RU2827489C2 - Epiaging: new ecosystem for healthy aging management - Google Patents

Abstract

FIELD: biotechnology.

SUBSTANCE: invention relates to biotechnology. Described is a method for calculating the biological age of a subject, which includes the following steps: (a) extraction of DNA from a substrate obtained from a subject; (b) measuring DNA methylation in DNA extracted from a substrate to obtain a DNA methylation profile; (c) analysis of the DNA methylation profile from 13 human CG sites located in the putative antisense region of the ElovL2 gene, the ElovL2 AS1 region, as specified in SEQ ID NO: 1, to obtain a polygenic estimate based on the following equation for calculating biological age: biological age = (CG1*87.5643+CG2*6.3301+CG3*-0.8691+CG4*1.9468+CG5*40.0336+CG6*49.4303+CG7*-14.7868+CG8*22.9042+CG9*-49.7942+CG10*111.7467+CG11*41.8108+CG12*0.4144+CG13*-150.8005)-71.6872, using multiple linear regression equations or neural network analysis; and (d) determining the biological age of the subject based on the polygenic assessment, where DNA extraction includes extraction of genomic DNA from saliva obtained from the subject. Also described is a kit for determining the biological age of a subject for implementing said method, comprising means and reagents for collecting and stabilizing a substrate from the subject, barcode scanner in the kit and instructions for collecting and stabilizing the substrate, where the substrate is saliva of the subject and where the substrate stabilization is intended for mailing the collected substrate for DNA extraction to measure DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile of the subject to determine the biological age of the subject.

EFFECT: invention extends the range of means for calculating the biological age of a subject.

12 cl, 7 dwg, 1 tbl, 6 ex

Description

This application claims priority under Section 119(e) of U.S. Provisional Patent Application No. 62/854,226, filed May 29, 2019, entitled “EpiAging; A Novel Ecosystem for Managing Healthy Aging,” the contents of which are incorporated herein by reference.

List of sequences

This application contains a Sequence Listing that has been submitted electronically in ASCII format and is incorporated herein by reference in its entirety. The ASCII copy created on May 28, 2020 is named TPC57505_Seq List_ST25.txt and is 4096 bytes in size.

Field of technology to which the invention relates

The present invention relates to epigenetics and DNA methylation signatures in human DNA in general and, in particular, to methods for determining epigenetic aging of an individual and managing healthy aging based on DNA methylation signatures. More specifically, the present invention relates to a method using DNA methylation signatures for molecular diagnostics, health management and lifestyle changes for personalized healthy aging using digital technology implemented in an application.

State of the art of the invention

Chronological age refers to the number of years a person has been alive, while biological age, also called physiological age, refers to how old a person appears to be. Because people age at different rates, it is difficult to determine a person’s biological age. Some people “look” and “feel” older than their chronological age, while others look younger than their chronological age. Although a person’s chronological age generally correlates with their biological age, this is not always the case. Biological age is a better indicator of a person’s health, well-being, and longevity than chronological age. As the equivalent of physiological age, biological age is a reflection of and influenced by several lifestyle factors, including diet, exercise, sleep habits, etc. However, estimating a person’s biological age remains challenging. It is important to note that the need to determine true biological age is driven by the idea that it may lead to the testing and development of intervention technologies that will slow the rate of biological aging. Over the past decades, considerable efforts have been made to identify various parameters that can predict biological aging and lifespan, such as frailty rates (Ferrucci et al., 2002), hair graying, skin aging (Yanai, Budovsky, Tacutu, & Fraifeld, 2011), and levels of different types of white blood cells. However, most of these markers have been found to offer no advantage over knowledge of chronological age.

More recently, advances in molecular biology have allowed the introduction of new molecular indices of aging. “Telomere length” (Monaghan, 2010) and “metabolic indices” (Hertel et al., 2016) have been used to predict biological age. However, although telomere length changes with age, the correlation between chronological age and telomere length has been weak, and the predictive power for life expectancy has been low. In addition, the method for measuring telomere length is technically complex, and technical errors make age estimation difficult. Another index that has been used is the “metabolic age index”, which measures various metabolites in urine (Hertel et al., 2016). This technology requires a sophisticated way of measuring various components of urine.

A paradigm shift in the search for markers of biological age occurred with Horvath’s discovery of the “epigenetic clock” (Horvath, 2013). This clock is based on measuring the DNA methylation clock at 353 CG positions in DNA. The degree of methylation of genes included in the methylation clock was found to correlate with an individual’s chronological age better than any previous measure, including telomere length and other measures of aging in hair, skin, frailty, etc. More importantly, although for most people the DNA methylation clock is very close to the chronological clock, for some people the clock runs faster than the chronological clock, so that an individual’s epigenetic age can be much older than their chronological clock. Recent studies have shown that such advances in the DNA methylation clock predict early death from a variety of causes. A recent analysis of 13 different studies involving 13,089 people found that the epigenetic clock could predict all-cause mortality independently of several risk factors, such as age, body mass index (BMI), education, smoking, physical activity, alcohol consumption, smoking status, and some comorbidities (Chen et al., 2016).

A recent review by Jylhävä, Pedersen, and Hägg in EBiomedicine concluded that: “Although telomere length is the most well-studied predictor of biological age, many new predictors are emerging, the epigenetic clock is currently the best predictor of biological age as it correlates well with age and predicts mortality (Jylhava, Pedersen, & Hagg, 2017). Technical error in measuring telomere length may also contribute to the lack of consistent results.” The authors summarize that “In summary, telomere length is well validated but has low predictive power. A composite biomarker is not well validated but may be a stronger predictor than telomere length, as is a metabolic age measure. The epigenetic clock currently performs best when both aspects are taken into account (Jylhava et al., 2017).”

A comparison of telomere length and the epigenetic clock as predictors of biological age in the 1,895-person Berlin Aging Study II by Valentin Max Wetteret et al. found that while “as described previously, telomere length was significantly shorter in the younger age group than in the older age group in the BASE-II cohort, telomere length and chronological age were only weakly negatively correlated in BASE-II (Rs 2 = 0.013).” In contrast, this study found that “our results showed a positive and significant correlation between DNA methylation age estimation (epigenetic clock) and chronological age (R2sRs2 = 0.47) that persisted after adjustment for covariates (sex, leukocyte distribution, alcohol consumption, and smoking).” The authors conclude that: “Thus, as expected, we found that DNA age is a much more accurate predictor of chronological age than telomere length (Vetter et al., 2018).”

In a Scottish study of two birth cohorts, a combined cohort analysis found that telomere length explained 2.8% of the variation in age, while the epigenetic clock explained 34.5% of the variation in age. In the same study, also in a combined cohort analysis, a one standard deviation increase in baseline epigenetic age was associated with a 25% increased mortality risk, while in the same model, a one standard deviation increase in baseline telomere length was independently associated with an 11% decreased mortality risk (P < 0.047) (Marioni et al., 2018).

Although it is becoming clear that the “epigenetic clock” is the most accurate indicator of biological age to date, the available tests require testing a large number of sites using blood, which is an invasive and expensive sample, which is not suitable for large-scale voluntary use by patients. Although the available methods are suitable for scientific and clinical research, they are not suitable for consumer-oriented use of this test. Thus, there is a need for an accurate, reliable, high-throughput, and non-invasive test.

The present invention provides a solution to this problem in the form of a system for integrating an accurate, reliable, saliva-based "EpiAging test" using novel CG sites within a health ecosystem for self-training, self-regulated healthy aging using consumer-performed, user-accessible, repeatable epigenetic clock testing integrated with a machine-readable medium, also called an application (App), that enables data collection and communication with the consumer, data sharing, and machine learning technologies. Current methods are expensive (require DNA methylation analysis for many CG sites across different regions of the genome), invasive (use blood), stand-alone, and provide no guidance for improving age estimates. Although the medical literature provides general advice on behaviors that have a positive impact on health, the exact personalized combination of lifestyle changes that may be beneficial for a specific individual is unknown. The present invention relates to a system that integrates a consumer-administered "DNA methylation age test" using saliva with an app-driven health and lifestyle management environment that combines data sharing, machine learning, and personalized interventions to ensure a healthy lifestyle managed by the consumer via the app. The data is completely private and is shared only with consumers and no other third parties. The incentive for the consumer to share data is that he/she receives better health advice by participating in the sharing community, so that the benefit of data sharing is delivered to the consumer dynamically and periodically by receiving higher quality lifestyle assessment and recommendations for improvement.

Objectives of the invention

The main objective of the present invention relates to a method for calculating the biological age of a subject, including the steps of extracting DNA from a substrate in a subject, measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile, analyzing the DNA methylation profile to obtain a polygenic score, and determining the biological age of the subject based on the polygenic score, characterized in that the DNA extraction includes extracting genomic DNA from saliva or blood obtained from the subject.

Another object of the present invention relates to a method for calculating the biological age of a subject based on a polygenic score obtained from a measured DNA methylation profile performed for polygenic DNA methylation biomarkers, which comprises measuring the methylation status of CG sites in any of the human CG sites and combinations thereof that are located in the putative antisense region for the ElovL2 gene, the ElovL2 AS1 region, as set forth in SEQ ID NO: 1.

Another object of the present invention relates to a method for calculating the biological age of multiple subjects, comprising the steps of extracting DNA from multiple substrates from multiple subjects, measuring DNA methylation in the extracted DNA from multiple substrates to obtain a DNA methylation profile, analyzing the DNA methylation profile to obtain a polygenic score, and determining the biological age of multiple subjects based on the polygenic score, characterized in that the DNA extraction includes extracting genomic DNA from saliva or blood obtained from multiple subjects.

Another object of the present invention relates to a kit for determining the biological age of a subject, comprising means and reagents for collecting and stabilizing a substrate from a subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, where the substrate is saliva or blood of the subject, and where the stabilization of the substrate is intended to send the collected substrate by mail for DNA extraction for measuring DNA methylation in DNA extracted from the substrate to obtain a DNA methylation profile of the subject to determine the biological age of the subject.

Another object of the present invention relates to a computer-implemented method for providing lifestyle change recommendations, the method comprising the following steps: evaluating a machine-readable storage medium account obtained by sharing user data from a subject, matching said account with a set for determining the biological age of the subject as obtained from said subject to determine the biological age of said subject, calculating the biological age of the subject using a method for calculating the biological age of the subject or a method for calculating the biological age for several subjects to obtain a calculated biological age, integrating said calculated biological age into a machine learning model for said subject by performing statistical analysis using the evaluation of the machine-readable storage medium account obtained by sharing user data from the subject to obtain a combined data report, preparing a dynamic report for said subject by analyzing the combined data report with a sequence of responses to a questionnaire obtained by sharing user data from said subject over time and comparing them with recommendations from national associations, and exchanging the dynamic a report on a machine-readable medium with the specified subject to provide recommendations for lifestyle changes.

An alternative object of the present invention relates to a method for developing a machine-readable medium, the method comprising the following steps: storing data obtained from multiple subjects, analyzing the stored data and building a model, where the step of storing data obtained from multiple users includes a cloud SQL database, where the step of analyzing the stored data includes a group selected from deep machine learning, reinforcement learning and machine learning or a combination thereof, and where the step of building a model includes matching the input data of the questionnaire and the difference between the age by DNA methylation and chronological age as output data, as well as other physiological and psychological output data, such as pain, blood pressure, BMI and mood.

The essence of the invention

Accordingly, the present invention relates to methods and materials suitable for assessing age progression, lifestyle effects, and providing personalized lifestyle recommendations for lifestyle changes based on the calculation of biological age by analyzing DNA methylation of CG sites or CG positions upstream of the gene encoding an antisense mRNA directed against the ElovL2 gene (ElovL2 AS1 region) in substrates from a single subject or from multiple subjects in DNA extracted from substrates including blood and saliva.

One embodiment of the present invention relates to a method for calculating the biological age of a subject, the method comprising the following steps: extracting DNA from a substrate obtained from the subject, measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile, analyzing the DNA methylation profile to obtain a polygenic score, and determining the biological age of the subject based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject.

Since it has been found in the present invention that age progression is strongly correlated with the methylation of CG positions or CG sites that are or are located in the region upstream of the gene encoding the antisense mRNA directed against the ElovL2 gene (referred to as the ElovL2 AS1 region), then another embodiment of the present invention relates to a method for calculating the biological age of a subject, characterized in that the method comprises a DNA methylation measurement step that is performed for polygenic DNA methylation biomarkers, which comprises measuring the methylation status of CG sites at any of the human CG sites as described in Table 1, which indicates the CG positions on human chromosome 6 as described herein, and combinations thereof, that are located in the putative antisense region for the ElovL2 gene, the ElovL2 AS1 region, as further set forth in SEQ ID NO: 1. It has been found in the present invention that sequencing of the target amplicon of this region revealed the aforementioned novel combination of 13 sites CG, as described in Table 1, which indicates the locations of CG on human chromosome 6, as described herein in the ElovL2 AS1 region, as indicated in SEQ ID NO: 1, the methylation of which was strongly correlated with biological age determined from saliva. The linear regression equation in the present invention identified the regression coefficients of these regions with age, where the combined weighted equation of these regions accurately predicts biological age.

One embodiment of the present invention relates to a method for calculating the biological age of multiple subjects, the method comprising the following steps: extracting DNA from multiple substrates from multiple subjects, measuring DNA methylation in the DNA extracted from the multiple substrates to obtain a DNA methylation profile, analyzing the DNA methylation profile to obtain a polygenic score, and determining the biological age of multiple subjects based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject. The present invention relates to a method that accurately measures DNA methylation age from saliva by determining DNA methylation at a polygenic array of CG sites as described in Table 1, which indicates the CG positions on human chromosome 6, as described herein, in the ElovL2 AS1 region, as indicated in SEQ ID NO: 1, in hundreds of individuals simultaneously, by sequential amplification with target-specific primers and then with barcoding primers and multiplex sequencing in a single Miseq next generation sequencing reaction, data extraction and methylation quantification. The present invention also relates to measuring methylation of said CG DNA methylation sites as described in Table 1, which indicates the CG positions on human chromosome 6, as described herein, in the ElovL2 AS1 region, as indicated in SEQ ID NO: 1. using pyrosequencing, methylation-specific PCR or digital PCR. The present invention discloses the calculation of a polygenic weighted methylation score that predicts age.

One embodiment of the present invention relates to a kit for determining the biological age of a subject, comprising means and reagents for collecting and stabilizing a substrate from a subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, where the substrate is saliva or blood of the subject, and where the stabilization of the substrate is intended to send the collected substrate by mail for DNA extraction for measuring DNA methylation in DNA extracted from the substrate to obtain a DNA methylation profile of the subject to determine the biological age of the subject.

One embodiment of the present invention relates to a computer-implemented method for providing lifestyle change recommendations, the method comprising the following steps: evaluating a machine-readable storage medium account obtained by sharing user data from a subject, matching said account with a set as disclosed herein and obtained from said subject to determine a biological age of said subject, calculating the biological age of the subject using a method for calculating the biological age of a subject as disclosed herein or a method for calculating biological age for multiple subjects to obtain a calculated biological age, integrating the calculated biological age into a machine learning model for said subject by performing statistical analysis using the evaluation of the machine-readable storage medium account to obtain a combined data report, preparing a dynamic report for said subject by analyzing the combined data report with a sequence of responses to a questionnaire obtained by sharing user data from said subject over time and comparing them with recommendations from national associations, and sharing the dynamic report on the machine-readable storage medium with said subject to provide lifestyle change recommendations lifestyle changes. Thus, the present invention relates to a computer-implemented method as disclosed herein, which is a novel process that combines repeated measurements of DNA methylation age in saliva to determine biological age with dynamic lifestyle changes using a computer-readable medium, alternatively referred to as an application, that manages these changes. Since only saliva is required to determine DNA methylation age, the disclosed method of the present invention uses consumer-initiated ordering of tests via a computer-readable medium or an application, by placing a saliva sample in a saliva collection kit, which is mailed to a laboratory for DNA extraction followed by DNA methylation analysis. Lifestyle changes are recorded in the Application; methylation data as well as lifestyle data are recorded in a database and are continuously and iteratively analyzed by machine learning programs such as neural networks.

One embodiment of the present invention relates to a method for developing a machine-readable medium, the method includes the following steps: storing data obtained from multiple subjects, analyzing the stored data and building a model, where the step of storing the data obtained from multiple users comprises a cloud SQL database, where the step of analyzing the stored data includes a group selected from deep machine learning, reinforcement learning and machine learning or a combination thereof, and where the step of building a model includes matching the input data of a questionnaire and the difference between the age by DNA methylation and chronological age as output data, as well as other physiological and psychological output data, such as pain, blood pressure, BMI and mood. Data shared by a plurality of consumers is continuously analyzed to build a model that relates the input changes in lifestyle to the output data of the difference between the age by DNA methylation or biological age determined by the method of the present invention and chronological age. The model disclosed herein is applied to personal data, and the model outputs recommendations for personal changes in lifestyle. The lifestyle inputs and DNA methylation age or biological age outputs as described here are measured repeatedly and used to further reinforcement learning with additional advice fed into consumer apps.

The present invention relates to methods that can be used by anyone skilled in the art to measure biological age and the relationship between lifestyle changes and DNA methylation age. DNA methylation markers (CGIDs), as described in Table 1, which indicates selected CG positions in the upstream region on human chromosome 6 in the newly discovered ElovL2 AS1 gene, as indicated in SEQ ID NO: 1, as described in the present invention, are suitable for consumer-initiated saliva-based tests for DNA methylation age or biological age, as well as for reporting and changing lifestyle parameters using a "common" application or machine-readable medium and a machine learning system. It is shown that the present invention can be used to measure "biological age" using a polygenic score based on the DNA methylation measurement methods described herein, which include sequencing a target amplicon of the ELOVL2 AS1 antisense region as set forth in SEQ ID NO: 1 disclosed in the present invention in hundreds of individuals or subjects simultaneously or using other DNA methylation measurement methods available to those skilled in the art, such as next-generation bisulfite sequencing, pyrosequencing, MeDip sequencing, Ion Torrent sequencing, Illumina 450 K chips and Epic microarrays, etc. The present invention also relates to the use of the present invention for integrating DNA methylation measurements into a comprehensive lifestyle change plan using the EpiAging application disclosed herein, which can be developed by anyone skilled in the art using open source programs and other programs such as Build Fire JS, Ionic, Appcelerator's Titanium SDK, Mobile angular UI and Siberian CMS. The data will be stored in a database such as MySQL on a cloud server such as Azure cloud or Amazon cloud, which can be operated by anyone skilled in the art. The data will be analyzed by a machine learning platform such as neural networks using open source programs such as Tensor flow or R-statistics, available to those skilled in the art. The present invention relates to the use of the present invention in providing consumers with dynamic "personalized" reports with recommendations on a combination of lifestyle changes that can affect their healthy aging. The present invention also relates to the use of the EpiAging DNA methylation test and application for measuring the impact of interventions on one's biological age by submitting saliva to measure DNA methylation age before and after recommended lifestyle changes.

Other objects, features and advantages of the present invention will become apparent to those skilled in the art from the following detailed description. However, it should be understood that the detailed description and specific examples, while indicating certain embodiments of the present invention, are given by way of illustration and not limitation. Many changes and modifications within the scope of the present invention can be made without departing from its spirit, and the invention includes all such modifications.

Brief description of the drawings

Fig. 1. Since DNA methylation at CG sites in the antisense region upstream of the ElovL2 gene, called the ElovL2 AS1 region, correlates with age. Shown is the IGV Browser view of the human genomic region around the CG sites and the position of two ElovL2 AS1 CG regions, namely cg16867657 and cg21572722, as described in Table 1 below. Pearson correlation between genome-wide CGID methylation states in blood cells on publicly available Illumina 450K arrays and age revealed that the top CG was cg16867657 with a Pearson mixed moment correlation coefficient of r = 0.934 (p = 0) and the adjacent site cg21572722 whose correlation coefficient with age was r = 0.81004 (p = 0), indicating that the methylation state of the detected CG sites correlates with age in the ElovL2 AS1 region. Examination of the genomic position of this CG revealed that it is a member of a sequence of 13 CGs (annotated) that reside in a previously uncharacterized region, the antisense region of the ElovL2 gene (ElovL2 AS1).

Fig. 2. CG sites of the ElovL2 AS1 region are strongly correlated with age in saliva. A correlation is shown between the methylation score of the weighted methylation levels of the CG sites in the ElovL2 AS1 region, namely cg16867657, cg21572722 located on chromosome 6 and cg09809672 located on chromosome 1 (see Table 1 for genomic positions) and age in publicly available Illumina 450K blood cell arrays (GSE40279 n = 656 and GSE2219, n = 60). The analysis shows a strong correlation between methylation and age in all age groups.

Figure 3. Correlation of methylation at ElovL2 AS1 CG sites (from saliva analysis) and age and comparison with Horvath epigenetic clock. The correlation of methylation of ElovL2 AS1 CG sites, namely cg1687657 and cg21572722, and age from saliva analysis are shown and compared with Horvath epigenetic clock. A. Correlation between the pooled methylation score of cg1687657 and cg21572722 (HKG) and age using saliva DNA methylation profiles from GSE78874. B. Correlation between the gold standard Horvath methylation clock using the same Illumina 450K data. C. Comparison of the accuracy of the two tests. The pooled score of these two sites has a lower standard error in predicted age than the gold standard Horvath clock.

Fig. 4. Age prediction using the 13 CG ElovL2 AS1 polygenic score from saliva. The applicability of the present invention is demonstrated. A. Age predictive methylation scores calculated using a linear regression equation predicting age as a function of weighted methylation levels of 1, 5, 6, 9 CG sites in the ElovL2 AS1 region (see Table 1 for genomic locations) using methylation levels of DNA from saliva of 65 individuals. The ElovL2 AS1 region, as described in Figure 1, was amplified from bisulfite-treated saliva DNA and subjected to multiplexed next-generation sequencing on a Miseq next-generation sequencer. B. Methylation scores predicting age calculated using the linear regression equation predicting age as a function of weighted methylation levels of CG sites 1, 2, 5, 6, 9 in the ElovL2 AS1 region (see Table 1 for genomic positions). C. Methylation scores predicting age calculated using the linear regression equation predicting age as a function of weighted methylation levels of CG sites 1, 2 in the ElovL2 AS1 region (see Table 1 for genomic positions). D. Methylation scores predicting age calculated using the linear regression equation predicting age as a function of weighted methylation levels of all CG sites in the ElovL2 AS1 region (see Table 1 for genomic positions). E. Comparison of the predictive value of different methylation scores. The equation that includes all 13 CG sites outperforms any other combination.

Fig.5. EpiAging application. The EpiAging application home page is displayed.

Fig. 6. Health ecosystem based on the EpiAging app. The EpiAging app is shown; and lifestyle management through the DATA-BASED RECOMMENDATIONS feature. At the center of the health ecosystem is the EpiAging app. The app allows consumers to learn about DNA methylation, biological aging, nutritional supplements. The app allows consumers to enter lifestyle data, cardiovascular health, mood, nutrition, sleep sex, and pain data. The app allows the customer to order a saliva test kit through the marketplace. The saliva test kit is delivered by mail and the buyer scans a barcode that assigns the buyer an ID that links the phone ID to the test ID. The buyer sends the saliva test kit to the lab by mail, pre-paid. The methylation age lab data and recurring lifestyle data are transmitted from the app and the lab to a programmed SQL database. Likewise, other consumers submit their lifestyle data and DNA methylation data to this database. A machine learning algorithm analyzes the data using deep learning and repeated inputs. A model is calculated that determines the weight of each account in determining the outcome. Personal data is analyzed by the model and lifestyle changes that are predicted to impact the outcome (DNA delta methylation vs. chronological age) are delivered to the app. The consumer makes a lifestyle change and orders a new saliva test, and the cycle repeats with iterative analysis and further recommendations based on the direction and range of change in DNA methylation age relative to chronological age.

Fig. 7. Health Epiecosystem. The health epiecosystem and its usefulness for healthy aging are depicted. The EpiAging app is at the center of the health ecosystem. It enables real-time delivery of health advice from the reputable National Medical Association. It creates a marketplace for health care and lifestyle product providers, as well as a marketplace for various new tests. The app sends data to a shared data server, which iteratively analyzes all the information and provides lifestyle recommendations, possible tests to take, and information on health care and product providers based on the analysis results.

Detailed description of the invention

In describing embodiments of the invention, reference may be made to the accompanying drawings, which form a part of the invention, and in which is shown by way of illustration a specific embodiment in which the invention may be practiced. It will be understood that other embodiments may be used and structural changes may be made without departing from the scope of the present invention. Many of the methods and procedures described or referred to herein are well understood and commonly used by those skilled in the art. Unless otherwise defined, all terms in the art, notations and other scientific terms or terminology used herein have the meanings commonly understood by those skilled in the art to which the present invention pertains. In some cases, terms with conventional meanings are defined herein for clarity and/or ease of reference, and the inclusion of such definitions in the present invention should not necessarily be construed as representing a substantial difference from what is commonly understood in the art.

All illustrations in the drawings are intended to describe selected embodiments of the present invention and are not intended to limit the scope of the present invention.

All publications cited in this disclosure are herein incorporated by reference to disclose and describe aspects, methods, and/or materials in connection with the cited publications.

DNA methylation refers to chemical modifications of the DNA molecule. It has been found that technology platforms such as the Illumina Infinium microarray or DNA sequencing-based technologies allow highly reliable and reproducible measurements of human DNA methylation levels. There are more than 28 million CpG loci or CG sites in the human genome. Therefore, unique identifiers are assigned to specific loci, such as those found in the Illumina CpG loci or CG site database. These CG locus designation identifiers are used in the present invention.

Definitions:

As used herein, the term "CG" refers to a dinucleotide sequence in DNA containing cytosine and guanosine bases. CG positions, referred to herein as "CG sites," are positions in the human genome defined by chromosome and nucleotide position in the hg19 human reference genome.

The term "beta value" as used herein refers to the calculation of the methylation level at a CGID position obtained by normalization and quantification of Illumina 450K or EPIC chips using the intensity ratio between methylated and unmethylated probes and the formula: beta value = methylated C intensity / (methylated C intensity + unmethylated C intensity) from 0 to 1, where 0 is completely unmethylated and 1 is completely methylated.

As used here, the term "decision trees" is a type of data mining algorithm that selects from a set of variables and interactions between variables those that are most predictive of the response or outcome to be explained (Mann et al., 2008).

The term random forests, as used here, is a type of data mining algorithm that can select the most important variables to determine a given outcome or answer (Shi, Seligson, Belldegrun, Palotie, & Horvath, 2005; Svetnik et al., 2003).

As used here, the term lasso regression is a variable selection method for linear regression models that identifies the minimum subset of predictors needed to predict the outcome (response variable) with minimal prediction error (Kim, Kim, Jeong, Jeong, & Kim, 2018).

As used here, the term K-means clustering is an unsupervised machine learning algorithm that partitions observations into a smaller set of clusters, where each observation belongs to one cluster (Beauchaine & Beauchaine, 2002; Kakushadze & Yu, 2017).

The term reinforcement learning used here refers to the feedback from data analysis and trial-and-error learning. A sequence of successful decisions will lead to reinforcement of the process (Zhao, Kosorok, & Zeng, 2009).

The term “penalized regression” as used herein refers to a statistical method that aims to identify the smallest number of predictors needed to predict an outcome from a larger list of biomarkers, as implemented, for example, in the R statistical package, “penalized” as described in Goeman, J. J., L1 penalized estimation in the Cox proportional hazards model. Biometrical Journal 52(1), 70–84.

As used here, the term "clustering" refers to the grouping of a set of objects such that objects in one group (called a cluster) are more similar (in some sense) to each other than to objects in other groups (clusters).

The term “neural networks and deep learning” as used here refers to a machine learning method that involves neural networks at multiple layers to iteratively learn from data. A neural network views various inputs, such as lifestyle variables, as a collection of connected units or nodes called artificial neurons that have multiple interactions, like neurons in the brain (De Roach, 1989; Mupparapu, Wu, & Chen, 2018; Sherbet, Woo, & Dlay, 2018). These interactions determine the outcome in the form of biological aging, which is measured as accelerated or decelerated relative to chronological age.

The term “multiple or polygenic linear regression” as used herein refers to a statistical method that evaluates the relationship between multiple “independent variables” or “predictors,” such as the percentage of methylation at several CG identifiers, and a “dependent variable,” such as chronological age. This method determines the “weight” or coefficient of each CG identifier in predicting the “outcome” (the dependent variable, such as age) when multiple “independent variables,” such as CG identifiers, are included in the model.

As used here, the term “Pearson correlation” refers to a statistical method that estimates the correlation between an “independent variable” or “predictor,” such as CG methylation percentage, and a “dependent variable,” such as chronological age. The Pearson product moment correlation coefficient r quantitatively weights the correlation between 0, indicating no correlation, and 1, indicating perfect correlation (Hardy & Magnello, 2002).

The method disclosed in the present invention is based on the discovery of sites in the human genome whose methylation state correlates with age, which were discovered by performing a series of Pearson correlations between age and DNA methylation on 450K sites in the genome in publicly available datasets (GSE61496, GSE98876) and confirmed the DNA methylation markers thus discovered using data from GSE40729. The analysis identified cg16867637 as the top site correlating with age (r = 0.934827, p = 0). In the present invention, a fragment of the human genome on chromosome 6 has been discovered that is an antisense sequence for the previously described age-related gene ElovL2, the ElovL2 AS1 region that contains 13 CGs, referred to herein as CG sites, are dinucleotide sequence positions as described herein below in Table 1, whose combined methylation scores, calculated in a multiple linear regression equation, provide a polygenic estimate of biological age from saliva with higher accuracy than using previously discovered genomic positions. These sites have not been previously described because they were not included in the Illumina chips. Thus, the present invention relates to new CG sites whose combined weighted methylation levels correlate with age. The present invention further demonstrates herein that methylation of all 13 CG sites can be accurately measured by amplifying a single amplicon and using indexed next generation sequencing to measure simultaneously in hundreds of individuals, reducing costs and increasing throughput using the EpiAging assay described herein. A subject or consumer orders a saliva collection kit, spits into a collection tube, and ships the kit back to the lab where the DNA is extracted, bisulfite treated, and the ElovL2 AS1 region amplified and indexed. Amplicons from 200 subjects are sequenced in a single Miseq reaction. The FastQ files are analyzed and the methylation level of the 13 CG sites is determined. Using an equation that relates the weighted methylation values of the 13 CG sites to age, a biological age is calculated and reported to the customer.

The present invention further relates to and contemplates the use of the methods disclosed herein in dynamically calculating biological age to improve healthy aging of a consumer by recommending lifestyle changes. The present invention shows that effective interventions can be recommended based on "machine learning" of the relationship between multiple lifestyle variables and the difference between DNA methylation age and biological changes. Because lifestyle data and methylation age or biological age data are dynamically collected from multiple subjects/users, the machine learns how a combination of changes in lifestyle parameters is associated with an increase or decrease in the difference between DNA methylation and biological age. The present invention combines a DNA methylation test with subject/consumer-oriented data sharing, learning, and lifestyle changes. The subject/consumer orders and communicates their lifestyle decisions using the EpiAging application or a machine-readable medium as disclosed herein in the present invention. Improving health is a two-way partnership and collaborative effort, not a one-way flow of instructions from an “educated, omniscient” health professional (healthcare provider) to a “compliant” and passive patient (healthcare consumer). The best scientific advice from the most reputable national medical associations is presented to the consumer via the EpiAging app. The consumer decides which recommendations to act on. The consumer shares their decisions via a “fully blind” app. The consumer receives an identifier linked to their mobile ID but “separated” from personal information such as address, name, email, etc. The results of lifestyle actions for one and more users are re-analyzed, just like the DNA methylation age test results, combining both physical and mental results. The data is analyzed using state-of-the-art machine learning algorithms such as neural networks, for example using Tensor Flow. A model is created that maps various input parameters to an output delta between DNA methylation age and chronological age. The model uses the user's personal data, and the change suggestions are personalized and delivered to the consumer. The present invention relates to a large platform for creating an ever-evolving dynamic model of scientifically based recommendations. The present invention provides an "evolutionary" platform that dynamically improves using an ever-expanding array of data. The consumer's well-being and learning environment simultaneously evolve in a dynamic interaction between DNA methylation tests, lifestyle modifications, common data, and machine learning, as disclosed in the present invention.

The present invention has several embodiments. In one aspect of the present invention, the present invention relates to polygenic DNA methylation markers of biological age for lifestyle management in healthy aging, said set of polygenic DNA methylation markers being obtained using Pearson correlation analysis between age and genome-wide DNA methylation data obtained using mapping methods such as Illumina 450K or 850K arrays, whole-genome bisulfite sequencing using various next-generation sequencing platforms, methylated DNA immunoprecipitation (MeDIP) sequencing or hybridization to oligonucleotide arrays, or a combination of these methods.

In one embodiment, the present invention relates to a method for calculating the biological age of a subject, the method comprising the steps of: (a) extracting DNA from a substrate obtained from the subject; (b) measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and (d) determining the biological age of the subject based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject. The present invention discloses a method for calculating the biological age of a subject, wherein the measurement of DNA methylation is performed using methods including DNA pyrosequencing, mass spectrometry (Epityper™), PCR-based methylation analysis, next-generation bisulfite sequencing of a target amplicon on a platform selected from the group of HiSeq, MiniSeq, MiSeq and NextSeq sequencers, Ion Torrent sequencing, methylated DNA immunoprecipitation (MeDIP) sequencing or hybridization to oligonucleotide arrays. A method for calculating the biological age of a subject disclosed herein, wherein the DNA methylation measurement is performed for polygenic DNA methylation biomarkers, which comprises measuring the methylation status of CG sites in any of the human CG sites and combinations thereof, which are the positions of the dinucleotide sequences described below in Table 1, as disclosed in the present invention, which are located in the antisense region of the ElovL2 gene in human chromosome 6, the ElovL2 AS1 region, as set forth in SEQ ID NO: 1

(CGCCCTCGCGTCCGCGGCGTCCCCTGCCGGCCGGGCGGCGATTTGCAGGTCCAGCCGGCGCCGGTTTCGCGCGGCGGCTCAACGTCCACGGAGCCCCAGGAATACCCACCCGCTGCCCAGATCGGCAGCCGCTGCTGCGGGGAGAAGCAGTATCGTGCAGGGCGGCACGCTGGTCTTGCTTACAGTTGGGCTTCGGTGGGTTTGAAGCACACATTAGGGGGAAATGGCTCTGTTCC TGCAGGTTTGCGCAGTCTGGGTTTCTTAG).

Table 1: Positions containing CG methylation sites (CG sites) corresponding to the 13 CG sites upstream of the ElovL2 antisense gene, ElovL2 AS1 region, as set forth in SEQ ID NO: 1, and useful in embodiments of the present invention.

The position in the human genome of the selected 13 CG dinucleotide in the CG sites used in the various embodiments of the present invention is indicated in Table 1 included herein, which also indicates the position of CG in chromosome 1 as shown in the Figures and Examples of the present application.

Illumina 450K CGID Human chromosome Start of position End of position After-
valency cg16867657 Chromosome 6 11044877 11044878 CG cg21572722 Chromosome 6 11044894 11044895 CG n.d.* Chromosome 6 11044861 11044862 CG n.d.* Chromosome 6 11044864 11044865 CG n.d.* Chromosome 6 11044867 11044868 CG n.d.* Chromosome 6 11044873 11044874 CG n.d.* Chromosome 6 11044875 11044876 CG n.d.* Chromosome 6 11044880 11044881 CG n.d.* Chromosome 6 11044888 11044889 CG n.d.* Chromosome 6 11044916 11044917 CG n.d.* Chromosome 6 11044928 11044929 CG n.d.* Chromosome 6 11044935 11044936 CG n.d.* Chromosome 6 11044943 11044944 CG cg09809672 Chromosome 1 236557682 236557683 CG

n.d. means no data

In one embodiment, the present invention relates to a method for calculating the biological age of a subject, the method comprising the steps of: (a) extracting DNA from a substrate obtained from the subject; (b) measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and (d) determining the biological age of the subject based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject, wherein the DNA methylation measurement is performed using DNA pyrosequencing using primers as set forth in SEQ ID NO: 2 (AGGGGAGTAGGGTAAGTGAG) for the forward biotinylated primer, SEQ ID NO: 3 (ACCATTTCCCCCTAATATATACTT) for the reverse primer, and SEQ ID NO: 4 (GGGAGGAGATTTGTAGGTTT) for the pyrosequencing primer.

In one embodiment, the present invention relates to the use of DNA methylation analysis by pyrosequencing for DNA methylation age determination using the ElovL2 AS1 region comprising CG sites and combinations thereof representing dinucleotide sequence positions as described in Table 1 as disclosed in the present invention, using primers as described in the present invention and standard pyrosequencing reaction conditions recommended by the manufacturer (Pyromark, Qiagen), wherein the primers comprise a forward (biotinylated) primer as set forth in SEQ ID NO: 2, an Elovl2_Rv primer as set forth in SEQ ID NO: 3, and an Elovl2_Seq primer as set forth in SEQ ID NO: 4.

In one embodiment, the present invention relates to a method for calculating the biological age of a subject, the method comprising the steps of: (a) extracting DNA from a substrate obtained from the subject; (b) measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and (d) determining the biological age of the subject based on a polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject, wherein the DNA methylation measurement is performed using next generation bisulfite sequencing of the target amplicon on a platform selected from the group of HiSeq, MiniSeq, MiSeq and NextSeq sequencers, with primers set forth in SEQ ID NO: 5 (ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNYGGGYGGATTTGTAGGTTTAGT) for the forward primer and SEQ ID NO: 6 (GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTCCCTACACRATACTACTTCCCCC) for the reverse primer.

In one embodiment, the present invention relates to the use of DNA methylation analysis with bisulfite sequencing of a polygenic multiplex amplicon for DNA methylation age determination in saliva using the ElovL2 AS1 region containing CG sites and combinations thereof representing dinucleotide sequence positions as described in Table 1 as disclosed in the present invention, using primers as disclosed in the present invention and standard conditions that include bisulfite conversion, sequential amplification comprising the use of: (a) target-specific primers (PCR 1) and (b) barcoding primers (PCR 2), and multiplex sequencing on a single Miseq next-generation sequencer (Illumina), demultiplexing using Illumina software, data extraction and methylation quantification using standard methylation assays including Methylkit, followed by calculation weighted DNA methylation score for calculating the biological age of a subject, wherein the target-specific primers (PCR 1) have the sequences set forth in SEQ ID NO: 5 (ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNYGGGYGGYGATTTGTAGGTTTAGT) for the forward primer and SEQ ID NO: 6 (GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTCCCTACACRATACTACTTCTCCCC) for the reverse primer, and wherein the barcoding primers (PCR 2) have the sequences set forth in SEQ ID NO: 7 (AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGAC) for the forward primer and SEQ ID NO: 8 (CAAGCAGAAGACGGCATACGAGATAGTCATCGGTGACTGGAGTTCAGACGTG) for the reverse primer, which is barcode index primer. In the sequence SEQ ID NO: 5 disclosed herein ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNYGGGYGGYGATTTGTAGGTTTAGT, the black bases are adapters and the red bases are target sequences. In the sequence SEQ ID NO: 6 disclosed herein GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTCCCTACACRATACTACTTCTCCCC, the black bases are adapters and the red bases are target sequences. In the sequence SEQ ID NO: 8 disclosed herein of the barcoding primer CAAGCAGAAGACGGCATACGAGATAGTCATCGGTGACTGGAGTTCAGACGTG, the red bases are the index; up to 200 variants of this index are used.

In one embodiment, the present invention relates to a method for calculating the biological age of a subject, the method comprising the steps of: (a) extracting DNA from a substrate obtained from the subject; (b) measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and (d) determining the biological age of the subject based on the polygenic score, wherein the DNA methylation measurement is performed using PCR-based methylation assays selected from the group of methylation-specific PCR and digital PCR.

In one embodiment, the present invention relates to a method for calculating the biological age of a subject, the method comprising the steps of: (a) extracting DNA from a substrate obtained from the subject; (b) measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and (d) determining the biological age of the subject based on the polygenic score, wherein analyzing the DNA methylation profile to obtain the polygenic score comprises using multiple linear regression equations or neural network analysis.

In one embodiment, the present invention relates to a method for calculating the biological age of multiple subjects, the method comprising the steps of: (a) extracting DNA from multiple substrates obtained from multiple subjects; (b) measuring DNA methylation in the DNA extracted from the multiple substrates to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and determining the biological age of multiple subjects based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject.

In one embodiment, the present invention relates to a method for calculating the biological age of multiple subjects, the method comprising the steps of: (a) extracting DNA from multiple substrates obtained from multiple subjects; (b) measuring DNA methylation in the DNA extracted from the multiple substrates to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and determining the biological age of multiple subjects based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject, and wherein measuring DNA methylation in the DNA extracted from the multiple substrates comprises the steps of: (i) amplifying the genomic DNA extracted from the multiple substrates with target-specific primers to obtain a PCR product 1; (ii) amplifying the PCR product 1 of step (i) with barcoding primers to obtain a PCR product 2; (iii) performing multiplex sequencing in a single Miseq next-generation sequencing reaction using the PCR product from step 2 of (ii); (iv) extracting data from the multiplex sequencing of step (iii); and quantifying DNA methylation from the extracted data of step (iv) to obtain a DNA methylation profile for each substrate. In an alternative embodiment, the present invention relates to a method for calculating the biological age of multiple subjects, wherein the target-specific primers for producing the PCR product 1 primers have the sequences set forth in SEQ ID NO: 5 (ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNYGGGYGGYGATTTGTAGGTTTAGT) for the forward primer and SEQ ID NO: 6 (GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTCCCTACACRATACTACTTCTCCCC) for the reverse primer, and wherein the barcoding primers for producing the PCR product 2 include primers having the sequences set forth in SEQ ID NO: 7 (AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGAC) for the forward primer and SEQ ID NO: 8 (CAAGCAGAAGACGGCATACGAGATAGTCATCGGTGACTGGAGTTCAGACGTG) for the reverse primer, which is the barcode index primer.

In one embodiment, the present invention relates to a combination of DNA methylation biomarkers for calculating biological age, wherein the combination of DNA methylation biomarkers comprises human CG sites and combinations thereof that are located in the putative antisense region of the ElovL2 gene, the ElovL2 AS1 region, as set forth in SEQ ID NO: 1. In one embodiment, the present invention identifies 13 CG sites located in the putative antisense region of the ElovL2 gene, the ElovL2 AS1 region, as set forth in SEQ ID NO: 1, which can be used alone or in combination to measure biological age. In one embodiment, the present invention relates to the use of CG sites, and combinations thereof, that are positions of a dinucleotide sequence as described in Table 1, as disclosed in the present invention.

In one embodiment, the present invention relates to a kit for determining the biological age of a subject, comprising means and reagents for collecting and stabilizing a substrate from a subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, wherein the substrate is saliva or blood of the subject, and wherein the stabilization of the substrate is for sending the collected substrate by mail for DNA extraction for measuring DNA methylation in DNA extracted from the substrate to obtain a DNA methylation profile of the subject to determine the biological age of the subject. In an alternative embodiment, the present invention relates to a kit for determining the biological age of a subject, wherein the kit is a saliva collection kit that a consumer or subject orders, spits saliva into a collection tube from the kit, which is sent to a laboratory for DNA extraction followed by DNA methylation analysis. In one embodiment, the present invention relates to a kit for collecting customer saliva samples, comprising means and reagents for collecting and stabilizing customer saliva. In one embodiment, the present invention relates to a kit comprising means and reagents for measuring DNA methylation of CG sites, and combinations thereof, representing positions of the dinucleotide sequence as described in Table 1, as disclosed in the present invention.

In one embodiment, the present invention relates to an application (App) for managing the order, sending, receiving of test results and lifestyle management in DNA methylation age testing. In one embodiment, the application is developed using open source development tools for storing test information, a virtual shopping cart for ordering a test, a scanning function for scanning a saliva kit barcode and a function for receiving test results from a laboratory. In one embodiment, the present invention relates to questionnaires for inclusion in the application that will examine lifestyle functions that can affect "healthy aging", these include vital signs, weight, height, blood pressure, heart rate, etc., self-assessment of mood, McGill pain severity questionnaire, diet and nutrition questionnaire, exercise questionnaire and questions on lifestyle, such as alcohol, drugs and smoking. In one embodiment of the invention, the method includes performing a statistical analysis of the survey responses and providing a dynamic report to consumers in the Application that describes the evolution of the survey responses over time compared to recommendations from national associations, such as associations studying cancer, heart disease and stroke, and diabetes.

In one embodiment, the present invention relates to storing data obtained from multiple users in a cloud-based SQL database and using "machine learning" to analyze the data and build a model that relates questionnaire input measurements and the difference between DNA methylation age and chronological age as output, as well as other physiological and psychological outputs such as pain, blood pressure, BMI, and mood. In a further embodiment of the present invention, as disclosed herein, the machine learning is selected from the group of "deep machine learning" data mining algorithms, including neural networks, or "reinforcement learning" through consumer feedback used to enhance the most effective lifestyle changes, or a "machine learning" data mining algorithm, including "random forest" analysis, or a "machine learning" data mining algorithm, including K-means cluster analysis, or a "machine learning" platform, including Amazon Machine Learning (AML), or "machine learning" programs, including H2O.ai products on platforms such as Apache Hadoop distributed file system, Amazon EC2 Google Compute Engine and Microsoft Azure.

In one embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, the method comprising the steps of: (a) evaluating an account on a computer-readable medium obtained by sharing user data from a subject; (b) associating the account of step (a) with a kit for determining the biological age of the subject, comprising means and reagents for collecting and stabilizing a substrate obtained from the subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, wherein the substrate is saliva or blood of the subject, and wherein the stabilization of the substrate is for sending the collected substrate by mail for extracting DNA therefrom for measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile of the subject to determine the biological age of said subject; (c) calculating the biological age of the subject using the method comprising: (i) extracting DNA from the substrate obtained from the subject; (ii) measuring DNA methylation in DNA extracted from a substrate to obtain a DNA methylation profile; (iii) analyzing the DNA methylation profile to obtain a polygenic score; and (iv) determining the biological age of the subject from the polygenic score, wherein the DNA extraction comprises isolating genomic DNA from saliva or blood obtained from the subject to obtain an estimated biological age; (d) integrating the estimated biological age of step (c) into a machine learning model for said subject by performing a statistical analysis using the score of step (a) to obtain an integrated data report; (e) preparing a dynamic report for said subject by analyzing the integrated data report of step (d) with a sequence of questionnaire responses obtained by sharing user data from said subject over time and comparing them to national association recommendations; and (f) sharing the dynamic report of step (e) on a machine-readable medium with said subject to provide lifestyle change recommendations. In a further embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, wherein the computer-readable medium comprises open source development tools comprising information about a test for calculating biological age based on the method disclosed herein, a virtual shopping cart for ordering said test, a scanning function for scanning a barcode of a kit for determining the biological age of a subject as disclosed herein, a function for receiving test results from a laboratory, wherein the open source development tools include questionnaires included in the computer-readable medium for studying lifestyle functions that influence healthy aging, including vital signs, weight, height, blood pressure, heart rate, self-rated mood, the McGill pain questionnaire, a diet and nutrition questionnaire, an exercise questionnaire, and questions about lifestyle, including alcohol, drug use, and smoking, as well as a combination thereof.

In an alternative embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, the method comprising the steps of: (a) evaluating an account on a computer-readable medium obtained by sharing user data from a subject; (b) associating the account of step (a) with a kit for determining the biological age of the subject, comprising means and reagents for collecting and stabilizing a substrate obtained from the subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, wherein the substrate is saliva or blood of the subject, and wherein the stabilization of the substrate is for sending the collected substrate by mail for extracting DNA therefrom for measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile of the subject to determine the biological age of said subject; (c) calculating the biological age of the subject using a method comprising: (i) extracting DNA from multiple substrates obtained from multiple subjects; (ii) measuring DNA methylation in extracted DNA from multiple substrates to obtain a DNA methylation profile; (iii) analyzing the DNA methylation profile to obtain a polygenic score; (iv) determining the biological age of multiple subjects based on the polygenic score, wherein the DNA extraction comprises isolating genomic DNA from saliva or blood obtained from the subject to obtain an estimated biological age; (d) integrating the estimated biological age of step (c) into a machine learning model for said subject by performing a statistical analysis using the score of step (a) to obtain a combined data report; (e) preparing a dynamic report for said subject by analyzing the combined data report of step (d) with a sequence of questionnaire responses obtained by sharing user data from said subject over time and comparing them to national association recommendations; and (f) sharing the dynamic report of step (e) on a machine-readable medium with said subject to provide lifestyle change recommendations. In a further embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, wherein the computer-readable medium comprises open source development tools comprising information about a test for calculating biological age based on the method disclosed herein, a virtual shopping cart for ordering said test, a scanning function for scanning a barcode of a kit for determining the biological age of a subject as disclosed herein, a function for receiving test results from a laboratory, wherein the open source development tools include questionnaires included in the computer-readable medium for studying lifestyle functions that influence healthy aging, including vital signs, weight, height, blood pressure, heart rate, self-rated mood, the McGill pain questionnaire, a diet and nutrition questionnaire, an exercise questionnaire, and questions about lifestyle, including alcohol, drug use, and smoking, as well as a combination thereof.

In yet another alternative embodiment, the present invention relates to a computer-implemented method of providing lifestyle change recommendations, the method comprising the steps of: (a) evaluating an account on a computer-readable medium obtained by sharing user data from a subject; (b) associating the account of step (a) with a kit for determining the biological age of the subject, the kit comprising means and reagents for collecting and stabilizing a substrate obtained from the subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, wherein the substrate is saliva or blood of the subject, and wherein the stabilization of the substrate is for sending the collected substrate by mail for extracting DNA therefrom for measuring DNA methylation in the DNA extracted from the substrate for obtaining a DNA methylation profile of the subject for determining the biological age of said subject; (c) calculating the biological age of the subject using a method comprising: (i) extracting DNA from multiple substrates obtained from multiple subjects; (ii) measuring DNA methylation in extracted DNA from multiple substrates to obtain a DNA methylation profile; (iii) analyzing the DNA methylation profile to obtain a polygenic score; (iv) determining the biological age of multiple subjects based on the polygenic score, wherein the DNA extraction comprises isolating genomic DNA from saliva or blood obtained from the subject to obtain an estimated biological age, wherein measuring DNA methylation in the extracted DNA from the multiple substrates comprises the steps of: (1) amplifying the extracted genomic DNA from the multiple substrates with target-specific primers to obtain PCR product 1; (2) amplifying the PCR product of step 1 of (1) with barcoding primers to obtain PCR product 2; (3) performing multiplex sequencing in a single Miseq next-generation sequencing reaction using the PCR product of step 2 of (2); (4) extracting data from the multiplex sequencing of step (3); (5) quantifying DNA methylation from the extracted data in step (d) to obtain a DNA methylation profile for each substrate; (d) integrating the calculated biological age of step (c) into a machine learning model for said subject by performing a statistical analysis using the assessment of step (a) to obtain a combined data report; (e) preparing a dynamic report for said subject by analyzing the combined data report of step (d) with a sequence of questionnaire responses obtained by sharing user data from said subject over time and comparing them with national association recommendations; and (f) sharing the dynamic report of step (e) on a machine-readable medium with said subject to provide lifestyle change recommendations. In a further embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, wherein the computer-readable medium comprises open source development tools comprising information about a test for calculating biological age based on the method disclosed herein, a virtual shopping cart for ordering said test, a scanning function for scanning a barcode of a kit for determining the biological age of a subject as disclosed herein, a function for receiving test results from a laboratory, wherein the open source development tools include questionnaires included in the computer-readable medium for studying lifestyle functions that influence healthy aging, including vital signs, weight, height, blood pressure, heart rate, self-rated mood, the McGill pain questionnaire, a diet and nutrition questionnaire, an exercise questionnaire, and questions about lifestyle, including alcohol, drug use, and smoking, as well as a combination thereof.

In one embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, the method comprising the steps of: (a) evaluating an account on a computer-readable medium obtained by sharing user data from a subject; (b) associating the account of step (a) with a kit for determining the biological age of a subject, comprising means and reagents for collecting and stabilizing a substrate obtained from the subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, wherein the substrate is saliva or blood of the subject, and wherein the stabilization of the substrate is for sending the collected substrate by mail for extracting DNA therefrom for measuring DNA methylation in the DNA extracted from the substrate for obtaining a DNA methylation profile of the subject for determining the biological age of said subject; (c) calculating the biological age of the subject using the method for calculating the biological age of a subject as disclosed in the present invention; (d) integrating the calculated biological age of step (c) into a machine learning model for said subject by performing statistical analysis using the assessment of step (a) to obtain a combined data report; (e) preparing a dynamic report for said subject by analyzing the combined data report of step (d) with a sequence of questionnaire responses obtained by sharing user data from said subject over time and comparing them with the recommendations of the national association; and (f) sharing the dynamic report of step (e) on a machine-readable medium with said subject to provide lifestyle recommendations, where the method includes using an Android, Apple or WeChat widget for personalized lifestyle recommendations, creating a health ecosystem focused on normalizing or slowing down the biological aging of the subject, or storing data in an object store on a server or cloud server, including Amazon, Ali Cloud or Microsoft Azure using a standard data pipeline and management systems such as Cloud dataprep, across multiple subjects. In a further embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, wherein the method includes using a set of artificial intelligence algorithms such as a random forest (RF), a support vector machine (SVM), a linear discriminant analysis (LDA), a generalized linear model (GLM) and deep learning (DL) to calculate a weighted contribution of various lifestyle indicators to the biological age of a subject or several subjects, which is dynamically updated to provide personalized lifestyle change recommendations.

In one embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, the method comprising the steps of: (a) evaluating an account on a computer-readable medium obtained by sharing user data from a subject; (b) associating the account of step (a) with a kit for determining the biological age of the subject, comprising means and reagents for collecting and stabilizing a substrate obtained from the subject; a scanner for reading a barcode in the kit; and instructions for collecting and stabilizing the substrate, wherein the substrate is saliva or blood of the subject, and wherein the stabilization of the substrate is for sending the collected substrate by mail for extracting DNA therefrom for measuring DNA methylation in the DNA extracted from the substrate for obtaining a DNA methylation profile of the subject for determining the biological age of said subject; (c) calculating the biological age of the subject using the method for calculating the biological age of several subjects as disclosed in the present invention; (d) integrating the calculated biological age of step (c) into a machine learning model for said subject by performing statistical analysis using the assessment of step (a) to obtain a combined data report; (e) preparing a dynamic report for said subject by analyzing the combined data report of step (d) with a sequence of questionnaire responses obtained by sharing user data from said subject over time and comparing them with the recommendations of the national association; and (f) sharing the dynamic report of step (e) on a machine-readable medium with said subject to provide lifestyle recommendations, where the method includes using an Android, Apple or WeChat widget for personalized lifestyle recommendations, creating a health ecosystem focused on normalizing or slowing down the biological aging of the subject, or storing data in an object store on a server or cloud server, including Amazon, Ali Cloud or Microsoft Azure using a standard data pipeline and management systems such as Cloud dataprep, across multiple subjects. In a further embodiment, the present invention relates to a computer-implemented method for providing lifestyle change recommendations, wherein the method includes using a set of artificial intelligence algorithms such as a random forest (RF), a support vector machine (SVM), a linear discriminant analysis (LDA), a generalized linear model (GLM) and deep learning (DL) to calculate a weighted contribution of various lifestyle indicators to the biological age of a subject or several subjects, which is dynamically updated to provide personalized lifestyle change recommendations.

In one embodiment, the present invention relates to a method for developing a machine-readable medium, the method comprising the following steps: (a) storing data obtained from multiple subjects; (b) analyzing the stored data of step (a); and (c) building a model, where the step of storing data obtained from multiple users includes a cloud SQL database, where the step of analyzing the stored data includes a group selected from deep machine learning, reinforcement learning and machine learning or a combination thereof, and where the step of building the model includes matching the input data of the questionnaire and the difference between the age by DNA methylation and chronological age as the output data, as well as other physiological and psychological output data, such as pain, blood pressure, BMI and mood. In a further embodiment, the present invention relates to a method for developing a machine-readable medium, where the machine learning includes a group selected from a data mining algorithm including random forest analysis or a data mining algorithm including K-means cluster analysis, or a platform including Amazon Machine Learning (AML) or software including H2O.ai products on platforms including Apache Hadoop distributed file system, Amazon EC2 Google Compute Engine and Microsoft Azure, or a combination thereof.

In one embodiment, the present invention relates to a method of calculating the biological age of a subject, the method comprising the steps of: (a) extracting DNA from a substrate obtained from the subject; (b) measuring DNA methylation of DNA extracted from the substrate to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and (d) determining the biological age of the subject based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject for use in a method of assessing the effect of a biological intervention on the biological age of the subject, the method comprising the steps of: (i) calculating the biological age of the subject using the method disclosed herein to obtain an initial biological age prior to the biological intervention; (ii) performing a biological intervention on said subject; (iii) repeating step (i) on a further substrate obtained from said subject after step (ii) is performed to obtain a biological age following the biological intervention; (iv) integrating the biological age following the biological intervention into a machine learning model for said subject to estimate the effect of the biological intervention on the biological age of said subject, wherein the biological intervention in step (ii) is selected from the group consisting of dietary supplements, vitamins, therapy, administration of a test substance, dietary manipulation, metabolic manipulation, surgical manipulation, social manipulation, behavioral manipulation, environmental manipulation, sensory manipulation, hormonal manipulation, and epigenetic manipulation, or combinations thereof, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject, and wherein the integration of the biological age following the biological intervention into the machine learning model for said subject comprises the biological age estimated in step (iii) and physiological parameters obtained by sharing user data from said subject.

In one embodiment, the present invention relates to a method of calculating the biological age of multiple subjects, the method comprising the steps of: (a) extracting DNA from multiple substrates obtained from multiple subjects; (b) measuring DNA methylation of DNA extracted from the multiple substrates to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and determining the biological age of the multiple subjects based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject for use in a method of assessing the effect of a biological intervention on the biological age of the subject, the method comprising the steps of: (i) calculating the biological age of the subject using the method disclosed herein to obtain an initial biological age prior to the biological intervention; (ii) performing a biological intervention on said subject; (iii) repeating step (i) on a subsequent substrate obtained from said subject after step (ii) is performed to obtain a biological age following the biological intervention; (iv) integrating the biological age following the biological intervention into a machine learning model for said subject to estimate the effect of the biological intervention on the biological age of said subject, wherein the biological intervention in step (ii) is selected from the group consisting of dietary supplements, vitamins, therapy, administration of a test substance, dietary manipulation, metabolic manipulation, surgical manipulation, social manipulation, behavioral manipulation, environmental manipulation, sensory manipulation, hormonal manipulation, and epigenetic manipulation, or combinations thereof, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject, and wherein the integration of the biological age following the biological intervention into the machine learning model for said subject comprises the biological age estimated in step (iii) and physiological parameters obtained by sharing user data from said subject.

In one embodiment, the present invention relates to a method of calculating the biological age of a subject, the method comprising the steps of: (a) extracting DNA from a substrate obtained from the subject; (b) measuring DNA methylation of DNA extracted from the substrate to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and (d) determining the biological age of the subject based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject for use in a method of screening an agent for whether the agent has anti-aging activity, the method comprising the steps of: (i) calculating age from a substrate obtained from the subject using a method disclosed in the present invention to obtain an initial biological age prior to biological intervention; (ii) administering a test agent to said subject; (iii) repeating step (i) on a further substrate obtained from said subject after performing step (ii) to determine the biological age after administration of the test agent; (iv) integrating the biological age after administration of the test agent into a machine learning model for said subject to assess whether age reduction has been calculated by integrating into the machine learning model to determine the test agent as an anti-aging agent for said subject, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject, and wherein the integrating the biological age after administration of the test agent into the machine learning model for said subject includes the biological age estimated in step (iii) and physiological parameters obtained from the exchange of user data from said subject.

In one embodiment, the present invention relates to a method of calculating the biological age of multiple subjects, the method comprising the steps of: (a) extracting DNA from multiple substrates obtained from multiple subjects; (b) measuring DNA methylation of the DNA extracted from the multiple substrates to obtain a DNA methylation profile; (c) analyzing the DNA methylation profile to obtain a polygenic score; and determining the biological age of multiple subjects based on the polygenic score, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from a subject for use in a method of screening an agent for whether the agent has anti-aging activity, the method comprising the steps of: (i) calculating age from a substrate obtained from a subject using a method disclosed in the present invention to obtain an initial biological age prior to biological intervention; (ii) administering a test agent to said subject; (iii) repeating step (i) on a further substrate obtained from said subject after performing step (ii) to determine the biological age after administration of the test agent; (iv) integrating the biological age after administration of the test agent into a machine learning model for said subject to assess whether age reduction has been calculated by integrating into the machine learning model to determine the test agent as an anti-aging agent for said subject, wherein the DNA extraction comprises extracting genomic DNA from saliva or blood obtained from the subject, and wherein the integrating the biological age after administration of the test agent into the machine learning model for said subject includes the biological age estimated in step (iii) and physiological parameters obtained from the exchange of user data from said subject.

Examples

The following examples are provided to illustrate the present invention and therefore should not be construed as limiting the scope of the present invention.

Example 1: Identification of 13 CG sites contained in the DNA region upstream of the ElovL2 gene, the ElovL2 AS1 region, whose weighted DNA methylation levels enable age prediction from salivary DNA.

In this example, the present invention relates to an "epigenetic clock" that is currently recognized as the most accurate measure of biological age. However, the tests that have been available to date require measuring DNA methylation at many sites (~350) using blood, which is an invasive and expensive method that is not applicable as a widespread consumer product. Although the available methods are suitable for scientific and clinical research, they are not suitable for widespread use of this test by consumers. Therefore, the present invention relates to a method that is an accurate, reliable, high-throughput and non-invasive test of biological aging based on the "epigenetic clock", in particular DNA methylation. The present invention in this example relates to polyCG DNA methylation markers of biological age for lifestyle management and healthy aging.

Identification of CG sites that correlate with age determined by DNA analysis from blood

In accordance with the present invention, the publicly available Illumina 450K whole genome blood DNA methylation assay chips (GSE40729) were subjected to Pearson correlation analysis. A smaller number of CG sites that had not been previously reported were selected and analyzed. Two of these CG sites were found to be located upstream of the antisense region of the ElovL2 gene, referred to as the ElovL2 AS1 region, as shown in the representative example form in Figure 1, as disclosed herein for the physical map, where, as shown herein, they were found to be highly correlated with age (with a Pearson correlation coefficient of r>0.9 and p=0). Then, in accordance with the present invention, it was determined that the combined weighted DNA methylation measurement for both of said CG sites accurately predicts age from blood DNA in an independent cohort (GSE40279 n=656 and GSE2219 n=60), as shown in Figure 2, as disclosed in the present invention. It was observed that methylation of the CG sites in the ElovL2 AS1 region as indicated in SEQ ID NO: 1, as disclosed herein, progressed from nearly 0% in fetuses to nearly 90% in 90-year-old humans. Thus, this ElovL2 AS1 region as indicated in SEQ ID NO: 1 and the CG sites found in said region, which are the dinucleotide sequence positions as described in Table 1, as disclosed herein, when applied once, showed a nearly perfect correlation with age, suggesting that a small number of said CG sites may be sufficient to determine biological age.

CG sites in the ElovL2 AS1 region enable age prediction from DNA in saliva samples

In order to assess the broad applicability of the disclosed DNA methylation age determination tests, it is important that they do not require skilled health professionals to obtain biological material. In this example, according to the present invention, it was determined whether the disclosed highly correlated CG sites, which are dinucleotide sequence positions as described in Table 1, located in the ElovL2 AS1 region as indicated in SEQ ID NO: 1 disclosed herein, can be used as age predictors in saliva by testing publicly available 450K Illumina DNA methylation chips from saliva (GSE78874, n = 259). The present invention discloses a methylation score consisting of weighted methylation measurements for cg16867657 and cg21572722, which are dinucleotide sequence positions as described in Table 1, and are CG site positions in the region as set forth in SEQ ID NO: 1, the ElovL2 AS1 region, together with cg09809672, which is a CG site on chromosome 1 as described in Table 1, as disclosed herein, which predicted age with a mean error of 5.62 years and a mean error of 4.74 years. The present invention then compares the accuracy of the model described herein to the gold standard, the Horvath clock. As seen in Figure 3, the ElovL2 AS1 region site scores were slightly better than the Horvath clock. It should be noted that the value of the ElovL2 gene for age prediction is known in the art. However, the present invention discloses that said prior art overlooks the fact that the two CG sites (namely, cg16867657 and cg21572722) that were thought to be in the ElovL2 gene are actually located upstream of another gene in an antisense orientation to the ElovL2 gene, the ElovL2-AS1 gene (see the physical description depicted in Figure 1 as disclosed herein), wherein said upstream region is referred to as the ElovL2 AS1 region and is disclosed in the present invention in SEQ ID NO: 1. This region upstream of the ElovL2-AS1 gene contains the selected 13 CG sites (see Table 1 as disclosed herein above).

Example 2: Bisulfite conversion, multiplex amplification and next generation sequencing and methylation calculation for 13 CGs in the ElovL2 AS1 region.

The present invention also provides that saliva, which is to be collected by the subject or consumer in DNA stabilization buffer (Tris 10 mM EDTA 10 mM, SDS 1%), which they will mail to the laboratory, is incubated with protease K (200 μg for 30 minutes at 37 ° C) in the laboratory. Genomic DNA is then purified using a Qiagen reagent kit. The purified DNA is treated with sodium bisulfite, using, for example, an EZ DNA bisulfite treatment kit. A library of target sequences is prepared using two-step PCR reactions using the following primers in a reaction with standard Taq polymerase:

For PCR 1 - amplification of the amplicon corresponding to the sequence indicated in SEQ ID NO: 1:

Forward primer indicated in SEQ ID NO: 5:

5’ACACTCTTTCCCTACACGACGCTCTTCCGATCTNNNNNYGGGYGGYGATTTGTAGGTTTAGT3’

Reverse primer indicated in SEQ ID NO: 6:

5’GTGACTGGAGTTCAGACGTGTGCTCTTCGATTCTCCCTACACRATACTACTTCTCCCC3’

For PCR 2 - for barcoding of samples, a second PCR reaction was performed with the following primers:

Forward primer indicated in SEQ ID NO: 7:

5’AATgATACggCgACCACCgAgATCTACACTCTTTCCCTACACgAC3 ’

Barcoding primer (reverse) indicated in SEQ ID NO: 8:

5’CAAGCAGAAGACGGCATACGAGATAGTCATCGGTGACTGGAGTTCAGACGTG3’ (the red bases represent the index; up to 200 variations of this index are used).

The second set of primers introduced an index for each sample, as well as primers for reverse and forward sequencing. PCR product 2 from all samples was pooled and purified on AMPpure-XP beads (NEB). The library was quantified by QPCR and loaded onto a MiSeq flow cell. The fastQ file was aligned to the corresponding genomic region using BisMark or other editing software.

Example 3: Superior characteristics of 13 CG sites in the region upstream of the ElovL2-AS1 gene region, the ElovL2 AS1 region, as indicated in SEQ ID NO: 1.

The present invention then determined whether the combined weighted methylation score of the described 13 CG sites provided superior predictive performance compared to 2 or 3 CG sites. Saliva samples were collected from 65 volunteers at the Hong Kong Science Park, and methylation levels at the 13 CG sites (see Table 1, as disclosed in the present invention above) were determined as described in Example 2. The present invention performed a series of multivariable linear regressions with different combinations of CG sites. The results, presented in Figure 4, in the present invention show that the combination of 13 CG sites (see Part D of Figure 4) performed better than a combination of either 4 CG sites (see Part A of Figure 4), 5 CG sites (see Part B of Figure 4), or 2 Illumina CG sites (see Part C of Figure 4). The result confirming the superiority of the 13 CG site combination over the smaller combinations of 4, 5 or 2 CG sites was further demonstrated by statistical comparison data as shown in Figure 4E. The Pearson product moment correlation coefficient r for the 13 CG methylation score was 0.95 (p = 1.8x10-33).

Example 4: Determination of biological age in consumer saliva samples.

As discussed above, biological age is an important parameter of our health. However, since the test described in the present invention is intended to be performed by subjects who are consumers at home outside of the professional health care system, it is important that the test is simple and does not require a health care professional to draw blood, as a preferred embodiment of the invention, since blood collection itself can be a moderately risky procedure, and that the test can be delivered to a central state-of-the-art laboratory by regular surface mail. Thus, the present invention discloses that 13 CG sites in the ElovL2 AS1 region form the basis of the EpiAging test, which provides such a possibility. According to the present invention, a consumer orders a saliva test kit, which contains a stabilizing buffer that maintains DNA stability for up to 1 month, via the EpiAging app, the Internet or email. The stabilizing buffer contains 20 mM Tris-HCl (pH 8.0), 20 mM EDTA, 0.5% SDS, and 1% Triton X-100. The barcoded kit arrives by mail to your home. The consumer scans the barcode with their phone scanner and registers their barcode in the app, which links the barcode to their phone's internal ID. Following the instructions provided with the EpiAging test kit, the consumer spits saliva into a collection tube, then transfers the saliva to a tube containing the stabilizing buffer, places the tube in a pre-paid postage stamped envelope, and mails it to the lab. In the lab, DNA is extracted, bisulfite treated (bisulfite chemical treatment converts unmethylated C to T, while methylated C remains as C), and the ElovL2 AS1 region is amplified as described in Example 3 and sequenced with samples from other patients on a MiSeq Illumina sequencer. fastQ files are analyzed and methylation values (m) are calculated for the 13 CG sites: mCGn = CGnC count / (CGnT count + CGnC count).

The values are then entered into the following equation to calculate biological age:

Biological age = (CG1 * 87.5643 + CG2 * 6.3301 + CG3 * -0.8691 + CG4 * 1.9468 + CG5 * 40.0336 + CG6 * 49.4303 + CG7 * -14.7868 + CG8 * 22.9042 + CG9 * -49.7942 + CG10 * 111.7467 + CG11 * 41.8108 + CG12 * 0.4144 + CG13 * -150.8005) -71.6872

The biological age value is then sent to the consumer in their EpiAging app, or the consumer can retrieve the results using their barcode ID. A significant increase in biological age over chronological age (+5 years) serves as a “red flag” for lifestyle changes. The consumer periodically (every 6-12 months) measures their biological age and assesses their progress in closing the gap between biological and chronological age.

Example 5: EpiAging app for managing biological age testing and lifestyle data.

The present invention relates to the EpiAging application (see Figure 5, which shows the home page of the application), which is compatible with the Apple and Android operating systems and provides information about the "EpiAging test", ordering methods, an order basket and a link to an electronic payment form such as PayPal or Alipay. The innovative aspect of the disclosed application of the present invention is that it combines the management of the "EpiAging test" based on customer data with a system of customer-driven management of lifestyle changes based on dynamic recommendations from authoritative national and international medical groups, "self-reporting", data sharing, machine learning, iterative changes driven by iterative machine learning, personalized reports for consumers and repeated assessments (see Figure 6). The "reinforcement learning" system suggests lifestyle changes that have the greatest impact on slowing down aging, as determined by the difference between DNA methylation age and chronological age. The application presented here is built using open source programs known to any specialist in this field of technology, such as Build Fire JS, Ionic, Appcelerator’s Titanium SDK, Mobile angular UI and Siberian CMS.

The application is downloaded from the Apple Store, Google Play Store or websites. The application requires registration and assignment of a consumer ID. The application activates a scanner that scans a barcode and associates a test barcode with a consumer ID. The data will be associated with these “blind” identifiers. Personal data and consumer data are separated, protected and labeled to ensure complete anonymization of “aging” and lifestyle data. The data management system does not have access to personal data. The system is built to recover personal identifiers that can only be initiated by the consumer using their email account, but are completely hidden from the data management system. Data anonymization is a fundamental function of the application.

The first page of the app has several buttons (see Figure 5). One button provides links to basic information about the "aging" tests, scientific citations, and PubMed links for more in-depth study of the field.

The "Info" function provides information on the relationship between lifestyle and "aging." The second button links to a page that contains a series of buttons that link to buttons for lifestyle and wellness areas such as "mood," "chronic pain," "nutrition" including supplements such as S-adenosyl-L-methionine, vitamins, etc., "physiological parameters" such as blood pressure, heart rate, weight, height, fasting sugar and other metabolic tests, medications, drug and alcohol use, smoking, and exercise data entered by the consumer. Each section is preceded by recommendations compiled from recommendations of authoritative associations such as the National Heart Association and the National Stroke Association or the Diabetes Association, the American Cancer Society, etc. The recommendations section provides a link to these associations so that the consumer can make his or her own judgment and decisions. The idea behind the lifestyle management section is consumer self-improvement and control over their lifestyle decisions. Data entry is done by sliding along a numerical scale. Yes-no entries are interpreted as 0 for No and 1 for Yes. Other quantifiable entries are entered by their number. Above each data entry scale is a rated representation of the recommendation, giving the consumer an assessment of their performance against the color-coded recommendations. The recommended range is shown in green. Deviations from the recommendations are shown in red if above and blue if below the recommended range. The save button, which the consumer clicks after completing each section's data entry, allows the data to be saved. A summary analysis report is provided after data entry. Charts are also provided, outlining progress over time against the national recommendation. Once the lab aging test is completed, the tests are delivered remotely to the app. User data, as well as data from other users, is stored in a cloud database for further analysis.

Example 6: Machine learning-based analysis of health and DNA methylation data and personalized lifestyle recommendations.

The data collected from multiple users is stored in a cloud-based SQL database (see Figure 7). Machine learning algorithms are used to analyze the data and build models that relate questionnaire inputs such as pain, blood pressure, BMI, and mood to the difference between DNA methylation age and chronological age as an output. For example, using methods such as neural networks, decision trees, random forest lasso regression, K-means clustering, reinforcement learning, and penalized regression.

The method of the present invention includes performing a statistical analysis of the responses to the questionnaire and providing consumers with a dynamic report in the Appendix that describes the development of the responses to the questionnaire over time compared to the recommendations of national associations, such as associations studying cancer, heart disease and stroke, and diabetes.

Although the invention has been explained with respect to its preferred embodiment, it will be understood that many other possible modifications and changes can be made without departing from the spirit and scope of the present invention.

Advantages of the invention

The innovative aspect of the method of the present invention, compared to solutions known in the art, is that the combination of 13 CG sites in a previously undescribed region of ElovL2 AS1 provides extremely high-precision age prediction from saliva samples in a single amplicon. The accuracy and simplicity are increased by multiplexing and the use of reliable next-generation sequencing. This approach significantly reduces the cost and makes the test suitable for use as a consumer product.

Links

Beauchaine, T. P., & Beauchaine, R. J., 3rd. (2002). A comparison of maximum covariance and K-means cluster analysis in classifying cases into known taxon groups. Psychol Methods, 7(2), 245-261.

Bybee, S. M., Bracken-Grissom, H., Haynes, B. D., Hermansen, R. A., Byers, R. L., Clement, M. J., Crandall, K. A. (2011). Targeted amplicon sequencing (TAS): a scalable next-gen approach to multilocus, multitaxa phylogenetics. Genome Biol Evol, 3, 1312-1323. doi:10.1093/gbe/evr106

Chen, B. H., Marioni, R. E., Colicino, E., Peters, M. J., Ward-Caviness, C. K., Tsai, P. C., Horvath, S. (2016). DNA methylation-based measures of biological age: meta-analysis predicting time to death. Aging (Albany NY), 8(9), 1844-1865. doi:10.18632/aging.101020

Colella, S., Shen, L., Baggerly, K. A., Issa, J. P., & Krahe, R. (2003). Sensitive and quantitative universal Pyrosequencing methylation analysis of CpG sites. Biotechniques, 35(1), 146-150.

De Roach, J. N. (1989). Neural networks—an artificial intelligence approach to the analysis of clinical data. Australas Phys Eng Sci Med, 12(2), 100-106.

Ferrucci, L., Cavazzini, C., Corsi, A., Bartali, B., Russo, C. R., Lauretani, F., Guralnik, J. M. (2002). Biomarkers of frailty in older persons. J Endocrinol Invest, 25(10 Suppl), 10-15.

Freire-Aradas, A., Phillips, C., Mosquera-Miguel, A., Giron-Santamaria, L., Gomez-Tato, A., Casares de Cal, M., Lareu, M. V. (2016). Development of a methylation marker set for forensic age estimation using analysis of public methylation data and the Agena Bioscience EpiTYPER system. Forensic Sci Int Genet, 24, 65-74. doi:10.1016/j.fsigen.2016.06.005

Hardy, A., & Magnello, M. E. (2002). Statistical methods in epidemiology: Karl Pearson, Ronald Ross, Major Greenwood and Austin Bradford Hill, 1900-1945. Soz Praventivmed, 47(2), 80-89.

Hertel, J., Friedrich, N., Wittfeld, K., Pietzner, M., Budde, K., Van der Auwera, S., Grabe, H. J. (2016). Measuring Biological Age via Metabonomics: The Metabolic Age Score. J Proteome Res, 15(2), 400-410. doi:10.1021/acs.jproteome.5b00561

Horvath, S. (2013). DNA methylation age of human tissues and cell types. Genome Biol, 14(10), R115. doi:10.1186/gb-2013-14-10-r115

Jylhava, J., Pedersen, N. L., & Hagg, S. (2017). Biological Age Predictors. EBioMedicine, 21, 29-36. doi:10.1016/j.ebiom.2017.03.046

Kakushadze, Z., & Yu, W. (2017). *K-means and cluster models for cancer signatures. Biomol Detect Quantif, 13, 7-31. doi:10.1016/j.bdq.2017.07.001

Kim, S. M., Kim, Y., Jeong, K., Jeong, H., & Kim, J. (2018). Logistic LASSO regression for the diagnosis of breast cancer using clinical demographic data and the BI-RADS lexicon for ultrasonography. Ultrasonography, 37(1), 36-42. doi:10.14366/usg.16045

Kristensen, L. S., Mikeska, T., Krypuy, M., & Dobrovic, A. (2008). Sensitive Melting Analysis after Real Time - Methylation Specific PCR (SMART-MSP): high-throughput and probe-free quantitative DNA methylation detection. Nucleic Acids Res.

Mann, J. J., Ellis, S. P., Waternaux, C. M., Liu, X., Oquendo, M. A., Malone, K. M., Currier, D. (2008). Classification trees suicide attempters in major psychiatric disorders: a model of clinical decision making. J Clin Psychiatry, 69(1), 23-31.

Marioni, R. E., Harris, S. E., Shah, S., McRae, A. F., von Zglinicki, T., Martin-Ruiz, C., Deary, I. J. (2018). The epigenetic clock and telomere length are independently associated with chronological age and mortality. Int J Epidemiol, 47(1), 356. doi:10.1093/ije/dyx233

Monaghan, P. (2010). Telomeres and life histories: the long and the short of it. Ann N Y Acad Sci, 1206, 130-142. doi:10.1111/j.1749-6632.2010.05705.x

Mupparapu, M., Wu, C. W., & Chen, Y. C. (2018). Artificial intelligence, machine learning, neural networks, and deep learning: Futuristic concepts for new dental diagnosis. Quintessence Int, 49(9), 687-688. doi:10.3290/j.qi.a41107

Sherbet, G. V., Woo, W. L., & Dlay, S. (2018). Application of Artificial Intelligence-based Technology in Cancer Management: A Commentary on the Deployment of Artificial Neural Networks. Anticancer Res, 38(12), 6607-6613. doi:10.21873/anticanres.13027

Shi, T., Seligson, D., Belldegrun, A. S., Palotie, A., & Horvath, S. (2005). Tumor classification by tissue microarray profiling: random forest clustering applied to renal cell carcinoma. Mod Pathol, 18(4), 547-557. doi:10.1038/modpathol.3800322

Svetnik, V., Liaw, A., Tong, C., Culberson, J. C., Sheridan, R. P., & Feuston, B. P. (2003). Random forest: a classification and regression tool for compound classification and QSAR modeling. J Chem Inf Comput Sci, 43(6), 1947-1958. doi:10.1021/ci034160g

Vetter, V. M., Meyer, A., Karbasiyan, M., Steinhagen-Thiessen, E., Hopfenmuller, W., & Demuth, I. (2018). Epigenetic clock and relative telomere length represent largely different aspects of aging in the Berlin Aging Study II (BASE-II). J Gerontol A Biol Sci Med Sci. doi:10.1093/gerona/gly184

Yanai, H., Budovsky, A., Tacutu, R., & Fraifeld, V. E. (2011). Is rate of skin wound healing associated with aging or longevity phenotype? Biogerontology, 12(6), 591-597. doi:10.1007/s10522-011-9343-6

Yu, M., Heinzerling, T. J., & Grady, W. M. (2018). DNA Methylation Analysis Using Droplet Digital PCR. Methods Mol Biol, 1768, 363-383. doi:10.1007/978-1-4939-7778-9_21

Zhao, Y., Kosorok, M. R., & Zeng, D. (2009). Reinforcement learning design for cancer clinical trials. Stat Med, 28(26), 3294-3315. doi:10.1002/sim.3720

Claims (37)

1. A method for calculating the biological age of a subject, which includes the following steps: (a) extraction of DNA from a substrate obtained from the subject; (b) measuring DNA methylation in the DNA extracted from the substrate to obtain a DNA methylation profile; (c) DNA methylation profiling of 13 human CG sites located within the putative antisense region of the ElovL2 gene, the ElovL2 AS1 region, as indicated in SEQ ID NO: 1, to generate a polygenic score based on the following biological age calculation equation: biological age = (CG1*87.5643+CG2*6.3301+CG3*-0.8691+CG4*1.9468+CG5*40.0336+CG6*49.4303+CG7*-14.7868+CG8*22.9042+CG9*-49.7942+CG10*111.7467+CG11*41.8108+CG12*0.4144+CG13*-150.8005)-71.6872, using multiple linear regression equations or neural network analysis; and (d) determining the biological age of the subject based on a polygenic assessment, wherein the DNA extraction comprises extracting genomic DNA from saliva obtained from the subject. 2. The method according to claim 1, characterized in that the measurement of DNA methylation is carried out using methods including DNA pyrosequencing, mass spectrometry (Epityper™), PCR-based methylation analysis, next-generation bisulfite sequencing of the target amplicon on a platform selected from the group of HiSeq, MiniSeq, MiSeq and NextSeq sequencers, Ion Torrent sequencing, sequencing using methylated DNA immunoprecipitation (MeDIP) or hybridization with oligonucleotide chips. 3. The method according to claim 1, characterized in that the measurement of DNA methylation is performed for polygenic DNA methylation biomarkers, which includes measuring the methylation status of CG sites in said 13 human CG sites that are located in the putative antisense region of the ElovL2 gene, the ElovL2 AS1 region, as indicated in SEQ ID NO: 1. 4. The method according to claim 1, characterized in that the measurement of DNA methylation is carried out using DNA pyrosequencing with primers indicated in SEQ ID NO: 2 for the forward biotinylated primer, SEQ ID NO: 3 for the reverse primer, and SEQ ID NO: 4 for the pyrosequencing primer. 5. The method according to claim 1, characterized in that the measurement of DNA methylation is carried out using next-generation bisulfite sequencing of the target amplicon on a platform selected from the group of HiSeq, MiniSeq, MiSeq and NextSeq sequencers, with primers indicated in SEQ ID NO: 5 for the forward primer and SEQ ID NO: 6 for the reverse primer. 6. The method according to claim 1, characterized in that the measurement of DNA methylation is carried out using PCR-based methylation assays selected from the group of methylation-specific PCR and digital PCR. 7. The method according to paragraph 1 for calculating biological age among several subjects, including the steps of the method according to paragraph 1 among several subjects. 8. The method according to claim 7, characterized in that the measurement of DNA methylation in DNA extracted from several substrates includes the following steps: (a) amplifying genomic DNA extracted from multiple substrates with target-specific primers to obtain PCR product 1; (b) amplifying PCR product 1 from step (a) with barcoding primers to obtain PCR product 2; (c) performing multiplex sequencing in a single Miseq next-generation sequencing reaction using PCR product 2 from step (b); (d) extracting data from the multiplex sequencing step (c); and (e) quantifying DNA methylation from the data extracted in step (d) to obtain a DNA methylation profile for each substrate, wherein the DNA extraction comprises extracting genomic DNA from saliva obtained from each subject of said plurality of subjects. 9. The method according to claim 8, characterized in that the target-specific primers for obtaining the PCR product 1 comprise the primers indicated in SEQ ID NO: 5 for the forward primer and SEQ ID NO: 6 for the reverse primer, and where the barcoding primers for obtaining the PCR product 2 comprise the primers indicated in SEQ ID NO: 7 for the forward primer and SEQ ID NO: 8 for the reverse primer, which is a barcode index primer. 10. A kit for determining the biological age of a subject for carrying out the method according to any one of claims 1-9, comprising means and reagents for collecting and stabilizing a substrate from the subject; a scanner for reading a barcode in the kit and instructions for collecting and stabilizing the substrate, where the substrate is the saliva of the subject and where the stabilization of the substrate is intended for sending the collected substrate by mail for DNA extraction for measuring DNA methylation in DNA extracted from the substrate to obtain a DNA methylation profile of the subject to determine the biological age of the subject. 11. The method according to any of paragraphs 1-9 for use in a method for assessing the impact of a biological intervention on the biological age of a subject, the method includes the following steps: (a) calculating the biological age of the subject using the method of claim 1 to obtain the initial biological age prior to the biological intervention; (b) performing a biological intervention on the said subject; (c) repeating step (a) on a further substrate obtained from said subject after step (b) has been performed to determine the biological age following the biological intervention; (d) integrating the biological age after the biological intervention into the machine learning model for said subject to estimate the effect of the biological intervention on the biological age of said subject, wherein the biological intervention in step (b) is selected from the group consisting of dietary supplements, vitamins, therapy, administration of a test substance, dietary manipulation, metabolic manipulation, surgical manipulation, social manipulation, behavioral manipulation, environmental manipulation, sensory manipulation, hormonal manipulation, and epigenetic manipulation, or combinations thereof, wherein the DNA extraction comprises extracting genomic DNA from saliva obtained from the subject, and where the integration of the biological age after the biological intervention into the machine learning model for said subject includes the biological age estimated in step (c) and physiological parameters obtained by exchanging user data from said subject. 12. The method according to any one of claims 1-9 for use in a method of screening an agent for whether the agent has anti-aging activity, the method comprising the following steps: (a) calculating age from a substrate obtained from the subject using the method of any one of claims 1-9 to obtain an initial biological age prior to administration of the test agent; (b) administering the test agent to the said subject; (c) repeating step (a) on a further substrate obtained from said subject after step (b) to determine the biological age following administration of the test agent; (d) integrating the biological age after administration of the test agent into the machine learning model for said subject to evaluate whether age reduction has been calculated by integrating into the machine learning model to determine the test agent as an anti-aging agent for said subject, wherein the DNA extraction comprises extracting genomic DNA from saliva obtained from the subject, and where the integration of the biological age after introducing the test agent into the machine learning model for said subject includes the biological age estimated in step (c) and physiological parameters obtained as a result of exchanging user data from said subject.