be made without departing from the scope of the invention as defined in the claims and without losing its intended advantages. Accordingly, such changes and modifications are intended to be encompassed by the appended claims.

[0071] Various preferred features and embodiments of the present invention will now be described by way of non-limiting examples. EXAMPLES

[0072] Example 1: Conversion of food frequency intake data to nutrients A publicly available citizen science project called the American Gut Project (AGP) (McDonald D, et al. mSystems. 2018) was converted to nutrient intakes using a tool made available to the public by the AGP called vioscreen.

[0073] Example 2: Building a model to estimate the relative abundance of Faecalibacterium prausnitzii (Fprau) A predictive model was constructed to determine the relative abundance of Faecalibacterium prausnitzii (Fprau) in individual subjects. In particular, the model predicted the Fprau relative abundance by several characteristic parameters to determine whether the subject had a "low" or "non-low", "high" or "non-high", or "low" or "high" Fprau abundance according to the categories defined above.

[0074] A cube root transformation was performed to make the Fprau quantities normally distributed before ranking them into different categories. The values ​​for the various definitions of ranking were 1st / lower quartile -0.2819, 3rd / upper quartile -0.4666, mean -std -0.1954, mean +std -0.5220.

[0075] To build the classification model, the data was split into a training set, "train", and a test set, "holdout / test set". For optimal model performance, we used downsampling to balance the imbalanced classes that may occur based on the class definition.

[0076] The training set was used by the machine learning algorithm to train the model. The training set included finding the variables (i.e., features) and thresholds (or coefficients) used to classify the groups. Learning from the data was done in a cross-validation fashion (k-fold cross-validation, e.g., 3-fold), where some parts of the training data were used to train the model and other parts were used for internal testing, or this method was also repeated several times (repeated k-fold cross-validation, e.g., 10-fold, 10 repetitions).

[0077] The holdout / test set was only used to check the performance of the final trained model. Therefore, this holdout / test data set was not used in the model training stage. We evaluated multiple statistical models (different machine learning algorithms) using freely available tools (R software, Python) to identify the best model for low vs. non-low, high vs. non-high, and low vs. high for Fprau.

[0078] Evaluation of model performance was important at all stages of modeling. Once the model was trained, it was applied to the holdout / test data that was not used during the training stage. The model calculated the probability of belonging to each group (e.g., "low", "non-low"). Based on this probability, the final decision was made, and therefore the use of a threshold was necessary. This threshold influenced the final classification of the subject, whether it was correctly classified or not. Therefore, the error was evaluated for different choices of threshold. For each given threshold, a confusion matrix was calculated. This confusion matrix essentially counts the number of correctly and incorrectly classified subjects. By using different thresholds, many confusion matrices were generated, which were then used to derive the sensitivity and specificity at different thresholds. These two metrics (sensitivity and specificity) were generally presented in the form of a receiver operating curve (ROC); this summarized the model performance over several thresholds.

[0079] Receiver operating characteristic (ROC) curves were generated for this model. We either defined a group of "low" subjects (and a "non-low" group) and predicted the probability that a subject would fall into this group; or we defined a subject as falling into a "high" group (and a "non-high" group) and predicted the probability that a subject would fall into this group; or we defined a subject as falling into a "low" group (and a "high" group) and predicted the probability that a subject would fall into this group.

[0080] As mentioned above, the dataset used for the predictive model examples comes from the American Gut Project (AGP) database (http: / / americangut.org).

[0081] Example 3: Estimation of "low" Fprau amounts from nutrient intake data (I) Using these parameters, a model of low vs. non-low Fprau doses was trained: Class definition: (mean-1 *std) vs. rest, feature cutoff: none, algorithm: RandomForest, training mode: cv-splits-3, cv-repeats-3, post-processing training size: 896, holdout / test size (original / pre-processing pre-train / test split): 764 (test percentage: 20.0%). The results obtained for training in cross-validation were accuracy -0.58 ± 0.02, sensitivity -0.61 ± 0.05, specificity -0.58 ± 0.03. The training ROC curve is shown in Figure 1A. The results obtained for the holdout / test set were accuracy -0.64, sensitivity -0.56, specificity -0.65. The holdout / test ROC curve is shown in Figure 1B. The important features and their association with the Fprau measure are shown in Figure 3.

[0082] Example 4: Estimation of "low" Fprau amounts from nutrient intake data (II) Another model for low vs. non-low Fprau dose was trained with these parameters: Class definition: 1st / lowest quartile vs. remaining, feature cutoff: none, algorithm: RandomForest, Train mode: cv-splits-3, cv-repeats-3, post-processing train size: 1554, holdout / test size (original / pre-processing train / test split): 764 (test percentage: 20.0%). The results obtained for the training in cross-validation were accuracy -0.58 ± 0.02, sensitivity -0.62 ± 0.03, specificity -0.57 ± 0.03. The training ROC curve is shown in Figure 2A. The results obtained for the holdout / test set were accuracy -0.59, sensitivity -0.57, specificity -0.59. The holdout / test ROC curve is shown in Figure 2B. The important features and their association with Fprau dose are shown in Figure 4.

[0083] Example 5: Recommendations for maintaining or improving Fprau levels For the model presented in Example 3, the top 30 features constituting the model are shown in FIG. 3. For the model presented in Example 4, the top 30 features constituting the model are shown in FIG. 4. (A) and (B) were both obtained by performing a SHaPley Additive Description (SHAP) value analysis (Lundberg SM, et al. Nat Mach. 2020). (A) shows the average impact per feature on the model output in order of importance from high to low. The leading / best feature was the top horizontal bar. The next best feature was the second horizontal bar, and so on. (B) shows in more detail the impact of the feature per instance / sample on the model output. The color gradation from gray to black indicates the low to high value of that feature. The vertical line at 0.00 defines the direction of the impact (the left side is a negative impact on the model output, and the right side is a positive impact on the model output). Here, the SHAP analysis output was for the reference class, which was "low".

[0084] If a feature has a black value toward the right of the 0.00 vertical line, this indicates that higher values ​​of this feature contribute positively to the model output. Vice versa, if a feature has a black value toward the left of the 0.00 vertical line, this indicates that higher values ​​of this feature contribute negatively to the model output. Similarly, if a feature has a gray value toward the right of the 0.00 vertical line, this indicates that lower values ​​of this feature contribute positively to the model output. Vice versa, if a feature has a gray value toward the left of the 0.00 vertical line, this indicates that lower values ​​of this feature contribute negatively to the model output.

[0085] As can be seen from Figures 3 and 4, by way of example, some of the key features of the model for predicting low vs. non-low salt were related to inositol (inositol in g), alphacar (alpha-carotene provitamin A carotenoid in mcg), betacar (beta-carotene provitamin A carotenoid in mcg), pectin (pectin in g), fiber (total dietary fiber in g, soluble dietary fiber in g, insoluble dietary fiber in g), and vitamin A (vita_iu-total vitamin A activity in IU, vita_rae-total vitamin A active retinol equivalents in mcg, vita_re-total vitamin A active retinol equivalents in mcg).

[0086] In Figure 5, for each feature, the SHAP dependency plot showed a point with the feature value on the x-axis and the corresponding Shapley value on the y-axis for each data instance / sample. SHAP explained the prediction of each instance by calculating the contribution of each feature to the prediction. The explanation of the Shapley value was expressed as a linear model, as an additive feature imputation method. The reference class here was "Low", so a positive coefficient of the SHAP value for the corresponding x-value of a feature indicates how much the model was influenced by this feature in predicting the "Low" class.

[0087] As can be seen, inositol influenced Fprau amount, which was one of the top features used by the model (Figures 3 and 4). As can be seen in Figure 5A, the specific intake value of inositol has a relationship with the influence on the model output, and low inositol intake tends to have Fprau status on the lower side, while higher intake of inositol tends to put Fprau status in the "non-low" class. Therefore, Fprau status benefits from more inositol intake from the diet, preferably more than 0.2g inositol per day, which can be obtained by eating fruits such as cantaloupe and orange.

[0088] Figures 3 and 4 show the importance of alphacar (alpha carotene provitamin A carotenoid in mcg), and Figure 5B shows the SHAP dependency plot for alphacar (alpha carotene provitamin A carotenoid in mcg). For all individuals with intakes below alphacar (all data points on the x-axis below a value of about 2000), the SHAP values ​​were positive, indicating that this was associated with being in the "low" class of Fprau status. Similarly, only for individuals with higher intakes of alphacar above about 2000, the SHAP values ​​were negative, indicating that this was associated with currently being in a "non-low" Fprau status. Therefore, the recommendation of the present invention is to consume yellow-orange vegetables such as carrots, sweet potatoes, pumpkins, and winter squash, which are reported to be rich in alpha-carotene, as well as dark green vegetables such as broccoli, kidney beans, green peas, spinach, turnip greens, collards, leaf lettuce, and avocado.

[0089] Based on the same reasoning and explanation above, looking at Figures 3, 4, and 5C as a whole, we can infer that an intake of betacar (β-carotene provitamin A carotenoid in mcg) greater than about 10,000 mcg was good for Fprau because it was associated with being in a "non-low" microbiome state. Based on these results, the present recommendation was to consume more yellow and orange fruits such as cantaloupe, mango, pumpkin, and papaya, as well as orange root vegetables such as carrots and sweet potatoes. Betacar is also present in green leafy vegetables such as spinach, kale, sweet potato leaves, and bitter melon leaves. It is also sold as a dietary supplement. The following table lists the main foods and their beta-carotene content (https: / / en.wikipedia.org / wiki / Beta-Carotene)

[0090] [Table 1]

[0091] Based on the same rationale and explanation as above, when Fig. 4 and Fig. 5D are taken together, it can be inferred that pectin intake (pectin, g) is associated with Fprau status. In particular, an increase in pectin intake above 4 g was associated with "non-low" Fprau status. Therefore, the recommendation of the present invention would be to consume more pectin, for example from pears, apples, guavas, quince, plums, gooseberries, and oranges, as well as other citrus fruits that are reported to contain large amounts of pectin. Typical levels of pectin in fresh fruits and vegetables are as follows: apples: 1-1.5%, apricots: 1%, cherries: 0.4%, oranges: 0.5-3.5%, carrots: 1.4%, citrus peels: 30%, rose hips: 15% (https: / / en.wikipedia.org / wiki / Pectin).

[0092] Summarizing from the SHAP analysis shown in Figures 3, 4, 5E, 5F and 5G, the conclusion was that increasing the amount of fiber positively impacted the microbiome. Fiber in this data was captured as total fiber-total dietary fiber (g) (fiber), insoluble fiber-insoluble dietary fiber (g) (fibinso), and soluble fiber-soluble dietary fiber (g) (fibh20). Based on the interpretation made here, the recommendation of the present invention was to have more than 40g of total fiber, more than 30g of insoluble fiber, and more than 10g of soluble fiber. Therefore, the recommendation of the present invention would be to consume more total fiber, consisting of both insoluble and soluble fiber, to increase the amount of Fprau that can be obtained from food sources. Dietary fiber is present in fruits, vegetables and whole grains. The amount of fiber contained in common foods is listed here (https: / / en.wikipedia.org / wiki / Dietary_fiber).

[0093] [Table 2]

[0094] Soluble fiber is present in varying amounts in all plant foods, including legumes (peas, soybeans, lupins and other beans), oats, rye, chia and barley, some fruits (including figs, avocados, plums, prunes, berries, ripe bananas, and the skin of apples, quince and pears), certain vegetables such as broccoli, carrots and Jerusalem artichokes, tuberous and root vegetables such as sweet potatoes and onions (their skins are also a source of insoluble fiber), psyllium seed husks (mucilage soluble fiber) and flaxseed, and nuts, with almonds being the highest in dietary fiber.

[0095] Sources of insoluble fiber include: whole grain foods, wheat and corn bran, legumes such as beans and peas, nuts and seeds, potato skins, lignans, beans, vegetables such as cauliflower, zucchini (sturgeon), celery, and nopales, some fruits including avocados, and the skins of some fruits including unripe bananas, kiwifruit, grapes, and tomatoes.

[0096] Similarly, Figure 5H, Figure 5I, Figure 5J show that increasing the amount of vitamin A has the desired effect on Fprau amount. This was captured in the AGP data as vita_iu (total vitamin A activity in IU), vita_rae (total vitamin A active retinol activity equivalent in mcg), and vita_re (total vitamin A active retinol equivalent in mcg). 1 IU of retinol corresponds to approximately 0.3 micrograms (300 nanograms). According to Figure 5H, Figure 5I, Figure 5J, vita_iu>20000 IU, vita_rae>2000mcg, and vita_re>3000mcg had the desired effect on Fprau amount.

[0097] Dietary vitamin A comes from two sources. Animal products contain retinaldehyde and retinol, which are readily available in active forms such as retinoids. Precursors, called provitamins, must be converted to the active form and are obtained from fruits and vegetables that contain the yellow, orange, and dark green pigments known as carotenoids. The best known is beta-carotene. The amount of vitamin A is measured in retinol equivalents (RE). 1 RE is equivalent to 0.001 mg of retinol, or 0.006 mg of beta-carotene, or 3.3 international units of vitamin A. Retinoids are found naturally only in foods of animal origin. Each of the following contains at least 0.15 mg of retinoid per 1.75-7 oz (50-198 g): cod liver oil, butter, liver (beef, pork, chicken, turkey, fish), eggs, cheese, and milk.

[0098] Thus, the present invention's recommendation was to eat animal products such as eggs, liver, cod liver oil, etc. Additionally, synthetic retinols are commercially available as follows: Acon, Afaxin, Agiolan, Alphalin, Anatola, Aoral, Apexol, Apostavit, Atav, Avibon, Avita, Avitol, Axerol, Dohyfral A, Epiteliol, Nio-A-Let, Prepalin, Testavol, Vaflol, Vi-Alpha, Vitpex, Vogan, and Vogan-Neu. (https: / / en.wikipedia.org / wiki / Retinol)

[0099] The final recommendation was the result of a complex multivariate analysis in which traits were related to each other and the final effect on an individual's Fprau status was a combination of different factors.

[0100] The system of the present invention with its user-friendly digital interface incorporates these recommendations in order to communicate them directly to the user to improve the status of their own microbiome.

[0101] Example 6 Fecal samples were collected from healthy adult donors under human testing protocols. After receiving the fecal samples, small aliquots were prepared using storage buffer (PBS and 10% glycerol) and stored at -80°C before use. During each experiment, 250 μL fecal aliquots were inoculated into 10 mL Hungate tubes filled with minimal bacterial culture medium under strictly anaerobic conditions (oxygen < 3 ppm) in an anaerobic chamber. Different nutrients or combinations of nutrients, as shown in Table 1, were added to the culture medium at time 0 and the tubes were incubated at 37°C for 24 or 48 hours. Growth of Faecalibacterium prausnitzii (F.prau) was examined by two methods, namely quantitative PCR specifically targeting F.prau and 16S microbial rRNA gene sequencing.

[0102] [Table 3]

[0103] First, we examined the absolute amount of F. prau in the community after 24 or 48 hours (Figure 6A and Figure 6B). At 24 hours, F. prau in Pump_full, B vitamins + inositol, and inulin is at least twice as much as the control. However, only inulin was able to maintain a large amount of F. prau after 48 hours.

[0104] F. prau is heterogeneous and genetically diverse. Therefore, in this next experiment, we tested whether nutrients or combinations of nutrients favor certain F. prau to grow in a mixed community. A total of 14 genetically distinct Faecalibacterium species were found in the fermentation experiment, and most of them (ASV1, 6, 9, 12 and 13) responded positively to inulin as expected. Interestingly, ASV6 also responded to Pump_full and B vitamins + inositol at 24 hours (Figure 7A), and to a lesser extent at 48 hours (Figure 7B).

[0105] In conclusion, our results demonstrate that certain nutrient combinations (Pump_full and B vitamins + inositol) provide F. prau with a growth benefit in mixed communities, although the effect was not sustained for up to 48 hours. More importantly, the benefit of these nutrient combinations was only seen in certain F. prau, not all F. prau, suggesting that these nutrient combinations can be used in combination with F. prau-enhancing fibers such as inulin, or alone when inulin is not available in products or tolerated by people.

Claims

1. 1. A method for determining intestinal Faecalibacterium prausnitzii (Fprau) status, comprising: (i) determining intestinal Fprau status in a subject; (ii) providing recommendations for improving or maintaining the Fprau status in the subject; and A method comprising:

2. 2. The method of claim 1, wherein the determining of the intestinal Fprau status is by a food frequency survey to determine nutritional intake to predict the subject's Fprau status.

3. The method of claim 1, wherein the determination of the intestinal Fprau status is additionally performed by a biological sample to quantify the microbiome diversity of the subject.

4. The method of claim 1 , wherein the method is computer-implemented.

5. The method of claim 1 , wherein the method comprises assessing a characteristic parameter associated with intestinal Fprau status as low, medium, or high.

6. A computer-implemented method according to any one of claims 1 to 5, comprising: (i) determining intestinal Fprau status in a subject; (ii) providing personalized fiber composition recommendations; and (iii) delivering personalized nutrition recommendations.

7. A kit comprising a food frequency survey to determine nutritional intake to predict a subject's Fprau status, and a computer-implemented tool for making dietary recommendations to maintain or improve relative Fprau levels.

8. 1. A method for optimizing one or more dietary interventions in a subject, comprising: (i) determining the Fprau status of a subject according to the method of any one of claims 1 to 5; (ii) administering said dietary intervention to said subject.

9. The kit of claim 7, wherein the recommendation is a nutritional composition selected from the group consisting of a food product, a beverage product, or a dietary supplement, or a combination thereof in a kit-of-parts delivered to an individual.

10. The dietary intervention, (i) consuming total fiber, consisting of both insoluble and soluble fiber, provided by dietary sources such as fruits, vegetables, and whole grains; (ii) following the Mediterranean diet; (iii) Eating foods containing pectin, such as pears, apples, guavas, quince, plums, gooseberries, and oranges, as well as other citrus fruits that have been reported to contain large amounts of pectin; (iv) drinking red wine; (v) consuming raisins; (vi) Eating animal products such as eggs, liver, or cod liver oil; (vii) increasing the amount of Vitamin A; (viii) increasing intake of alpha-carotene, such as by consuming yellow-orange vegetables, such as carrots, sweet potatoes, pumpkins, winter squash, and dark green vegetables, such as broccoli, kidney beans, green peas, spinach, turnip greens, collards, leaf lettuce, and avocados; (ix) increasing intake of beta-carotene, such as consuming more yellow and orange fruits, such as cantaloupe, mango, pumpkin, and papaya, and orange root vegetables, such as carrots and sweet potatoes.

11. The method of any one of claims 1 to 5, wherein the recommendation is a nutritional composition selected from the group consisting of a food product, a beverage product, or a dietary supplement, or a combination thereof in a kit-of-parts delivered to the individual.

12. The method of claim 6, wherein the recommendation is a nutritional composition selected from the group consisting of a food product, a beverage product, or a dietary supplement, or a combination thereof in a kit-of-parts delivered to the individual.

13. The method of claim 8, wherein the recommendation is a nutritional composition selected from the group consisting of a food product, a beverage product, or a dietary supplement, or a combination thereof in a kit-of-parts delivered to the individual.

Images