WO2018137203A1 - 确定人群样本生物指标集、预测生物学年龄的方法及其应用 - Google Patents

确定人群样本生物指标集、预测生物学年龄的方法及其应用 Download PDF

Info

Publication number
WO2018137203A1
WO2018137203A1 PCT/CN2017/072675 CN2017072675W WO2018137203A1 WO 2018137203 A1 WO2018137203 A1 WO 2018137203A1 CN 2017072675 W CN2017072675 W CN 2017072675W WO 2018137203 A1 WO2018137203 A1 WO 2018137203A1
Authority
WO
WIPO (PCT)
Prior art keywords
age
biological
sample
indicators
indicator
Prior art date
Application number
PCT/CN2017/072675
Other languages
English (en)
French (fr)
Inventor
齐彦伟
柴相花
李伟阳
聂超
王书元
陈志华
张现东
李尉
甄贺富
谭美华
张爱萍
张彩芬
李睿
赵昕
Original Assignee
深圳华大基因研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳华大基因研究院 filed Critical 深圳华大基因研究院
Priority to CN201780084324.XA priority Critical patent/CN110392740A/zh
Priority to PCT/CN2017/072675 priority patent/WO2018137203A1/zh
Publication of WO2018137203A1 publication Critical patent/WO2018137203A1/zh

Links

Images

Classifications

    • CCHEMISTRY; METALLURGY
    • C12BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12QMEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids

Definitions

  • the present invention relates to the field of biotechnology, and in particular to the field of biological age prediction technology, and more particularly to a method for determining a population sample biological indicator set, predicting a biological age, and an application thereof.
  • the biological age is related to the occurrence time of certain events in human growth and development. It is the age inferred from the normal human physiology and anatomy, indicating the actual state of the human body's tissue structure and physiological functions.
  • Biological age is a comprehensive index of human health status and an objective expression of the degree of aging of the body.
  • the biological age can be inconsistent with the real age, and its determination has various methods and modes. Because the function of the cardiovascular system is strongly dependent on the age of the human body, it also reflects the health of the body. Calculating the biological age of an individual can effectively assess the state of aging and understand the health of the human body.
  • the present invention aims to solve at least one of the technical problems existing in the prior art. Accordingly, it is an object of the present invention to at least provide a means for accurately determining the biological age of an individual and assessing the level of aging thereof.
  • the present invention intends to eliminate the collinearity problem caused by multi-index fitting by proposing a screening algorithm for biological age-related indicators, and using the large-scale population death age distribution data to correct the computational biology age as the individual biological age, the strategy is The biological age calculation model established for the present invention.
  • the inventors attempted to establish an aging assessment baseline model, filter outlier samples, calculate the biological age confidence interval of the population, and finally assess the relative state of the individual at the baseline of aging population to achieve the purpose of guiding or intervening aging.
  • the inventors succeeded in collecting biological indicators of large-scale samples, screening biological indicators related to age or aging, calculating the age of individual biology, and establishing different groups (sex Group, age group) aging baseline, and ultimately achieve the purpose of accurate quantitative assessment of individual aging levels.
  • the invention provides a method of determining a biological age predictive biological indicator set for a population sample.
  • the method comprises the steps of: obtaining data of candidate biological indicators of all individuals in the population sample; grouping candidate biological indicator data of all individuals in the population sample by gender to obtain a male candidate biomarker set and a female candidate biometric indicator set; respectively, the male candidate biometric indicator set and the female candidate biometric indicator set are separately filtered by a variance expansion factor algorithm to obtain a male effective candidate biometric indicator set respectively And a set of valid alternative biometric indicators for the female; each of the set of effective candidate biomarkers and the set of valid alternative biometric indicators of the female are grouped according to age groups, respectively, so as to obtain effective males of different age groups An alternative biometric indicator set and a plurality of female effective candidate biometric indicator sets of different age groups; respectively determining the male effective candidate biometric indicator set of the plurality of different age groups and the female effective candidate of the plurality of different age groups A set
  • this method can effectively determine the biological age prediction biological indicator set of the population sample, and then use these biological indicator sets to accurately determine the individual biological age and the aging evaluation baseline of the population sample, and further, based on The obtained individual biological age and the aging evaluation baseline of the population sample can effectively evaluate the relative aging degree of the individual, and the evaluation result is accurate and highly reliable.
  • the invention provides a method of determining the biological age of an individual to be tested.
  • the method comprises the steps of: determining, according to the method for determining a biological indicator set of a biological age of a population sample, the biological age predictions by sex age classification of the sample of the population to be tested a set of biological indicators; calculating a preliminary estimate of the biological age of the test subject based on the biological age predicted biological indicator set corresponding to the sex age corresponding to the test subject And preliminary estimates of the biological age with reference to sample age distribution data for different populations A maximum posterior probability calculation process is performed to determine the predicted biological age BA of the individual to be tested.
  • the inventors have surprisingly found that the method can accurately and effectively determine the biological age of an individual, and then based on the aging evaluation baseline of the population sample, can effectively assess the relative aging degree of the individual, and the evaluation result is accurate and highly reliable. .
  • the invention provides a method of determining a baseline for aging assessment of a population sample.
  • the method comprises the steps of: determining, according to the foregoing method for determining a biological indicator set of a biological age of a population sample, determining a biological age-predicted biological indicator set classified by sex age of the population sample; Re-filtering treatment of each biological age-predicted biological indicator set classified by sex age, wherein the biological indicator set for each biological age classified by sex age is larger than the biological age-predicted biological index
  • the sample size is less than the number of biological age-predicted biological indicators, remove the highest 5% of the Mahalanobis Distance samples and filter the organism using the variance expansion factor algorithm.
  • the aging assessment baseline of the population sample can be effectively determined by using the method, and then the biological age of the individual to be tested is compared with the aging assessment baseline of the gender age group in which the patient is located, and the test can be effectively evaluated.
  • the relative aging of the individual, and the assessment results are accurate and highly reliable.
  • the invention provides a method of determining the relative aging of an individual to be tested.
  • the method comprises the steps of: determining a biological age BA of the test subject according to the method for determining the biological age of the individual to be tested, as described above; determining the aging of the sample of the population according to the foregoing a method for assessing a baseline, determining a baseline of aging assessment of each gender age group of the sample of the population to which the test subject belongs; and determining the aging of the biological age BA of the test subject and the sex age group of the test subject Assess baselines for comparison to determine The relative aging degree of the test subject, wherein when the biological age BA of the test subject is within a range of the aging evaluation baseline of the sex age group in which the test subject is located, determining the test subject relative to The sample of the population to be tested is at a normal aging level; when the biological age BA of the test subject deviates from the range of the aging assessment
  • the inventors have surprisingly found that the relative aging degree of the test subject can be effectively evaluated by using the method, and the evaluation result is good in accuracy and high in credibility.
  • FIG. 1 shows a flow chart of a biological age assessment model in accordance with an embodiment of the present invention
  • FIG. 2 shows a flow chart of a aging baseline estimation model in accordance with an embodiment of the present invention
  • FIG. 3 is a graph showing the results of aging baseline distribution of biological ages BA of different genders and different age groups according to an embodiment of the present invention
  • Figure 4 shows the results of an indicator associated with biological age in a female 20-25 age group, in accordance with an embodiment of the present invention
  • Figure 5 shows the results of an indicator associated with biological age in a female 25-30 age group, in accordance with an embodiment of the present invention
  • Figure 6 shows the results of an indicator associated with biological age in a 30-35 year old female population, in accordance with an embodiment of the present invention
  • Figure 7 shows the results of indicators associated with biological age in the 35-40 year old age group of women in accordance with an embodiment of the present invention
  • Figure 8 shows the results of indicators associated with biological age in the 40-100 year old age group of women in accordance with an embodiment of the present invention
  • Figure 9 shows the results of an indicator associated with biological age in a male 20-25 age group, in accordance with an embodiment of the present invention.
  • Figure 10 shows the results of an indicator associated with biological age in a male 25-30 age group, in accordance with an embodiment of the present invention
  • Figure 11 shows the results of an indicator associated with biological age in a male 30-35 age group, in accordance with an embodiment of the present invention
  • Figure 12 shows indicators associated with biological age in the male 35-40 age group, in accordance with an embodiment of the present invention. result
  • Figure 13 shows the results of indicators associated with biological age in the male 40-100 age group, in accordance with an embodiment of the present invention
  • Figure 14 shows the results of biological age and aging levels measured by sample No. 167 in accordance with Example 2 of the present invention.
  • the prior art has too simple to evaluate the biological age (BA), and only obtains individual physical indicators through biochemical indicators, psychological test reports, etc., and does not filter the indicators that truly affect the biological age, but directly fits the population.
  • the so-called psychological age, immunization age and other practices are too rough.
  • differences in the contribution of each indicator to biological age were not considered for different genders and age groups; therefore, biological indicators were assessed. These differences have affected the calculation of biological age and the assessment of aging to some extent.
  • the inventors first provided a screening determination method for a biological indicator set for biological age prediction of a population sample that is more scientific and rigorous than the prior art.
  • the invention provides a method of determining a biological age predictive biological indicator set for a population sample.
  • the method comprises the steps of: obtaining data of candidate biological indicators of all individuals in the population sample; grouping candidate biological indicator data of all individuals in the population sample by gender to obtain a male candidate biomarker set and a female candidate biometric indicator set; respectively, the male candidate biometric indicator set and the female candidate biometric indicator set are separately filtered by a variance expansion factor algorithm to obtain a male effective candidate biometric indicator set respectively And a set of valid alternative biometric indicators for the female; each of the set of effective candidate biomarkers and the set of valid alternative biometric indicators of the female are grouped according to age groups, respectively, so as to obtain effective males of different age groups An alternative biometric indicator set and a plurality of female effective candidate biometric indicator sets of different age groups; respectively determining the male effective candidate biometric indicator set of the plurality of different age groups and the female effective candidate of the plurality of different age groups A set of valid biological indicators for each of the biological indicators in order to obtain effective males of different age groups a set of indicators and a set of female effective bio
  • this method can effectively determine biological age predictive biological indicators of population samples.
  • the set, and then the use of these sets of biological indicators can accurately determine the individual's biological age and the aging assessment baseline of the population sample, and further, based on the obtained individual biological age and the aging assessment baseline of the population sample can effectively assess the relative aging of the individual Degree, and the results obtained are accurate and reliable.
  • the candidate biological indicator is at least one selected from the group consisting of a longevity gene, a mitochondrial DNA copy number, a telomere length, an overall methylation level, and a hormone level.
  • the “alternative biological indicator” described herein may be a biological indicator of any measurable characteristic in the population, including all known and unknown indicators associated with aging, and is not limited to the above. Several major indicators. In addition, in early studies it was determined that aging is relevant, like physiological and biochemical blood, and characterizations that may be detected in the future can be included. The range of indicators should be broad, and the method of the present invention (sometimes referred to as a model) can discriminate whether the known and unknown indicators are related to aging or age, thereby determining the set of indicators for calculating the biological age.
  • the data of the biological indicators of all individuals in the population sample conforms to standard quality control.
  • all the biometric indicator data of the male candidate biometric indicator set and the female candidate biometric indicator set are the same batch detection, or meet the requirement of inter-batch detection CV value, or are not satisfied.
  • Inter-batch detection CV values have been corrected using the LMM algorithm, and the sample size of each group should be greater than the number of alternative biometrics.
  • the variance expansion factor algorithm filtering is performed by the following steps:
  • regression coefficients a 0 , a 1 , a 2 , ..., a m-1 , a m are obtained :
  • the VIF corresponding to each candidate biological indicator or effective candidate biological indicator is obtained based on the following formula:
  • the biological age prediction biological indicator set of the obtained population sample is reliable, and is applied to the biological age calculation, the baseline determination of the population sample aging evaluation, and the final determination of the relative aging degree of the individual, the accuracy. Good and reliable.
  • the grouping by age group is performed at predetermined age intervals.
  • the age group interval scale of the age grouping is theoretically not limited, and the actual operation can be selected according to the sample size to be studied and the accuracy of the evaluation.
  • the predetermined age interval of the grouping may be 5 years or 10 years.
  • the predetermined age interval of the grouping is selected to be 5 years, specifically, between 20 and 40 years old, grouped every 5 years, and individuals older than 40 years old are grouped into one group.
  • the individual's overall age change is not too large on the scale of 3-5 years, on the other hand, it is limited by the sample size, the finer the age group, the less each group of samples; as for 40 Above age, in theory, it is also grouped by age, but in the actual example, the samples of older samples are not much; while the samples under 20 are too small to be ignored.
  • the age-related association filtering is performed based on the following formula using the Pearson correlation analysis method:
  • y represents the age of the age CA
  • x represents the detection value corresponding to the effective biological indicator
  • the biological age prediction biological indicator set of the selected population samples is reliable, and is applied to the biological age calculation, the baseline determination of the population sample aging assessment, and the final determination of the relative aging degree of the individual, with good accuracy and high credibility. .
  • the inventors based on the aforementioned method for determining the biological indicator set of the biological age of the population sample, that is, screening the biological age-related index algorithm, eliminating the collinearity problem of multi-index fitting, and using the large-scale population death age distribution.
  • the data, corrected for the biological age, as the individual biological age, the strategy is the biological age calculation model established by the present invention.
  • the invention provides a method of determining the biological age of an individual to be tested.
  • the method comprises the steps of: determining, according to the method for determining a biological indicator set of a biological age of a population sample, the biological age predictions by sex age classification of the sample of the population to be tested a set of biological indicators; calculating a preliminary estimate of the biological age of the test subject based on the biological age predicted biological indicator set corresponding to the sex age corresponding to the test subject And preliminary estimates of the biological age with reference to sample age distribution data for different populations A maximum posterior probability calculation process is performed to determine the predicted biological age BA of the individual to be tested.
  • the inventors have surprisingly found that the method can accurately and effectively determine the biological age of an individual, and then based on the aging evaluation baseline of the population sample, can effectively assess the relative aging degree of the individual, and the evaluation result is accurate and highly reliable. .
  • a preliminary estimate of the biological age of the test subject is calculated based on the KD model method based on the following formula
  • C is the age of age CA
  • j is the biological age prediction biological indicator BM
  • m is the biological age prediction biological indicator set BMs type
  • k j is the biological indicator BM to CA for each biological age
  • the slope of the combination q j is the intercept of the biological indicator BM for the age-specific CA
  • x j is the value of the j-th biological age-predicted biological indicator BM of the sample x.
  • the maximum a posteriori probability calculation is performed based on the following formula to determine the predicted biological age BA:
  • ⁇ C is the actual age C
  • ⁇ and ⁇ 0 are the standard deviations of the likelihood function and the prior function, respectively.
  • the final biological age is accurate and reliable.
  • the invention provides a method of determining a baseline for aging assessment of a population sample.
  • the method is scientific and rigorous, and the identified population sample aging assessment has good reliability and practical value.
  • the method comprises the steps of: determining a biological age-predicted organism classified by sex age of a population sample according to the method for determining a biological indicator set of biological age of a population sample as described above a set of indicators for each biological age-predicted biological indicator group classified by sex age, wherein the biological indicator set for each biological age classified by sex age is greater than the biological age predicting organism
  • the number of indicators is reduced, the highest 5% of the Euclidean Distance is removed.
  • the sample size is less than the number of biological age-predicted biological indicators, the highest 5% of the Mahalanobis Distance is removed and the variance expansion factor is used.
  • the algorithm filters the biological age-predicted biological indicators; uses the age-based CA to linearly fit all biological age-predicted biological indicators, removes the Cook's Distance>1 sample, and removes the biological age-predicted biological indicators with a correlation of ⁇ 0.1 with CA. In order to screen out the initial criteria for meeting the established baseline.
  • This biological age and predictive biomarker set biomarker prediction set based on the baseline established to meet the initial sample and standard biological age, biological age of the population is calculated for each individual sample preliminary estimate
  • Preliminary estimates of the biological age based on sample age distribution data from different populations Perform a maximum posterior probability calculation process to determine the predicted biological age BA for each individual; linearly fit the chronological age CA using the predicted biological age BA of each individual, remove the sample of Cook's Distance>1, and repeat this Steps to no sample Cook's Distance>1 to screen out a set of sample and biological age-predicted biological indicators that meet established baseline requirements; and calculate ages for each gender based on the sample and biological age-predicted biological indicator sets that meet the established baseline requirements
  • the 95% confidence interval for the biological age prediction of the group, the 95% confidence interval for the biological age prediction is the baseline for aging assessment for each gender age group.
  • the aging assessment baseline of the population sample can be effectively determined by using the method, and then the biological age of the individual to be tested is compared with the aging assessment baseline of the gender age group in which the patient is located, and the test can be effectively evaluated.
  • the relative aging of the individual, and the assessment results are accurate and highly reliable.
  • a preliminary estimate of the biological age of each individual is calculated by the KD model method based on the following formula
  • C is the age of age CA
  • j is the biological age prediction biological indicator BM
  • m is the biological age prediction biological indicator set BMs type
  • k j is the biological indicator BM to CA for each biological age
  • the slope of the combination q j is the intercept of the biological indicator BM for the age-specific CA
  • x j is the value of the j-th biological age-predicted biological indicator BM of the sample x.
  • k j , q j and s are determined according to the following steps:
  • the maximum a posteriori probability calculation is performed based on the following formula to determine the predicted biological age BA:
  • ⁇ C is the actual age C
  • ⁇ and ⁇ 0 are the standard deviation of the likelihood function and the standard deviation of the prior function, respectively.
  • screening the population sample based on a linear fit result comprises removing a sample of Cook's Distance>1. Therefore, the screening effect is good, and the final determined population sample aging evaluation baseline is reliable and the application value is high.
  • the invention provides a method of determining the relative aging of an individual to be tested.
  • the method comprises the steps of: determining a biological age BA of the test subject according to the method for determining the biological age of the individual to be tested, as described above; determining the aging of the sample of the population according to the foregoing a method for assessing a baseline, determining a baseline of aging assessment of each gender age group of the sample of the population to which the test subject belongs; and determining the aging of the biological age BA of the test subject and the sex age group of the test subject Evaluating a baseline for comparison to determine a relative senescence level of the subject to be tested, wherein when the biological age BA of the subject to be tested is within a range of aging assessment baselines of the sex age group in which the subject is to be tested, Determining that the sample to be tested is at a normal aging level relative to the population sample to be tested; and determining the test to be
  • the biological age of the test subject, the baseline of the aging evaluation of the population sample, and the biological age of the test subject can be compared with the baseline of the aging assessment of the sex age group in which the test is performed,
  • the relative aging degree of the test subject can be effectively evaluated, and the evaluation result is good in accuracy and high in credibility.
  • the present invention also provides apparatus, apparatus and systems suitable for implementing the methods described above, for example, corresponding to the method of determining a biological age predicting biological indicator set of a population sample of the present invention.
  • the above-described devices, devices and systems suitable for implementing the various methods are structurally composed of corresponding units, devices or devices suitable for implementing
  • the corresponding device the device for determining the biological age of the individual to be tested, includes: a biological age for determining the sample of the population a device for predicting a biological indicator set for determining a biological age-predicted biological indicator set classified by sex age according to a biological age prediction biological indicator set method for determining a population sample according to the foregoing; a school age preliminary estimation value calculation device, wherein the biological age preliminary estimation value calculation device is connected to the device for determining a biological age prediction biological indicator set of the population sample, and is configured to classify the sex-based age corresponding to the individual to be tested a biological age predicting biological indicator set, calculating a preliminary estimate of the biological age of the test subject And a biological age prediction device coupled to the biological age preliminary estimate calculation device for using the sample age distribution data of different populations as a reference, and preliminary estimates of the biological age A maximum posterior probability calculation process is performed to determine the predicted biological age BA
  • the present invention has at least one of the following effects:
  • the rigorous biological age calculation model can accurately calculate the biological age
  • the method for determining the relative aging degree of an individual to be tested generally includes the following steps:
  • determining a biological age prediction biological indicator set of a population sample determining a biological age prediction biological indicator set classified by sex age of the sample of the population to be tested;
  • Re-filtering treatment of each biological age-predicted biological indicator set classified by sex age wherein The biological indicator set of biological ages classified by sex age, when the sample size is greater than the number of biological age-predicted biological indicators, the highest 5% of the Euclidean Distance is removed; when the sample size is less than the biological age prediction When the number of biological indicators is removed, the highest 5% of the Mahalanobis Distance samples are removed, and the biological age predictive biological indicators are filtered using the variance expansion factor algorithm;
  • the 95% confidence interval for the biological age prediction is the gender age Group of aging assessment baselines
  • the detection method of the aging correlation index is separately described from the technical methods or processes of data acquisition; and the model for evaluating aging is described in detail: the biological age calculation model and the aging assessment baseline model, and the specific steps are as follows:
  • longevity gene loci To synthesize the latest research on longevity genes at home and abroad, use longevity gene loci to calculate the genetic background risk assessment index of individual biological age. Therefore, a long series of longevity genes that have been validated in the Chinese population were selected from the longevity gene database. Combining large-scale genotyping data from Chinese Han population, we have multiple longevity genes (such as SIRT1, APOE, The SNP locus on FOXO1,3, IL6, TOMM40, APOC1), through the Bayesian prior probability formula, established the OR-GRS risk assessment model, and calculated the risk contribution coefficient of each locus genotype to longevity. OR-GRS logistic regression expression. The regression equation is used to calculate the longevity risk index for each individual.
  • Table 1-1 The functions of the longevity gene and related loci are as follows:
  • Table 1-2 PCR primer amplification sequence information of longevity gene-related sites are as follows:
  • Instruments and consumables 9700, 96-well plate.
  • Buffer needs to be taken out and melted in advance. Mix well before use and centrifuge at low speed. After the reaction mixture was disposed, it was placed at room temperature.
  • reaction mixture configuration is performed at 10% loss.
  • the sample was placed in a PCR machine, and the following PCR reaction was carried out (note: the PCR apparatus required a hot lid), and the reaction system was 100 ⁇ L. If necessary, the sample after the reaction was placed in a refrigerator at -20 ° C.
  • OR-GRS Odds Ratio-Genetic Risk Scores
  • Mitochondria are at the center of metabolism and bioenergy conversion. Loss or mutation of mitochondrial DNA (mtDNA) leads to oxidative phosphorylation and abnormal energy supply, leading to defects in electron transport complexes or other substances that cause mitochondrial disorders, which may cause excessive reactive oxygen species (ROS) production, while ROS and DNA, RNA, and protein Or reactions such as lipids are likely to cause many pathological changes, causing aging or a variety of diseases, even cancer. Thus, the instability of the mitochondrial genome reflects the aging level of the body to a certain extent, and studies have shown that mitochondrial DNA copies serve as biological indicators of aging.
  • ROS reactive oxygen species
  • the Ct values of the mitochondrial gene and the internal reference gene were determined by real-time fluorescent quantitative PCR, respectively.
  • the copy number of the mitochondrial gene and the internal reference gene were calculated according to the curve simulated by the standard.
  • the ratio of mitochondrial (Mi) copy number to single copy gene (N), ie Mi/N ratio, can be calculated to give the relative copy number of mitochondria.
  • the standard must be set in each reaction, so the stability of the different batches of experiments can be evaluated by the curve of the standard.
  • Instruments and consumables 9700, 96-well plate.
  • Typesetting 5 pairs of standard products and samples for each pair of primers to make a double hole, that is, 1 sample for a total of 4 wells.
  • telomere and internal reference gene RPLPO ribosomal large subunit PO protein gene.
  • T telomere
  • S single copy gene
  • the T/S calculation formula is as follows:
  • T/S 2 - ⁇ CT , the value of which indicates the comparison of the telomere length of the sample with the standard product.
  • a value less than 1 indicates that the telomere length is smaller than the standard telomere length, and a value greater than 1 indicates that the telomere length is greater than Standard telomere length.
  • Typesetting Set 2 control samples, each sample has duplicate holes for each pair of primers, and one sample has 4 holes.
  • a hydrolysis reaction solution taking a certain amount of deionized water, adding 2 moles per liter of potassium acetate (Potassium Acetate), 1 mole per liter of magnesium acetate (Magnesium Acetate), 0.05 mole per liter of dithiothreitol ( DL-Dithiothreitol, DTT), 1 mole per liter of Tris-acetate having a pH of 7.9.
  • the ratios of methyl acetate, magnesium acetate, dithiothreitol, Tris-acetic acid and deionized water were 1:40, 1:100, 1:50, 1:50, respectively.
  • the ratio of DNA to the hydrolysis reaction solution is such that each 1 ug of DNA is dissolved in 100 ul of the hydrolysis reaction solution.
  • the final concentrations of the five gradient concentration points of mdC, hmdC, and dG are as follows.
  • A, T, C, G, dA, U, dC, mC are kept constant as background, and the final concentration is as follows.
  • the concentration calculated according to the quality control standard was derived, and the data stability and DNA global methylation level were analyzed by a Perl script.
  • the calculation formula is:
  • Sample preparation and loading transfer to the corresponding hole of the SPE board, and pressurize the column at normal temperature and normal pressure (or the low pressure of the column under the 96-well positive pressure extraction device, preferably the liquid outflow speed of 1 to 2 drops/second) After all the liquid has flowed out, give a large pressure to let the liquid flow out completely; discard the waste liquid; pay attention to keep the column wet.
  • 600uL ddH 2 O Replace the 2mL waste collection plate under the completely dried SPE plate with 2mL receiving plate 1. Add 600 ⁇ L of ultrapure water to each well, and pressurize the column at normal temperature and pressure (or 96-well positive pressure extraction). Under the device, the low pressure is over the column, and the liquid outflow speed of 1 ⁇ 2 drops/second is appropriate); finally, the large pressure is given for 1s to let the liquid completely flow out;
  • 600uLDCM (dichloromethane): replace the 2mL waste liquid collection plate under the completely dried SPE plate with 2mL receiving plate 2, add 600 ⁇ LDCM to each well, and pressurize the column at normal temperature and normal pressure (or 96-well positive pressure extraction device). The lower air pressure is passed through the column, and the liquid flow rate is preferably 1-2 drops/second.) Finally, a large pressure of 1 s is allowed to allow the liquid to completely flow out; then the high pressure is used to pass the column to completely dry the SPE column;
  • 600uL acetonitrile replace the collection plate 2 under the completely dried SPE plate with the collection plate 3, add 600 ⁇ L of acetonitrile to each well, and pass the column at normal temperature and normal pressure (or the low pressure column under the 96-well positive pressure extraction device). It is preferable to use a liquid effluent speed of 1 to 2 drops/second; finally, a large pressure of 1 s is allowed to allow the liquid to completely flow out; then, the SPE column is completely dried by using a high pressure column;
  • Hydrophilic phase 20 ⁇ L DCM mixed with acetonitrile, add 15 ul ddH2O, mix on a vortex mixer for 1 min, centrifuge at 25,000 rcf for 10 min, then take 33 ⁇ L into the injection bottle; wait for the machine to test.
  • the detected salivary hydrophobic hormones include 8 kinds, namely 25OHVD3, DHEA, DHEAS, E1, E2, F, T, P
  • the ion source is replaced by the Turbo Spray APCI source, the hardware is activated, the hormone detection method is called, and the pre-LC-MS/MS system is balanced for 20 minutes;
  • phase A ddH 2 O
  • phase B methanol
  • the detected salivary hydrophilic hormones include three kinds, namely 8-OHdG, CML, Melatonin
  • the ion source is changed to the ESI source, the hardware is activated, the saliva hydrophilicity index detection method is called, and the pre-LC-MS/MS system is balanced for 20 min;
  • phase A ddH 2 O
  • phase B methanol
  • the aging assessment model is divided into two sub-models: (a) First, establish a Biological Age Calculating Model (BACM) to calculate the biological age of each individual; (b) Second, establish a baseline estimation model for aging (Aging) The Baseline Evaluation Model (ABEM) calculates the aging baseline range of the sample and assesses the relative aging of each individual.
  • ABM Biological Age Calculating Model
  • ABEM Baseline Evaluation Model
  • test data corresponding to a series of samples is obtained, and it is judged whether it meets the first-level quality control standard;
  • test data conforming to the standard quality control shall be grouped according to the gender and age group.
  • the principles of grouping are as follows:
  • the indicators related to aging do not change linearly throughout the life course
  • VIF Variant Inflation Factor
  • the threshold is filtered by VIF>5 or VIF>10, that is, when VIF>5 or VIF>10, it is considered that there is strong collinearity between the corresponding biomarker and other biomarkers, then the BM is removed. .
  • C is the age of age CA
  • j is BM
  • m is the type of BMs
  • k j is the slope of each BM paired with CA
  • q j is the intercept of each BM to CA
  • x j is the j of sample x
  • the value of BM is the value of BM.
  • ⁇ 0 is the actual age C
  • ⁇ and ⁇ 0 are the standard deviation of the likelihood function and the prior function, respectively.
  • BA is the final calculated biological age.
  • the specific implementation process please refer to the flow chart of the biological age calculation model in Figure 1.
  • step 6.16 Repeat step 6.15 until there are no samples larger than the Cook's Distance threshold and then calculate the 95% confidence interval for the physiological age prediction for different age groups.
  • the biological age confidence interval is the calculated baseline range for aging assessment.
  • Figure 2 the aging baseline estimation model flow chart.
  • the screening of biological indicators and the relative aging of the individuals to be tested are performed based on 294 normal individual blood and urine samples, specifically:
  • This embodiment intends to use the sample multiple biological indicators to assess the physiological age and aging rate of the body.
  • the inventors collected 294 normal individual blood and urine samples, detected nearly 60 biological indicators, and successfully screened aging biological indicators based on the aforementioned "general method" of the present invention, using the aging evaluation model of the present invention.
  • the biological age algorithm screens biological indicators that are linearly associated with age, and uses these indicators to calculate the biological age.
  • the aging assessment baseline algorithm is used to eliminate outliers and establish biological age baselines for different genders and age groups. Assess the individual's aging status.
  • the biological indicators selected in this example cover genomic instability, telomere loss, mitochondrial function loss, hormone metabolism, protein homeostasis, and cell senescence and other aging factors.
  • the first biological indicators tested by technology development are methylation level (mcDNA-level), telomere length (MTL), mitochondrial copy number (mtCN), mitochondrial cumulative mutation (mtHP), 8-hydroxydeoxyguanosine (8- OHdG), carboxymethyllysine (CML), dehydroepiandrosterone (DHEA), dehydroepiandrosterone sulfate (DHEAS), 25 hydroxy VD3 (25OHVD3), hydrocortisone (F), estrone ( E1), estradiol (E2), progesterone (P), testosterone (T).
  • a blood sample of 10 volunteers was randomly selected, and the extracted DNA was subjected to second generation sequencing, and sequenced using BGISEQ500SE100 to obtain genotyping results as shown in the following table.
  • the longevity site contribution OR-GRS model calculation is performed. The longevity risk score values for each individual of S1-S10 were obtained.
  • telomere TEL
  • RPLPO ribosomal large subunit PO protein gene
  • the ratio of the telomere (T) repeat copy number to the single copy gene (S), that is, the T/S ratio, can be calculated to obtain the relative length of the telomeres, and the T/S ratio is proportional to the telomere length.
  • the T/S calculation formula is as follows:
  • T/S 2 - ⁇ CT , the value of which indicates the comparison of the telomere length of the sample with the standard product.
  • a value less than 1 indicates that the telomere length is smaller than the standard telomere length, and a value greater than 1 indicates that the telomere length is greater than Standard telomere length.
  • telomere length quantitative data is shown in the table below.
  • the stability of the method is determined, and on the other hand, the batch test.
  • the sample size of both groups is larger than the BM data, and the biological indicators are filtered using the VIF algorithm to obtain the male group and the female respectively.
  • a set of valid alternative BMs for the group is
  • the 6.6 age segment is divided into two phases.
  • the first phase is between 20 and 40 years old, with an interval of five years.
  • the second phase is for individuals older than 40 years.
  • the two gender groups are divided into five age groups. If the number of samples in the age group is less than the number of BM, no filtering is performed; for the age group, the number of samples larger than the number of BMs is also filtered using the VIF algorithm, resulting in an effective candidate BM set for biological age calculation.
  • Figure 4 Indicators associated with biological age in the 20-25 age group of women; in the age group of 25-29 years old, the indicators associated with biological age are LYM, LYMR, RDW-CV, LDL, UA ,TRIG,BMI,CML,F,mt-CN,MCHC,MSIZEDRATE,Creatinine,Melatonin,8-OHdG,HDL,PLCR,T,mcDNA-level,TBIL,E1,MCV,Height,P,FBG,Alkphosph,RBC , GGT, DHEAS, AST, PCT, etc., as shown in Figure 5, the age-related index of women in the 25-30 age group; in the 30-35 age group of women, the index associated with biological age is combined with MCV.
  • the indicators associated with biological age are RBC, Height, MCV, P, FBG, E1, Albumin, Urea, DHEA, 8-DHdG, MCHC, Creatinine, 25OHVD3, DHEAS, MTL, Melatonin, TRIG, LYMR, T, Globulin, Alkphosph, RDW-SD, HDL, LYM, GGT, LDL, F, PCT, PLCR, CML, BMI, etc.
  • the indicators associated with biological age are BMI, PDW, Height, PLCR, LDL, Melatonin, MTL, TRIG, DHEAS, mtCN, AST, PCT, CML, RDW-CV.
  • the indicators associated with biological age are 8-OHdG, Albumin, GGT, Globulin, PCT, AST, LDL, TRIG, BMI, E2, Hemoglobin, HDL, RDW-SD, Melatonin, UA, DHEAS, MTL, IBIL, T, CML, Alkphosph, Creatinine, LYMR, 25OHVD3, DHEA, etc., see Figure 10 for the age-related indicators in the male 25-30 age group; in the male 30-34 age group, The indicators associated with biological age are RDW-SD, WBC, Hemoglobin, Height, Creatinine, RDW-CV, CML, FBG, LYMR, Urea, Albumin,
  • the indicators associated with biological age are Albumin, T, IBIL, Height, MCHC, GGT, LDL, UA, MTL, BMI, Alkphosph, F, PCT, mtCN, Melatonin, Hemoglobin, PWD, PLCR, AST, Creatinine, Globulin, DHEA, 8-OHdG, E1, FBG, etc., are shown in Figure 13 for indicators associated with biological age in the male 40-100 age group.
  • the types of indicators related to biological age in different age groups are different, and the contribution of the same indicators to biological age in different age groups is also different.
  • a comparative analysis of the indicators of biological age associated between male and female groups found that there were significant differences in the types of indicators associated with biological age in different age groups in men and women, and the contribution of different indicators to aging in different age groups. It is also different. These indicators related to biological age are the selected aging indicators. The above results also prove that our aging assessment model strategy is feasible, which effectively indicates that biological age-related biological indicators will change with gender and age. different.
  • sample data of the age grouping selected in step 6.7 of the biological age calculation model is used. Judging each age group, when the number of samples is larger than the BM number, the highest 5% of the Euclidean Distance samples are removed, and the effective candidate BM set under the gender grouping is obtained; when the number of samples is less than the BM number, the horse is removed.
  • the highest 5% of the Mahalanobis Distance samples were filtered using the VIF algorithm described above.
  • the biological age confidence interval is the calculated baseline range for aging assessment. When an individual develops abnormal aging, it appears outside the baseline of the aging baseline, and in Figure 3, the outliers are off-base. When the individual is above the upper limit of the confidence interval, the biological age tends to be aging, and when the individual is below the lower limit of the confidence interval, the physiological age tends to be young.
  • the age of the age is 38.13 years according to the above steps, and the content or score of various biological indicators is obtained, as shown in the following table.
  • the BA calculated using the KD algorithm calculated using the biological age calculation model was 38.01 years old, within the 95% confidence interval; the final BA was calculated using the MAP algorithm to be 37.51 years old, within the 95% confidence interval.
  • the aging baseline range of the male 35-40 age group established using the aging assessment baseline model, which did not deviate from the overall range, was a better physiological health status (see the sample labeled in Figure d, Figure d, and Figure 14). This means that the biological biology of the sample is in the normal range, the body is in good condition, and it is in the normal range of aging in the population aging baseline.
  • the method for determining a biological age prediction biological indicator set of a population sample of the invention can be effectively used for determining a biological age prediction biological indicator set of a population sample, and then using the biological indicator set to accurately determine an individual biological age and a population
  • the aging assessment baseline of the sample, and further, based on the acquired individual biological age and the aging assessment baseline of the population sample can effectively assess the relative aging of the individual, and the assessment results are accurate and highly reliable.

Landscapes

  • Chemical & Material Sciences (AREA)
  • Organic Chemistry (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Zoology (AREA)
  • Wood Science & Technology (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Microbiology (AREA)
  • Immunology (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Biotechnology (AREA)
  • Biophysics (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Genetics & Genomics (AREA)
  • Investigating Or Analysing Biological Materials (AREA)

Abstract

提供了确定待测个体生物学年龄的方法及其应用,其中该方法包括以下步骤:根据前面所述的确定人群样本的生物学年龄预测生物指标集的方法,确定待测个体所属人群样本的以性别年龄分类的各生物学年龄预测生物指标集;基于待测个体对应的以性别年龄分类的生物学年龄预测生物指标集,计算所述待测个体的生物学年龄初步估计值DAEC;以及以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值DAEC进行最大后验概率计算处理,以便确定所述待测个体的预测生物学年龄DA。

Description

确定人群样本生物指标集、预测生物学年龄的方法及其应用
优先权信息
技术领域
本发明涉及生物技术领域,具体而言,涉及生物学年龄预测技术领域,更具体地涉及确定人群样本生物指标集、预测生物学年龄的方法及其应用。
背景技术
生物学年龄是与人体生长发育中的某些事件的出现时间有关,是根据正常人体生理学和解剖学的发育状态所推断出来的年龄,表明人体的组织结构和生理功能的实际状态。
生物年龄是人体健康状况的综合指数,是机体老化程度的客观表述。生物年龄可以与真实年龄不相符合,它的确定有各种各样的方法和模式。由于心血管系统的功能强烈依赖于人体的年龄,同时也反映着机体的健康状况。计算个体的生物学年龄,可以有效评估身体衰老状态,了解人体健康状况。
然而,现有技术对生物学年龄(Biological Age,BA)的计算和评估过于简单,仅通过简单的生化指标、心理测试报告等获得个体体质指标进行,做法过于粗糙,不够科学严谨。
因而,目前确定个体生物学年龄的方法仍有待进一步研究。
发明内容
本发明旨在至少解决现有技术中存在的技术问题之一。因而,本发明的目的至少在于提供一种能够准确确定个体生物学年龄和评估其衰老水平的手段。
需要说明的是,本发明是基于发明人的下列发现而完成的:
人的衰老是一个复杂并且缓慢的过程。随着人的生长发育,在不同时期(年龄段)会有不同的生理状态,反应或影响机体生物学年龄的指标数目或种类也会有所差异,对生物学年龄的贡献度也会变化;如在性别差异的生物学指标上,男性和女性在雌雄激素有着巨大的差异,而女性在40岁以后将进入更年期,性激素水平等多项生理指标也会有较大程度变化。因此,筛选有效的生物学指标评估不同性别生物学年龄是当前技术所面临的难点及创新点。
本发明拟通过提出筛选生物学年龄相关的指标算法,消除多指标拟合出现的共线性问题,并使用大规模人群死亡年龄分布数据,校正计算生物学年龄,作为个体生物学年龄,该策略即为本发明建立的生物学年龄计算模型。
此外,发明人还发现,目前一些研究或方案使用未做统计处理的多组指标预测生物学 年龄,仅比较个体的生物学年龄与年代年龄差异,最后仅使用大于或小于年代个体的年代年龄就判定机体异常衰老或年轻状态,过于简单武断。因此有效的衰老基线是评估衰老的难点和创新点。
因而,进一步,发明人尝试通过建立衰老评估基线模型,过滤离群样本,计算群体的生物学年龄置信区间,最终评估个体在群体衰老基线的相对状态,以达到指导或干预衰老的目的。
经过一系列科学的设计和精准的实验,发明人成功通过采集大规模样本的生物学指标数据,筛选到关联年龄或衰老的生物学指标,计算获得个体生物学年龄,建立得到不同组群(性别组,年龄段)的衰老基线,并最终实现了准确定量评估个体衰老水平的目的。
因而,在本发明的第一方面,本发明提供了一种确定人群样本的生物学年龄预测生物指标集的方法。根据本发明的实施例,该方法包括以下步骤:获得所述人群样本中所有个体的备选生物指标的数据;将所述人群样本中所有个体的备选生物指标数据按照性别进行分组,以便获得男性备选生物指标集和女性备选生物指标集;分别对所述男性备选生物指标集和所述女性备选生物指标集进行方差膨胀因子算法过滤,以便分别获得男性有效备选生物指标集和女性有效备选生物指标集;针对所述男性有效备选生物指标集和所述女性有效备选生物指标集的每一个,分别按照年龄段进行分组,以便获得多个不同年龄段的男性有效备选生物指标集和多个不同年龄段的女性有效备选生物指标集;分别确定所述多个不同年龄段的男性有效备选生物指标集和所述多个不同年龄段的女性有效备选生物指标集中每一个的有效生物指标集,以便获得多个不同年龄段的男性有效生物指标集和多个不同年龄段的女性有效生物指标集,其中,针对所述多个不同年龄段的男性有效备选生物指标集和所述多个不同年龄段的女性有效备选生物指标集的每一个,当其样本量大于有效备选生物指标的数目时,对其有效备选生物指标进行方差膨胀因子算法过滤,以便确定有效生物指标集;当其样本量小于有效备选生物指标的数目时,直接以其有效备选生物指标的集合作为有效生物指标集;以及将所述多个不同年龄段的男性有效生物指标集和所述多个不同年龄段的女性有效生物指标集,分别进行年代年龄关联性过滤,以便获得多个不同年龄段的男性生物学年龄预测生物指标集和多个不同年龄段的女性生物学年龄预测生物指标集,各个以性别年龄分类的生物学年龄预测生物指标集即为所述人群样本的生物学年龄预测生物指标集。
发明人惊奇地发现,利用该方法能够有效地确定人群样本的生物学年龄预测生物指标集,进而利用这些生物指标集可以准确地确定个体生物学年龄以及人群样本的衰老评估基线,并进一步,基于获得的个体生物学年龄以及人群样本的衰老评估基线能够有效地评估该个体的相对衰老程度,并且评估结果准确性好,可信度高。
在本发明的第二方面,本发明提供了一种确定待测个体生物学年龄的方法。根据本发明的实施例,该方法包括以下步骤:根据前面所述的确定人群样本的生物学年龄预测生物指标集的方法,确定待测个体所属人群样本的以性别年龄分类的各生物学年龄预测生物指标集;基于待测个体对应的以性别年龄分类的生物学年龄预测生物指标集,计算所述待测 个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000001
以及以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值
Figure PCTCN2017072675-appb-000002
进行最大后验概率计算处理,以便确定所述待测个体的预测生物学年龄BA。
发明人惊奇地发现,利用该方法能够准确有效地确定个体生物学年龄,进而基于人群样本的衰老评估基线,能够有效地评估该个体的相对衰老程度,并且评估结果准确性好,可信度高。
在本发明的第三方面,本发明提供了一种确定人群样本的衰老评估基线的方法。根据本发明的实施例,该方法包括以下步骤:根据前面所述的确定人群样本的生物学年龄预测生物指标集的方法,确定人群样本的以性别年龄分类的各生物学年龄预测生物指标集;对以性别年龄分类的各生物学年龄预测生物指标集分别进行再过滤处理,其中,针对以性别年龄分类的各生物学年龄预测生物指标集,当其样本量大于生物学年龄预测生物指标的数目时,去除欧式距离Euclidean Distance最高的5%的样本;当其样本量小于生物学年龄预测生物指标的数目时,去除马氏距离Mahalanobis Distance最高的5%的样本,并使用方差膨胀因子算法过滤生物学年龄预测生物指标;使用年代年龄CA对所有生物学年龄预测生物指标进行线性拟合,去除Cook’s Distance>1的样本,同时去除与CA关联性<0.1的生物学年龄预测生物指标,以便筛选出满足确立基线初始标准的样本和生物学年龄预测生物指标集;基于所述满足确立基线初始标准的样本和生物学年龄预测生物指标集,计算所述人群样本的每一个个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000003
以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值
Figure PCTCN2017072675-appb-000004
进行最大后验概率计算处理,以便确定每一个个体的预测生物学年龄BA;利用每一个个体的预测生物学年龄BA对年代年龄CA进行线性拟合,去除Cook’s Distance>1的样本,并重复此步骤至没有样本Cook’s Distance>1,以便筛选出满足确立基线要求的样本和生物学年龄预测生物指标集;以及基于所述满足确立基线要求的样本和生物学年龄预测生物指标集,计算各性别年龄组的生物学年龄预测的95%置信区间,所述生物学年龄预测的95%置信区间即为各性别年龄组的衰老评估基线。
根据本发明的实施例,利用该方法能够有效地确定人群样本的衰老评估基线,进而将待测个体的生物学年龄与其所处性别年龄组的衰老评估基线进行比较,能够有效地评估该待测个体的相对衰老程度,并且评估结果准确性好,可信度高。
在本发明的第四方面,本发明提供了一种确定待测个体相对衰老程度的方法。根据本发明的实施例,该方法包括以下步骤:根据前面所述的确定待测个体生物学年龄的方法,确定所述待测个体的生物学年龄BA;根据前面所述的确定人群样本的衰老评估基线的方法,确定所述待测个体所属人群样本的各性别年龄组的的衰老评估基线;以及将所述待测个体的生物学年龄BA与所述待测个体所处性别年龄组的衰老评估基线进行比较,以便确定 所述待测个体的相对衰老程度,其中,当所述待测个体的生物学年龄BA在所述待测个体所处性别年龄组的衰老评估基线的范围内时,判断所述待测个体相对所属人群样本处于正常衰老水平;当所述待测个体的生物学年龄BA偏离所述待测个体所处性别年龄组的衰老评估基线的范围时,判断所述待测个体相对所属人群样本处于异常衰老水平,其中,当所述待测个体的生物学年龄BA高于置信区间上限时,所述待测个体相对所属人群样本趋于衰老状态,当所述待测个体的生物学年龄BA低于置信区间下限时,所述待测个体相对所属人群样本趋于年轻状态。
发明人惊奇地发现,利用该方法能够有效地评估该待测个体的相对衰老程度,并且评估结果准确性好,可信度高。
本发明的附加方面和优点将在下面的描述中部分给出,部分将从下面的描述中变得明显,或通过本发明的实践了解到。
附图说明
本发明的上述和/或附加的方面和优点从结合下面附图对实施例的描述中将变得明显和容易理解,其中:
图1显示了根据本发明实施例的生物学年龄评估模型流程图;
图2显示了根据本发明实施例的衰老基线估计模型流程图;
图3显示了根据本发明的实施例,不同性别的生物学年龄BA及不同年龄群的衰老基线分布结果图;
图4显示了根据本发明的实施例,女性20-25岁年龄群中与生物学年龄关联的指标结果;
图5显示了根据本发明的实施例,女性25-30岁年龄群中与生物学年龄关联的指标结果;
图6显示了根据本发明的实施例,女性30-35岁年龄群中与生物学年龄关联的指标结果;
图7显示了根据本发明的实施例,女性35-40岁年龄群中与生物学年龄关联的指标结果;
图8显示了根据本发明的实施例,女性40-100岁年龄群中与生物学年龄关联的指标结果;
图9显示了根据本发明的实施例,男性20-25岁年龄群中与生物学年龄关联的指标结果;
图10显示了根据本发明的实施例,男性25-30岁年龄群中与生物学年龄关联的指标结果;
图11显示了根据本发明的实施例,男性30-35岁年龄群中与生物学年龄关联的指标结果;
图12显示了根据本发明的实施例,男性35-40岁年龄群中与生物学年龄关联的指标 结果;
图13显示了根据本发明的实施例,男性40-100岁年龄群中与生物学年龄关联的指标结果;
图14显示了根据本发明的实施例2,第167号样本测量的生物学年龄及衰老水平结果。
发明详细描述
下面详细描述本发明的实施例。下面通过参考附图描述的实施例是示例性的,仅用于解释本发明,而不能理解为对本发明的限制。
确定人群样本的生物学年龄预测生物指标集的方法
现有技术对生物学年龄(Biological Age,BA)评估过于简单,仅通过生化指标,心理测试报告等获得个体体质指标,未过滤真正影响生物学年龄的指标,而直接对人群进行了拟合计算得到所谓的心理年龄,免疫年龄等做法过于粗糙。另外,在指标选择上,并没有科学严谨的算法筛选与生物学年龄有关的指标。此外,未考虑不同性别和年龄段下,各指标对生物学年龄的贡献度差异;因此评估生物学指标。这些差异因素,一定程度上影响了生物学年龄的计算和衰老程度评估。
因而,发明人首先提供了一种相对于现有技术更科学严谨的用于人群样本的生物学年龄预测的生物指标集的筛选确定方法。具体地,在本发明的第一方面,本发明提供了一种确定人群样本的生物学年龄预测生物指标集的方法。根据本发明的实施例,该方法包括以下步骤:获得所述人群样本中所有个体的备选生物指标的数据;将所述人群样本中所有个体的备选生物指标数据按照性别进行分组,以便获得男性备选生物指标集和女性备选生物指标集;分别对所述男性备选生物指标集和所述女性备选生物指标集进行方差膨胀因子算法过滤,以便分别获得男性有效备选生物指标集和女性有效备选生物指标集;针对所述男性有效备选生物指标集和所述女性有效备选生物指标集的每一个,分别按照年龄段进行分组,以便获得多个不同年龄段的男性有效备选生物指标集和多个不同年龄段的女性有效备选生物指标集;分别确定所述多个不同年龄段的男性有效备选生物指标集和所述多个不同年龄段的女性有效备选生物指标集中每一个的有效生物指标集,以便获得多个不同年龄段的男性有效生物指标集和多个不同年龄段的女性有效生物指标集,其中,针对所述多个不同年龄段的男性有效备选生物指标集和所述多个不同年龄段的女性有效备选生物指标集的每一个,当其样本量大于有效备选生物指标的数目时,对其有效备选生物指标进行方差膨胀因子算法过滤,以便确定有效生物指标集;当其样本量小于有效备选生物指标的数目时,直接以其有效备选生物指标的集合作为有效生物指标集;以及将所述多个不同年龄段的男性有效生物指标集和所述多个不同年龄段的女性有效生物指标集,分别进行年代年龄关联性过滤,以便获得多个不同年龄段的男性生物学年龄预测生物指标集和多个不同年龄段的女性生物学年龄预测生物指标集,各个以性别年龄分类的生物学年龄预测生物指标集即为所述人群样本的生物学年龄预测生物指标集。
发明人惊奇地发现,利用该方法能够有效地确定人群样本的生物学年龄预测生物指标 集,进而利用这些生物指标集可以准确地确定个体生物学年龄以及人群样本的衰老评估基线,并进一步,基于获得的个体生物学年龄以及人群样本的衰老评估基线能够有效地评估该个体的相对衰老程度,并且所得结果准确性好,可靠度高。
根据本发明的一些优选实施例,所述备选生物指标为选自长寿基因、线粒体DNA拷贝数、端粒体长度、总体甲基化水平、激素水平的至少之一。其中,需要说明的是,本文中所述的“备选生物指标”可以为人群中任何可测量特征的生物指标,包含所有已知的和未知的与衰老关联的指标,不局限于上述的几种几大指标。此外,在早期研究中确定与衰老相关,像生理生化血液,未来可能检测到表征都是可以包含进来的。指标的范围应为很宽,本发明的方法(有时也称为模型)可以判别处理已知和未知指标与衰老或年龄是否相关,从而来确定计算生物学年龄的指标集。
根据本发明的实施例,所述人群样本中所有个体的所述生物指标的数据符合标准质控。
根据本发明的实施例,所述男性备选生物指标集和所述女性备选生物指标集中的所有生物指标数据均是同批次检测,或满足批次间检测CV值的要求,或不满足批次间检测CV值要求的已经使用LMM算法修正,并且,各组的样本量均应大于其备选生物指标的数目。由此,获得的人群样本的生物学年龄预测生物指标集可靠,应用于生物学年龄计算、人群样本衰老评估基线确定以及最终的个体相对衰老程度的确定时,准确性好、可信度高。
根据本发明的实施例,通过以下步骤,进行所述方差膨胀因子算法过滤:
(1)根据多重线性回归,基于下列公式,得到回归系数a0,a1,a2,...,am-1,am
Figure PCTCN2017072675-appb-000005
(2)根据所述回归系数a0,a1,a2,...,am-1,am,基于下列公式,求出偏差平方和q:
Figure PCTCN2017072675-appb-000006
(3)根据所述偏差平方和q,基于下列公式,求出复相关系数R:
Figure PCTCN2017072675-appb-000007
其中,
Figure PCTCN2017072675-appb-000008
Figure PCTCN2017072675-appb-000009
(4)根据所述复相关系数R,基于下列公式,求出每个备选生物指标或有效备选生物指标对应的VIF:
Figure PCTCN2017072675-appb-000010
以及
(5)采用VIF>5或VIF>10两种阈值对各备选生物指标或有效备选生物指标进行过滤, 其中,当VIF>5或VIF>10时,认为对应的生物学年龄相关指标与其他生物学年龄相关指标之间存在强共线性,则剔除该备选生物指标或有效备选生物指标。
由此,有利于后续步骤的进行,获得的人群样本的生物学年龄预测生物指标集可靠,应用于生物学年龄计算、人群样本衰老评估基线确定以及最终的个体相对衰老程度的确定时,准确性好、可信度高。
根据本发明的实施例,所述按照年龄段进行分组,是按照预定年龄段间隔进行的。此外,需要说明的是,年龄分组的年龄段间隔尺度理论上不设限,实际操作中可以依据所要研究的样本量和评估的精度选择。根据本发明的一些实施例,分组的预定年龄段间隔可以是5年或10年。根据本发明的一些具体示例,分组的预定年龄段间隔选择为5年,具体地,在20岁~40岁之间,每隔5年分为一组,而大于40岁的个体分为一组,这是因为:一方面是人在3-5年的尺度上个体年龄整体变化不会太大,另一方面是受限于样本量,年龄分组越细,每组样本就越少;至于40岁以上,理论上也是要年龄分组的,但实际实施例中大龄的样本取样不多;而20岁以下的样本太小,而忽略。
根据本发明的实施例,利用Pearson相关性分析法,基于以下公式,进行所述年代年龄关联性过滤:
Figure PCTCN2017072675-appb-000011
其中y表示年代年龄CA,x表示有效生物指标对应的检测值,
根据最大化保留有效生物指标的原则,当r<0.1时,过滤对应的有效生物指标。
由此,筛选得到的人群样本的生物学年龄预测生物指标集可靠,应用于生物学年龄计算、人群样本衰老评估基线确定以及最终的个体相对衰老程度的确定时,准确性好、可信度高。
应用
进而,发明人基于前述的确定人群样本的生物学年龄预测生物指标集的方法,即筛选生物学年龄相关的指标算法,消除多指标拟合出现的共线性问题,并使用大规模人群死亡年龄分布数据,校正计算生物学年龄,作为个体生物学年龄,该策略即为本发明建立的生物学年龄计算模型。
因而,在本发明的第二方面,本发明提供了一种确定待测个体生物学年龄的方法。根据本发明的实施例,该方法包括以下步骤:根据前面所述的确定人群样本的生物学年龄预测生物指标集的方法,确定待测个体所属人群样本的以性别年龄分类的各生物学年龄预测生物指标集;基于待测个体对应的以性别年龄分类的生物学年龄预测生物指标集,计算所述待测个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000012
以及以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值
Figure PCTCN2017072675-appb-000013
进行最大后验概率计算处理,以便确定所述待测个体的 预测生物学年龄BA。
发明人惊奇地发现,利用该方法能够准确有效地确定个体生物学年龄,进而基于人群样本的衰老评估基线,能够有效地评估该个体的相对衰老程度,并且评估结果准确性好,可信度高。
根据本发明的实施例,以KD模型法,基于以下公式,计算所述待测个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000014
Figure PCTCN2017072675-appb-000015
其中,
Figure PCTCN2017072675-appb-000016
是预测BA的值,C为年代年龄CA,
Figure PCTCN2017072675-appb-000017
为生物学年龄预测生物指标集BMs的方差,j为生物学年龄预测生物指标BM,m为生物学年龄预测生物指标集BMs的种类,kj为每个生物学年龄预测生物指标BM对CA拟合的斜率,qj为每个生物学年龄预测生物指标BM对年代年龄CA拟合的截距,xj为样本x的第j个生物学年龄预测生物指标BM的数值。
由此,有利于后续步骤的进行,最终确定的生物学年龄准确可靠。
根据本发明的实施例,按照以下步骤确定kj、qj和s:
(1)根据一元线性回归,基于以下公式,计算回归系数kj,qj(j=0,1,...,m-1):
y=kx+q,
Figure PCTCN2017072675-appb-000018
Figure PCTCN2017072675-appb-000019
其中
Figure PCTCN2017072675-appb-000020
(2)根据所述回归系数kj,qj(j=0,1,...,m-1),基于以下公式,计算偏差平方和q’:
Figure PCTCN2017072675-appb-000021
(3)根据偏差平方和q’,基于以下公式,计算平均标准偏差s:
Figure PCTCN2017072675-appb-000022
由此,有利于后续步骤的进行,最终确定的生物学年龄准确可靠。
根据本发明的实施例,基于以下公式进行最大后验概率计算,确定预测的生物学年龄BA:
Figure PCTCN2017072675-appb-000023
其中,
Figure PCTCN2017072675-appb-000024
为生理年龄初步估计值,μC为实际年龄C,σ与σ0分别为似然函数与先验函数标准差。由此,最终确定的生物学年龄准确可靠。
此外,现有技术中,一些研究或方案只简单计算生物学年龄,简单比较生物学年龄与年代年龄,大于或小于年代年龄,就说明异常衰老,不严谨且没有量化的范围基线。因此有效的衰老基线是评估衰老的难点和创新点。
而在本发明的第三方面,本发明提供了一种确定人群样本的衰老评估基线的方法。该方法科学严谨,确定的人群样本衰老评估基线可信度好,实用价值稿。
具体地,根据本发明的实施例,该方法包括以下步骤:根据前面所述的确定人群样本的生物学年龄预测生物指标集的方法,确定人群样本的以性别年龄分类的各生物学年龄预测生物指标集;对以性别年龄分类的各生物学年龄预测生物指标集分别进行再过滤处理,其中,针对以性别年龄分类的各生物学年龄预测生物指标集,当其样本量大于生物学年龄预测生物指标的数目时,去除欧式距离Euclidean Distance最高的5%的样本;当其样本量小于生物学年龄预测生物指标的数目时,去除马氏距离Mahalanobis Distance最高的5%的样本,并使用方差膨胀因子算法过滤生物学年龄预测生物指标;使用年代年龄CA对所有生物学年龄预测生物指标进行线性拟合,去除Cook’s Distance>1的样本,同时去除与CA关联性<0.1的生物学年龄预测生物指标,以便筛选出满足确立基线初始标准的样本和生物学年龄预测生物指标集;基于所述满足确立基线初始标准的样本和生物学年龄预测生物指标集,计算所述人群样本的每一个个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000025
以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值
Figure PCTCN2017072675-appb-000026
进行最大后验概率计算处理,以便确定每一个个体的预测生物学年龄BA;利用每一个个体的预测生物学年龄BA对年代年龄CA进行线性拟合,去除Cook’s Distance>1的样本,并重复此步骤至没有样本Cook’s Distance>1,以便筛选出满足确立基线要求的样本和生物学年龄预测生物指标集;以及基于所述满足确立基线要求的样本和生物学年龄预测生物指标集,计算各性别年龄组的生物学年龄预测的95%置信区间,所述生物学年龄预测的95%置信区间即为各性别年龄组的衰老评估基线。
根据本发明的实施例,利用该方法能够有效地确定人群样本的衰老评估基线,进而将待测个体的生物学年龄与其所处性别年龄组的衰老评估基线进行比较,能够有效地评估该待测个体的相对衰老程度,并且评估结果准确性好,可信度高。
根据本发明的实施例,以KD模型法,基于以下公式,计算每一个个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000027
Figure PCTCN2017072675-appb-000028
其中,
Figure PCTCN2017072675-appb-000029
是预测BA的值,C为年代年龄CA,
Figure PCTCN2017072675-appb-000030
为生物学年龄预测生物指标集BMs的方差,j为生物学年龄预测生物指标BM,m为生物学年龄预测生物指标集BMs的种类,kj为每个生物学年龄预测生物指标BM对CA拟合的斜率,qj为每个生物学年龄预测生物指标BM对年代年龄CA拟合的截距,xj为样本x的第j个生物学年龄预测生物指标BM的数值。
根据本发明的一些具体示例,按照以下步骤确定kj、qj和s:
(1)根据一元线性回归,基于以下公式,计算回归系数kj,qj(j=0,1,...,m-1):
y=kx+q,
Figure PCTCN2017072675-appb-000031
其中
Figure PCTCN2017072675-appb-000032
(2)根据所述回归系数kj,qj(j=0,1,...,m-1),基于以下公式,计算偏差平方和q’:
Figure PCTCN2017072675-appb-000033
(3)根据偏差平方和q’,基于以下公式,计算平均标准偏差s:
Figure PCTCN2017072675-appb-000034
根据本发明的实施例,基于以下公式进行最大后验概率计算,确定预测的生物学年龄 BA:
Figure PCTCN2017072675-appb-000035
其中,
Figure PCTCN2017072675-appb-000036
为生理年龄初步估计值,μC为实际年龄C,σ与σ0分别为似然函数标准差与先验函数标准差。
根据本发明的实施例,基于线性拟合结果对所述人群样本进行筛选包括:去除Cook’s Distance>1的样本。由此,筛选效果好,最终确定的人群样本衰老评估基线可靠,应用价值高。
在本发明的第四方面,本发明提供了一种确定待测个体相对衰老程度的方法。根据本发明的实施例,该方法包括以下步骤:根据前面所述的确定待测个体生物学年龄的方法,确定所述待测个体的生物学年龄BA;根据前面所述的确定人群样本的衰老评估基线的方法,确定所述待测个体所属人群样本的各性别年龄组的的衰老评估基线;以及将所述待测个体的生物学年龄BA与所述待测个体所处性别年龄组的衰老评估基线进行比较,以便确定所述待测个体的相对衰老程度,其中,当所述待测个体的生物学年龄BA在所述待测个体所处性别年龄组的衰老评估基线的范围内时,判断所述待测个体相对所属人群样本处于正常衰老水平;当所述待测个体的生物学年龄BA偏离所述待测个体所处性别年龄组的衰老评估基线的范围时,判断所述待测个体相对所属人群样本处于异常衰老水平,其中,当所述待测个体的生物学年龄BA高于置信区间上限时,所述待测个体相对所属人群样本趋于衰老状态,当所述待测个体的生物学年龄BA低于置信区间下限时,所述待测个体相对所属人群样本趋于年轻状态。
如前所述,利用该方法,能够有效地确定待测个体的生物学年龄、人群样本的衰老评估基线,进而将待测个体的生物学年龄与其所处性别年龄组的衰老评估基线进行比较,能够有效地评估该待测个体的相对衰老程度,并且评估结果准确性好,可信度高。
此外,在本发明的另一方面,本发明还提供了适于实施前面所述各方法的装置、设备和系统,例如,与本发明的确定人群样本的生物学年龄预测生物指标集的方法对应的,适于实施该方法,用于确定人群样本的生物学年龄预测生物指标集的装置;与本发明的确定待测个体生物学年龄的方法对应的,适于实施该方法,用于确定待测个体生物学年龄的方法的设备;与本发明的确定人群样本的衰老评估基线的方法对应的,适于实施该方法,用于确定人群样本的衰老评估基线的设备;与本发明的确定待测个体相对衰老程度的方法对应的,适于实施该方法,用于确定待测个体相对衰老程度的系统。其中,可以理解的是,上述的适于实施各方法的装置、设备和系统,其结构组成均是包含了适于实施各个步骤的相应单元、装置或设备。
具体地,以本发明的确定待测个体生物学年龄的方法为例,其对应的设备——用于确定待测个体生物学年龄的方法的设备,包括:用于确定人群样本的生物学年龄预测生物指 标集的装置,用于根据前面所述的确定人群样本的生物学年龄预测生物指标集方法,确定待测个体所属人群样本的以性别年龄分类的各生物学年龄预测生物指标集;生物学年龄初步估计值计算装置,所述生物学年龄初步估计值计算装置与所述用于确定人群样本的生物学年龄预测生物指标集的装置相连,用于基于待测个体对应的以性别年龄分类的生物学年龄预测生物指标集,计算所述待测个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000037
以及生物学年龄预测装置,所述生物学年龄预测装置与所述生物学年龄初步估计值计算装置相连,用于以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值
Figure PCTCN2017072675-appb-000038
进行最大后验概率计算处理,以便确定所述待测个体的预测生物学年龄BA。
前述的本发明的各方法的特征和优点同样适用于与其对应的本发明的用于确定人群样本的生物学年龄预测生物指标集的装置、用于确定待测个体生物学年龄的方法的设备、用于确定人群样本的衰老评估基线的设备以及用于确定待测个体相对衰老程度的系统,再次不再赘述。
根据本发明的实施例,还需要说明的是,本发明具有下列效果的至少之一:
(1)开发的液体活检技术(唾液和血液),能稳定高效检测生物学指标;
(2)基于人群大样本的数据,使用严谨的生物学年龄计算模型,能够准确计算生物学年龄;
(3)基于人群大样本的数据,能够建立不同性别、不同年龄段的衰老评估基线,并能够评估个体在群体下的相对衰老状态,从而能够科学规范的对异常衰老给出建议干预措施。
下面将结合实施例对本发明的方案进行解释。本领域技术人员将会理解,下面的实施例仅用于说明本发明,而不应视为限定本发明的范围。实施例中未注明具体技术或条件的,按照本领域内的文献所描述的技术或条件(例如参考J.萨姆布鲁克等著,黄培堂等译的《分子克隆实验指南》,第三版,科学出版社)或者按照产品说明书进行。所用试剂或仪器未注明生产厂商者,均为可以通过市购获得的常规产品,例如可以采购自Illumina公司。
一般方法:
本发明的确定待测个体相对衰老程度的方法一般包括以下步骤:
根据本发明的确定人群样本的生物学年龄预测生物指标集的方法,确定待测个体所属人群样本的以性别年龄分类的各生物学年龄预测生物指标集;
基于待测个体对应的以性别年龄分类的生物学年龄预测生物指标集,计算所述待测个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000039
以及
以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值
Figure PCTCN2017072675-appb-000040
进行最大后验概率计算处理,以便确定所述待测个体的预测生物学年龄BA;
对以性别年龄分类的各生物学年龄预测生物指标集分别进行再过滤处理,其中,针对 以性别年龄分类的各生物学年龄预测生物指标集,当其样本量大于生物学年龄预测生物指标的数目时,去除欧式距离Euclidean Distance最高的5%的样本;当其样本量小于生物学年龄预测生物指标的数目时,去除马氏距离Mahalanobis Distance最高的5%的样本,并使用方差膨胀因子算法过滤生物学年龄预测生物指标;
使用年代年龄CA对所有生物学年龄预测生物指标进行线性拟合,去除Cook’s Distance>1的样本,同时去除与CA关联性<0.1的生物学年龄预测生物指标,以便筛选出满足确立基线初始标准的样本和生物学年龄预测生物指标集;
基于所述满足确立基线初始标准的样本和生物学年龄预测生物指标集,计算所述人群样本的每一个个体的生物学年龄初步估计值
Figure PCTCN2017072675-appb-000041
以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值
Figure PCTCN2017072675-appb-000042
进行最大后验概率计算处理,以便确定每一个个体的预测生物学年龄BA;
利用每一个个体的预测生物学年龄BA对年代年龄CA进行线性拟合,去除Cook’s Distance>1的样本,并重复此步骤至没有样本Cook’s Distance>1,以便筛选出满足确立基线要求的样本和生物学年龄预测生物指标集;
基于所述满足确立基线要求的样本和生物学年龄预测生物指标集,计算各性别年龄组的生物学年龄预测的95%置信区间,所述生物学年龄预测的95%置信区间即为各性别年龄组的衰老评估基线;以及
将所述待测个体的生物学年龄BA与所述待测个体所处性别年龄组的衰老评估基线进行比较,以便确定所述待测个体的相对衰老程度,其中,当所述待测个体的生物学年龄BA在所述待测个体所处性别年龄组的衰老评估基线的范围内时,判断所述待测个体相对所属人群样本处于正常衰老水平;当所述待测个体的生物学年龄BA偏离所述待测个体所处性别年龄组的衰老评估基线的范围时,判断所述待测个体相对所属人群样本处于异常衰老水平,其中,当所述待测个体的生物学年龄BA高于置信区间上限时,所述待测个体相对所属人群样本趋于衰老状态,当所述待测个体的生物学年龄BA低于置信区间下限时,所述待测个体相对所属人群样本趋于年轻状态。
实施例1
本实施例从数据获取的技术方法或流程等方面分别阐述衰老关联指标的检测方法;并详细阐述评估衰老的模型:生物学年龄计算模型和衰老评估基线模型,具体步骤如下:
一、长寿基因对生物学年龄贡献度
综合国内外最新长寿基因研究进展,使用长寿基因位点用以计算个体生物学年龄的遗传背景风险评估指数。因此从长寿基因数据库选取经典的一系列在中国人群中有研究验证的长寿基因。结合中国汉族人群大规模基因分型数据,我们多个长寿基因(如SIRT1、APOE、 FOXO1,3、IL6、TOMM40,APOC1)上的SNP位点,通过贝叶斯先验概率公式,建立了OR-GRS风险评估模型,计算了每个位点基因型对长寿的风险贡献系数,得到OR-GRS逻辑回归表达式。使用回归方程计算每个个体的长寿风险指数。
表1-1 长寿基因的功能及相关位点具体信息如下:
Figure PCTCN2017072675-appb-000043
表1-2 长寿基因相关位点PCR引物扩增序列信息如下:
Figure PCTCN2017072675-appb-000044
Figure PCTCN2017072675-appb-000045
a)实验流程:
1.DNA制备
参考QIAGEN血液DNA提取试剂盒DNA Blood Mini Kit说明书,进行血液DNA制备,并做浓度测定;
2.PCR扩增
按以下条件进行PCR扩增:
PCR体系:
Figure PCTCN2017072675-appb-000046
PCR程序:
Figure PCTCN2017072675-appb-000047
仪器及耗材:9700,96孔板。
3.PCR产物pooling。
4.将每个样本10个PCR产物按1:1混合到一起,然后按5个不同标签的PCR混合物按1:1混合,并记录每个混合产物对应的样本。
5.pooling后PCR产物纯化
每个pooling后样本用1.5倍体积的AmpureXP Beads进行纯化,最后溶于39.5μl EB中,具体操作步骤如下:
5.1将磁珠震荡混匀,室温静置30min;
5.2加入样本体积1-1.5倍的磁珠,并吹打混匀(或封膜震荡)静置吸附5-10min,放磁力架吸附2-3min,吸掉废液;
5.3加入500ul 75%乙醇(板子加100ul),封口后颠倒冲洗10次,放磁力架吸附2-3min,吸掉废液;
5.4加入500ul 75%乙醇(板子加100ul),封口后颠倒冲洗10次,放磁力架吸附2-3min,吸掉废液;
5.5 37摄氏度烘干3-5min(观察到磁珠干裂即可);
5.6加入30-50ul水或Elution Buffer进行溶解,5000rpm后震荡,再次短离心,静置5min,放磁力架吸附2-3min。
5.7将下清液转至一新EP管保存。
6.文库构建
每个纯化后的pooling样本取200ng,按建库流程进行建库:
6.1末端修复(End-Repaired)
a)预先从-20℃保存的试剂盒中取出10x NEBNext ER Buffer和10mM dNTPs mix,将其置于冰上融解并充分混匀10x NEBNext ER Buffer。
b)在1.5ml的离心管中配制末端修复反应体系:
DNA(来自4的纯化产物) 10μL
10x NEBNext ER Buffer 7μL
T4 DNA Pol(3U/ul) 1.87μL
T4 PNK(10U/ul) 1.40μL
BSA(10mg/ml) 0.8μL
48.93μL
总体积 70μL
c)加入上述Mix后,轻轻震荡均匀,短暂离心,并置于20℃中孵浴30min。
d)用1.5倍体积的Axygen beads进行纯化,回溶40μl TE中,操作步骤同上。
6.2末端加“A”(A-Tailing)
a)预先从-20℃保存的试剂盒中取出10x NEBuffer2和100mM dATP,将其置于冰上融解并充分混匀。
b)在1.5ml的离心管中配制末端加“A”反应体系:
H2O 11.45μL
10x NEBuffer2 6μL
100mM dATP 0.15μL
Klenow Exo-(5U/ul) 2.4μL
总体积 20μL
c)在Thermomixer中37℃温浴30min。
d)用1.5倍体积(75μl)的Axygen beads进行纯化,回溶38μl TE中,操作步骤同上。
6.3 Adapter的连接(Adapter Ligation)
a)预先从-20℃保存的试剂盒中取出3x HB buffer和T-tailedΩadapter153,将其置于冰上融解并充分混匀。
b)在1.5ml的离心管中配制Adapter连接反应体系:
DNA 38μL
25μM T-tailedΩadapter153 2μL
H2O 1.24μL
3x HB buffer 25μL
8bp Index PE PCR-free Adapter oligo mix(50uM) 2μL
T4 DNA Ligase(600U/ul) 3.76μL
总体积 70μL
c)在Thermomixer中20℃温浴60min。
d)用35μL EDTA终止反应。然后用0.5倍体积的Axygen beads进行纯化,回溶25μl TE中,操作步骤同上。
6.4连接产物扩增(PCR)
提前5分钟左右准备PCR混合液涡旋,混匀:
Figure PCTCN2017072675-appb-000048
注:Buffer需要提前取出后融化,使用前震荡混匀,低速离心。反应混合物配置完成后置于常温。
如果需要配置多个样本反应,按照10%损耗进行反应混合物配置。
向80μL PCR混合物中加入20μl纯化的接头连接产物,枪头吹吸8次以混匀。
将样品置入PCR仪中,进行如下PCR反应(注意:PCR仪需要热盖),反应体系100μL,如需保存可将反应后样品置于-20℃冰箱。
Figure PCTCN2017072675-appb-000049
6.5扩增样品纯化(PCR纯化)
将已完成AdA扩增反应的样品,转移至新1.5ml不粘管中,加入100μL XP/Tween20的混合液,枪头吹打混匀磁珠和样品的混合液7-10次,室温结合5分钟后,再次用枪头吹打混匀7-10次,室温结合5分钟后置于磁力架上结合2分钟(至液体澄清),小心吸弃上清。
在MPC上向不粘管里加入500μL 75%乙醇,盖紧管盖,上下颠倒混匀5次,弃掉上清;500μL 75%乙醇重复洗1次,用小量程的移液器尽可能弃掉残留的乙醇,室温晾干。
用47μL TE/Tween重悬磁珠,枪头吹打混匀7-10次,室温结合5分钟后,再次用枪头吹打混匀7-10次,室温结合5分钟后置于磁力架上结合2分钟(至液体澄清),小心吸出45μL上清至新的1.5mL EP管中,准备进行下一步反应或保存至-20℃冰箱
6.6扩增样本质检(PCR质量检测)
取3μL PCR产物送交文库质检组进行Agilent 2100检测。
7.建立风险评估OR-GRS模型
基于优势比的遗传风险模型评估(Odds Ratio-Genetic Risk Scores,OR-GRS)模型计算:
(1)计算每个SNP的OR权重风险因子ω
ωORi=ln(ORi)
(2)计算每个SNP的风险等位基因型出现频次Gi及GRS
Figure PCTCN2017072675-appb-000050
(3)计算每个SNP基因型发生时,样本条件概率P(D=1|Gi)
Figure PCTCN2017072675-appb-000051
例如:当i=1,即SNP1基因型为GA时,
Figure PCTCN2017072675-appb-000052
(4)模型建立
Figure PCTCN2017072675-appb-000053
其中,Gi表示基因型贡献度(根据OR值,取Gi=0,1,2);α为常数项,β为各个位点多元回归系数,ω为OR权重风险因子。
(5)计算模型的参数
Figure PCTCN2017072675-appb-000054
(Notes:1.Signif.codes:0'***'0.001'**'0.01'*'0.05'.'0.1”1;2.Dispersion parameter for Gaussian family taken to be 0.01856851)
长寿基因型对生物学年龄贡献度的计算表达式:
LogitP(D=1|G)=0.032-0.49*G1+7.96*G2+0.64*G3-1.72*G4-1.70*G5-0.19*G6-0.96*G7+1.90*G8+0.71*G9+0.86*G10
二、线粒体DNA拷贝数
线粒体在新陈代谢和生物能量转换中处于中心地位。线粒体DNA(mtDNA)缺失或突变会导致氧化磷酸化及能量供应异常,导致电子传递复合物缺陷或其他引起线粒体紊乱的物质都可能造成过量活性氧(ROS)产生,而ROS与DNA、RNA、蛋白质或脂类等反应容易导致许多病理变化,从而引起衰老或多种疾病发生,甚至癌症。因而线粒体基因组的不稳定性在一定程度上反应机体衰老水平,研究表明线粒体DNA拷贝作为衰老的生物学指标。
2.1线粒体拷贝数检测—Mi/N法
1)引物:
2)标准品:
Figure PCTCN2017072675-appb-000056
3)检测方法及原理:
实时荧光定量PCR分别测定线粒体基因和内参基因(GAPDH基因)的Ct值。根据标准品模拟的曲线,分别计算线粒体基因和内参基因的拷贝数。最后计算线粒体(Mi)拷贝数与单拷贝基因(N)的比率,即Mi/N比率可以得出线粒体的相对拷贝数。标准品必须设置在每次上机反应中,因此可以通过标准品构成的曲线来评估不同批次实验间的稳定性。
4)试剂:KAPA SYBR Fast ABI
Figure PCTCN2017072675-appb-000057
5ml
5)仪器:StepOnePlusTM实时荧光定量PCR系统
6)实验操作流程:
DNA制备并稀释至5-20ng/μL;
体系配制与Q-PCR上机
QPCR体系(10μL体系):
试剂/样本 用量(单个反应)
KAPASYBR FASTqPCR Kit Master Mix(2X) 5μL
Primer Mix(1P/μL) 1μL
DNA 2μL
2μL
7)仪器及耗材:9700,96孔板。
8)排版:5对标准品及样本每对引物做复孔1次,即1个样本共4孔反应。
三、端粒长度定量
端粒长度检测—T/S法
1)引物:
Figure PCTCN2017072675-appb-000058
2)检测方法及原理:
实时荧光定量PCR分两部分进行,分别测定端粒和内参基因RPLPO(核糖体大亚基PO蛋白基因)的Ct值。端粒(T)重复拷贝数与单拷贝基因(S)的比率,即T/S比率可以得出端粒的相对长度,而T/S比率与端粒长度成正比关系。
T/S计算公式如下:
T/S=[2CT(telomeres)/2CT(single copy gene)]=2-ΔCT
引入标准品后,T/S=2-ΔΔCT,其数值表示样本端粒长度与标准品的比较短息,小于1表示其端粒长度小于标准品端粒长度,大于1表示其端粒长度大于标准品端粒长度。引入多个标准品可以通过标准品T/S值来评估不同批次实验间的稳定性。
3)试剂:KAPA SYBR Fast ABI
Figure PCTCN2017072675-appb-000059
5ml。
4)仪器:StepOnePlusTM实时荧光定量PCR系统。
5)实验操作流程:DNA制备并稀释至5-20ng/μL;
体系配制与Q-PCR上机
QPCR体系(10μL体系):
试剂/样本 用量(单个反应)
KAPASYBR FASTqPCR Kit Master Mix(2X) 5μL
Primer Mix(1P/μL) 2μL
DNA 2μL
1μL
6)仪器及耗材:9700,96孔板。
7)排版:设2个对照样本,每个样本每对引物做复孔,一个样本共4孔。
四、总体甲基化水平检测
4.1 DNA水解
a)配制水解反应液:取一定量的去离子水,往其中加入2摩尔每升乙酸钾(Potassium Acetate)、1摩尔每升乙酸镁(Magnesium Acetate)、0.05摩尔每升二硫苏糖醇(DL-Dithiothreitol,DTT)、1摩尔每升pH为7.9的Tris-乙酸(Tris-acetate)。乙酸甲、乙酸镁、二硫苏糖醇、Tris-乙酸与去离子水的比例分别为1:40、1:100、1:50、1:50。然后 按照每50ul上述溶液中加入2U脱氧核糖核酸酶I(Deoxyribonuclease I,DNase I),2U Lambda核酸外切酶(Lambda Exonuclease),2U虾碱性磷酸酶(Shrimp Alkaline Phosphatase,SAP)的比例配制成水解反应液。
b)取一定量的DNA加入到水解反应液中,混匀。DNA与水解反应液的比例为,每1ug DNA溶于100ul水解反应液中。
c)将以上混匀溶液,置于37℃条件下,孵育2h。
d)水解完成,水解产物可以通过1%琼脂糖凝胶电泳检测水解情况。
e)将水解产物放在-20℃保存,待质谱上机检测。
4.2液相色谱质谱联用检测
a)优化质谱条件。选用Q-Trap4500仪器,参数设置如下:
离子源 Turbo V
模式 正离子模式
IS 4500
CUR 30
CAD Medium
GS1 45
GS2 40
Tem 400
b)以单个核苷酸化合物的标准品为样本,构建化合物母离子/子离子对:
Figure PCTCN2017072675-appb-000060
c)优化液相条件。ACQUITY
Figure PCTCN2017072675-appb-000061
I-Class仪器,ACQUITY
Figure PCTCN2017072675-appb-000062
HSS T3色谱柱(2.1×100mm,1.7u)。配置A、B两种试剂做流动相。A:0.1%甲酸(水溶),B:0.1%甲酸(甲醇溶)。A、B按梯度混合做流动相。
配置标准曲线。mdC,hmdC,dG的5个梯度浓度点终浓度如下所示。A,T,C,G,dA,U,dC,mC保持恒定做背景,终浓度如下所以。
Figure PCTCN2017072675-appb-000063
d)标准曲线、QC及样本上机。每针上样5ul,首先重复两针wash,再按std1到std5的顺序上样标曲,wash一针,将QC、样本上机(QC间隔插入样本中)。
4.3下机数据处理
选用MultiQuant(AB SCIEX)积峰软件。设置:算法MQ4,高斯平滑2点。质控标准:S/N>3;Accuracy>80&<120;CV<15%;R2>0.90。
将按照质控标准计算出的浓度导出,通过Perl脚本分析数据稳定性及DNA全局甲基化水平。计算公式为:
Figure PCTCN2017072675-appb-000064
五、激素指标定量
5.1唾液前处理
a)样品准备
1)将所有样品按编排好的样品表,依次摆放在EP管架对应位置上;
2)按待测唾液样品计算,取2管去离子水用以配制空白和若干分装好的有效期内的混合唾液样品,每管800μL,用于配制QC;
3)唾液样本解冻后离心,或移取850ul上清后离心,25,000rcf 10min。
4)取离心后唾液样品800μL转移至1号EP管架上相应的EP管中备用,剩余唾液样品-80暂存到出结果;
5)一个人负责取样品,另一个人负责摆放样品及复核,不一致的部分需在样品表中标注;
6)在已经取好的样品、QC及空白中加入400μL超纯水和10μL同位素内标(ISTD,双空白孔不用加内标),用吸头吹打混匀;
7)准备与所有样品相应数目的内衬管、贴上标签纸的样品瓶和换好新衬垫的瓶盖。
b)SPE板活化与平衡
1)活化:常温打开提取试剂盒,取出SPE板,置于2mL废液收集板上。在将要用到的孔中依次加入1mL乙腈、甲醇,分别常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜);
2)平衡:SPE板置于2ml废液收集板,在将要用到的孔中加入1mL超纯水,常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜),重复2次,共过3次1ml超纯水;
c)样本loading与wash
1)样本准备与Loading:转移至SPE板的相应孔中,常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜),液体全部流完后,给大压力让液体完全流出;弃废液;注意保持柱子的湿润状态。
2)Wash:加200μL超纯水于各孔中,常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜),最后给1s大压力让液体完全流出;将96孔SPE板置于96孔正压提取装置下,使用高气压过柱,使SPE柱完全干燥。
d)萃取
1)600uL ddH2O:将完全干燥的SPE板下的2mL废液收集板换成2mL接收板1,在各孔中加入600μL超纯水,常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜);最后给1s大压力让液体完全流出;
2)600uL ddH2O Wash:将SPE板下的接收板1换成2mL废液收集板,在各孔中加入600μL超纯水,常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜),最后给1s大压力让液体完全流出,然后使用高气压过柱使SPE柱完全干燥;
3)600uLDCM(二氯甲烷):将完全干燥的SPE板下的2mL废液收集板换成2mL接收板2,在各孔中加入600μLDCM,常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜);最后给1s大压力让液体完全流出;然后使用高气压过柱使SPE柱完全干燥;
4)600uL乙腈:将完全干燥的SPE板下的收集板2换成收集板3,在各孔中加入600μL乙腈,常温常压过柱(或于96孔正压提取装置下低气压过柱,以1~2滴/秒的液体流出速度为宜);最后给1s大压力让液体完全流出;然后使用高气压过柱使SPE柱完全干燥;
e)抽干、复溶
1)抽干:接收板中的萃取液转移至编好号的新1.5mL离心管中或直接在收集板中,置于冷冻真空浓缩仪中离心抽干;
2)复溶:抽干后,600uL ddH2O加25ul柠檬酸铵复溶液;600uLDCM和600uL乙腈分别加30ul柠檬酸铵复溶液,在漩涡混合器上震荡5min,简单离心后混合。
f)合并液体
1)疏水相:DCM与乙腈混合液,25,000rcf离心10min后,取22ul于进样瓶;等待上机检测。
2)亲水相:20μL DCM与乙腈混合液,加15ul ddH2O相混合,漩涡混合器上震荡1min,25,000rcf离心10min后,取33μL于进样瓶;等待上机检测。
5.2质谱检测
a)疏水类激素检测
检测的唾液疏水类激素包括8种,即25OHVD3、DHEA、DHEAS、E1、E2、F、T、P
1)选用QTRAP5500质谱仪,换上色谱柱Hydro-RP,150mm×2mm,4μm,需要加上保护柱;
2)配制流动相和洗针液
3)将样品在自动进样器的进样盘中放好,做好瓶位置的记录
4)确认仪器质控合格后,离子源换为Turbo Spray APCI源,激活硬件,调用荷尔蒙检测方法,平衡预LC-MS/MS系统20min;
参数设置:
Figure PCTCN2017072675-appb-000065
5)连续进三针STD,若符合质量标准与评估中的MRM Intensity和柱效的上机质控标准,则开始上机样本;
6)建立batch和run样品
7)上机结束后,流动相换为ddH2O(A相)和甲醇(B相),先用高水相(90%A相)冲洗色谱柱20min,然后用高有机相(90%B相)冲洗20min,保存
8)数据上传:当质控数据正常后,需连同样本数据一起上传至大型机
b)亲水类激素检测
检测的唾液亲水类激素包括3种,即8-OHdG、CML、Melatonin
1)选用QTRAP5500质谱仪,换上色谱柱C8(kinetexR 2.6μm C8 100A),需要加上保护柱
2)配制流动相和洗针液
3)将样品在自动进样器的进样盘中放好,做好瓶位置的记录
4)确认仪器质控合格后,离子源换为ESI源,激活硬件,调用唾液亲水性指标检测方法,平衡预LC-MS/MS系统20min;
参数设置:
Figure PCTCN2017072675-appb-000066
Figure PCTCN2017072675-appb-000067
5)连续进三针STD,若符合质量标准与评估中的MRM Intensity和柱效的上机质控标准,则开始上机样本;
6)建立batch和run样品。
7)上机结束后,流动相换为ddH2O(A相)和甲醇(B相),先用高水相(90%A相)冲洗色谱柱20min,然后用高有机相(90%B相)冲洗20min,保存。
8)数据上传:当质控数据正常后,需连同样本数据一起上传至大型机。
六、衰老评估模型(Aging Evaluation Model,AEM)建立
衰老评估模型具体分为两个子模型:(a)首先,建立生物学年龄计算模型(Biological Age Calculating Model,BACM)计算每个个体的生物学年龄;(b)其次,建立衰老基线估计模型(Aging Baseline Evaluation Model,ABEM)计算样本的衰老基线范围,评估每个个体相对衰老程度。
(a)生物学年龄计算模型实施步骤如下:
6.1首先,通过实验和测序得到一系列样本对应的检测数据,并判断是否符合一级质控标准;
6.2将符合标准质控的检测数据,按照性别以及年龄段对样本数据进行分组,分组依据的原则如下:
1、一些指标对不同性别的样本影响效度不同;
2、与衰老相关的指标在整个生命历程中,并非呈线性变化;
3、相同年龄间隔进行年龄段的划分。
6.3我们首先依据性别分组,并对批次内获得数据进行质检,判断每个组别中各项BMs是否符合同批检测,对不同批次获得数据做批次间CV计算,若CV值不满足要求,我们将对批次间获得数据使用线性混合模型(Linear mixed model,LMM)修正,此模型使用的前提要求批次内样本满足LMM要求,LMM算法可去除批次间误差。具体做法是对每个点加一修正项。修正项的值为全样本拟合结果与该数据点所在批次的批次内拟合结果之差。
6.4使用满足同批检测的,或满足要求批次间CV值的,或最终使用LMM算法修正的样本及BMs,当样本量大于BMs数目时,进行方差膨胀因子算法过滤(Variance Inflation Factor,VIF),否则提示样本量过少并中断分析。
6.5针对每个性别组,进行VIF过滤,获得有效的性别分组BMs集,即BS,具体算法如下:
(1)根据多重线性回归(multiple linear regression,MLR),即:
Figure PCTCN2017072675-appb-000068
得到回归系数a0,a1,a2,...,am-1,am
(2)根据回归系数,求出偏差平方和q,即:
Figure PCTCN2017072675-appb-000069
(3)根据偏差平方和,求出复相关系数R,即:
Figure PCTCN2017072675-appb-000070
其中,
Figure PCTCN2017072675-appb-000071
Figure PCTCN2017072675-appb-000072
(4)根据复相关系数,求出每个BMS对应VIF,即:
Figure PCTCN2017072675-appb-000073
(5)根据实际数据特点,采用VIF>5或VIF>10两种阈值进行过滤,即当VIF>5或VIF>10时,认为对应biomarker与其他biomarker之间存在强共线性,则剔除该BM。
6.6分别对性别分组后的样本进行年龄段划分,并判断不同年龄分组中的样本量是否大于BMs数目。若是,则对有效的性别分组BMs集使用上述进行VIF过滤;否则使用有效的性别分组BMs集。
6.7针对每个年龄分组,继续进行与年代年龄(Chronological Age,CA)关联性的过滤,具体算法主要是Pearson相关性分析,具体如下:
Figure PCTCN2017072675-appb-000074
其中y表示CA,x表示BMs对应的检测值。
根据最大化保留有效biomarkers的原则,当r<0.1时,过滤对应的BMs。
6.8针对每个年龄分组,使用每个年龄组过滤后的BMs进行生理年龄初步估计值
Figure PCTCN2017072675-appb-000075
计算,核心算法为KD模型,计算步骤如下:
Figure PCTCN2017072675-appb-000076
其中,
Figure PCTCN2017072675-appb-000077
是预测BA的值;C为年代年龄CA;
Figure PCTCN2017072675-appb-000078
为BMs的方差;j为BM;m为BMs的种类;kj为每个BM对CA拟合的斜率;qj为每个BM对CA拟合的截距,xj为样本x的第j个BM的数值。
(1)根据一元线性回归,计算回归系数kj,qj(j=0,1,...,m-1),即:
y=kx+q
Figure PCTCN2017072675-appb-000079
Figure PCTCN2017072675-appb-000080
其中
Figure PCTCN2017072675-appb-000081
(2)根据回归系数,计算偏差平方和q,即:
Figure PCTCN2017072675-appb-000082
(3)根据偏差平方和,计算平均标准偏差,即:
Figure PCTCN2017072675-appb-000083
6.9在6.8计算了生理年龄估计值
Figure PCTCN2017072675-appb-000084
后,为更精确计算不同人群下的生物学年龄估计,使用不同人群的样本年龄分布的数据作为参考总体,进行最大后验概率计算(maximum a posteriori,MAP),以此可以消除过拟合状况;最后,得到预测的生物学年龄BA。其计算方法如下:
Figure PCTCN2017072675-appb-000085
其中,
Figure PCTCN2017072675-appb-000086
即生理年龄初步估计值,μ0为实际年龄C,σ与σ0分别为似然函数与先验函数标准差。BA为最终计算的生物学年龄。具体实施流程请参考图1生物学年龄计算模型流 程图。
(b)衰老评估基线模型实施步骤如下:
6.11使用生物学年龄评估模型中步骤6.7所选的年龄分组的样本数据。判断各年龄组别中,样本数是否大于BMs数;若是,去除欧式距离(Euclidean Distance)最高的5%的样本,使用性别分组下的备选BMs做后续计算;否,则去除马氏距离(Mahalanobis Distance)最高的5%的样本,使用上述VIF算法过滤BMs,使用过滤后的biomarkers进行后续计算。
6.12使用CA对所有BM进行拟合,去除Cook’s Distance>1的样本后,同时去除与CA关联性<0.1的BMs。此时即选出满足确立基线的样本。
6.13使用同步骤6.8算法完成生理年龄
Figure PCTCN2017072675-appb-000087
计算,并计算生理年龄标准差。同时,获取研究样本人群的死亡年龄数据库,计算先验的生理年龄标准差。
6.14使用最大后验概率(MAP)计算生理年龄BA。
6.15使用BA对CA进行拟合,判断有无样本大于Cook’s Distance所设阈值。如有,则去除该样本。
6.16重复步骤6.15,直至没有大于Cook’s Distance阈值的样本,然后计算不同年龄段的生理年龄预测的95%置信区间。所述生物学年龄置信区间即为计算的衰老评估基线范围。具体实施流程请参考图2衰老基线估计模型流程图。
实施例2
下面参照“一般方法”和实施例1,基于294个正常个体血液和尿液样本,进行生物学指标的筛选和待测个体的相对衰老程度的评估,具体地:
本实施例拟使用样本多项生物学指标评估机体生理年龄和衰老速率。具体地,发明人采集294个正常个体血液和尿液样本,检测近60项的生物学指标,并基于前述的本发明的“一般方法”成功筛选衰老生物学指标,使用本发明的衰老评估模型的生物学年龄算法筛选与年龄线性关联的生物学指标,并使用这些指标计算生物学年龄,使用衰老评估基线算法剔除离群样本,建立不同性别,不同年龄段群体的生物学年龄基线,用以评估个体的衰老状态。
其中,本实施例选取的生物学指标涵盖了基因组的不稳定、端粒缺失、线粒体功能丧失、激素代谢、蛋白质内稳态,细胞自身衰老等衰老影响因子。首先通过技术开发检测的生物学指标是甲基化水平(mcDNA-level)、端粒长度(MTL)、线粒体拷贝数(mtCN)、线粒体累计突变(mtHP)、8-羟基脱氧鸟苷(8-OHdG)、羧甲基赖氨酸(CML)、脱氢表雄酮(DHEA)、硫酸脱氢表雄酮(DHEAS)、25羟基VD3(25OHVD3)、氢化可的松(F)、雌酮(E1)、雌二醇(E2)、孕酮(P)、睾酮(T)。同时获取常规体检的39项生化指标,具体为白细胞计数(WBC)、淋巴细胞比率(LYMR)、中间细胞比率(MID)、中性粒细胞比率(GRANR)、淋巴细胞总数(LYM)、中间细胞总数(MTD)、中性粒细胞总数(GRAN)、红细胞总数(RBC)、血红蛋白(HGB)、红细胞压积(HCT)、红细胞平均体积(MCV)、平均红细胞血红蛋白含量(MCH)、平均红细胞血红 蛋白浓度(MCHC)、红细胞分布宽度CV(RDW-CV)、红细胞分布宽度SD(RDW-SD)、血小板总数(PLT)、平均血小板体积(MPV)、血小板分布宽度(PDW)、血小板压积(PCT)、大血小板比率(P_LCR)、天门冬氨酸氨基转移酶(AST)、丙氨酸氨基转移酶(ALT)、γ-谷氨酰转肽酶(GGT)、肌酐(Creatinine)、空腹血葡萄糖(FBG)、甘油三酯(TRIG)、总胆固醇(TCHO)、低密度脂蛋白胆固醇(LDLC)、高密度脂蛋白胆固醇(HDLC)、白蛋白(Albumin)、球蛋白(Globulin)、白蛋白/球蛋白比值(A/G)、间接胆红素(IBIL)、总胆红素(TBIL)、碱性磷酸酶(ALP)、尿酸(UA)、直接胆红素(DBIL)、总蛋白(TP)、尿素(Urea)。
一、长寿基因对生物学年龄贡献度
随机选取10名志愿者血液样品,对提取的DNA进行二代测序,使用BGISEQ500SE100测序,获得基因分型结果见下表。另外,进行长寿位点贡献度OR-GRS模型计算。得到S1-S10每个个体的长寿风险得分值。
Figure PCTCN2017072675-appb-000088
二、线粒体DNA拷贝数和线粒体DNA累积突变
2.1线粒体拷贝数
1.数据导出与数据分析。
本实施例,共测试5个样本,使用Q-PCR程序,对每个样本系统完成后会给出线粒体拷贝数Mi孔和单基因拷贝数N孔的拷贝数值,然后取复孔Mi-1,Mi-2平均值和复孔N-1,N-2平均拷贝数值的商,即为该样本线粒体相对拷贝数。
2.数据整理(上表K列为Quantity Mean值,即两复孔的平均拷贝数):
表2-1
Figure PCTCN2017072675-appb-000089
Figure PCTCN2017072675-appb-000090
3.线粒体相对拷贝数计算
表2-2
Figure PCTCN2017072675-appb-000091
三、端粒长度定量
本实施例中,共检测两个标准品和八个样本,实时荧光定量PCR分别测定端粒(TEL)和内参基因RPLPO(核糖体大亚基PO蛋白基因)的CT值。将检测仪器数据导出,整理计算各样本各引物对应的平均CT值。
首先,计算端粒(T)重复拷贝数与单拷贝基因(S)的比率,即T/S比率可以得出端粒的相对长度,而T/S比率与端粒长度成正比关系。T/S计算公式如下:
T/S=[2CT(telomeres)/2CT(single copy gene)]=2-ΔCT
引入标准品后,T/S=2-ΔΔCT,其数值表示样本端粒长度与标准品的比较短息,小于1表示其端粒长度小于标准品端粒长度,大于1表示其端粒长度大于标准品端粒长度。引入多个标准品可以通过标准品T/S值来评估不同批次实验间的稳定性。完成端粒长度定量检测。
端粒长度定量数据如下表格。
表3-1
Figure PCTCN2017072675-appb-000092
Figure PCTCN2017072675-appb-000093
四、总体甲基化水平
本实施例中,一方面确定方法的稳定性,另一方面批量测试。
(1)使用同一个样本进行多次重复,以确定检测方法的稳定性。使用QC样本进行6次重复,将按照标准质控流程得到的单核苷酸化合物(A,T,C,G,dA,U,dC,mC,mdC,dG,hmdC)浓度,使用mdC和dG浓度数据分析方法稳定性,并根据mdC和dG的浓度计算每个样本的全局甲基化水平。计算6次重复实验间的CV值均小于15%,稳定性好。具体数据如下表。
表4-1
Figure PCTCN2017072675-appb-000094
(2)此外,批量检测了12个唾液样本的总体甲基化水平稳定性表现较好。每个样本有2个重复,CV值均小于15%。具体数据如下表
表4-2
样品 1 2 3 4 5 6 7 8 9 10 11 12
重复1 0.902 0.882 1.109 0.807 0.837 1.233 1.426 1.270 1.276 1.436 1.408 1.281
重复2 0.960 0.882 1.212 0.766 0.852 1.176 1.262 1.316 1.257 1.452 1.242 1.367
CV% 4.444 0.035 6.283 3.695 1.237 3.357 8.614 2.502 1.107 0.820 8.839 4.561
五、激素指标定量
5.1选用MultiQuant(AB SCIEX)积峰软件。质控标准:S/N>3;Accuracy>80&<120;CV<15%;R2>0.90。
5.2将按照质控标准计算出的浓度导出,通过Perl脚本分析数据稳定性,批量测试34份样本各项激素数据,如下表5-1。
表5-1
Figure PCTCN2017072675-appb-000095
Figure PCTCN2017072675-appb-000096
六、衰老评估模型计算
6.1首先,对294个样本检测,得58个符合一级质控标准的各项生物学指标数据;
6.2将294样本按照年龄进行分组,男性和女性分别为154份和140份。
6.3我们首先对两个性别组的样本集中,对各项生物学指标数据进行归一化处理,消除批次间误差。(1)对同一批次下检测数据不做处理。(2)对不同批次获得数据做批次间CV计算,对CV值不满足<15%,我们将对批次间的生物学指标数据使用线性混合模型(Linear  mixed model,LMM)进行修正。
6.4使用满足6.3处理的生物学指标进行数目统计,分析男性组和女性组样本量和BM数目的关系,两组样本量均大于BM数据,使用VIF算法过滤生物学指标,分别得到男性组和女性组的有效备选BM集。
6.6年龄段划分分为两个阶段,第一阶段是在20岁~40岁之间,每五年作为一个年龄段间隔;第二个阶段为大于40岁个体。两个性别组分别分成5个年龄组。如果年龄组中,样本数目小于BM数目,不做过滤处理;对年龄组中,样本数目大于BM数目同样使用VIF算法过滤,得到用于进行生物学年龄计算的有效备选BM集。
6.7使用有效备选BM集对每个年龄组进行生物学年龄评估。使用BACM模型计算了每个个体在所属性别组和年龄组相对生物学年龄BA,计算的BA与CA拟合,计算获得R^2均大于0.6,甚至高达0.98,显著性均P<1e-4,如图3不同性别的生物学年龄BA及不同年龄群的衰老基线分布图中实线表示生物学年龄BA与年代年龄CA的拟合结果。
6.8生物学指标和生物学年龄的关联结果显示,在不同年龄段,女性20-25岁年龄群中,与生物学年龄关联的指标组合有TRIG,Lymph,HDL,PCT,BMI,Globulin,mtCN,UA,RDW-CV,Urea,MTL,Albumin,Alkphosph,FBG,P,T,PLCR,AST,GGT,E1,MCHC,Height,CML,DHEA,RDW-SD,Melatonin,mcDNA-level,25OHVD3,LDL等;图4女性20-25岁年龄群中与生物学年龄关联的指标;在女性25-29岁的年龄群中,与生物学年龄关联的指标组合有LYM,LYMR,RDW-CV,LDL,UA,TRIG,BMI,CML,F,mt-CN,MCHC,MSIZEDRATE,Creatinine,Melatonin,8-OHdG,HDL,PLCR,T,mcDNA-level,TBIL,E1,MCV,Height,P,FBG,Alkphosph,RBC,GGT,DHEAS,AST,PCT等,如图5女性25-30岁年龄群中与生物学年龄关联的指标;在女性30-35岁的年龄群中,与生物学年龄关联的指标组合有MCV,RDW-CV,PCT,RDW-SD,Globulin,msizedrate,UA,DHEA,Urea,MCHC,LYMR,Albumin,DHEAS,TRIG,LYM,LDL,8-OHdG,P,TBIL,Height,mcDNA-level,F,AST,BMI,HDL,mtCN,GGT等,详见如图6女性30-35岁年龄群中与生物学年龄关联的指标;在女性35-40岁的年龄群中,与生物学年龄关联的指标组合有RBC,Height,MCV,P,FBG,E1,Albumin,Urea,DHEA,8-DHdG,MCHC,Creatinine,25OHVD3,DHEAS,MTL,Melatonin,TRIG,LYMR,T,Globulin,Alkphosph,RDW-SD,HDL,LYM,GGT,LDL,F,PCT,PLCR,CML,BMI等,详见图7女性35-40岁年龄群中与生物学年龄关联的指标;在女性大于40岁的年龄群中,与生物学年龄关联的指标有DHEA,P,Urea,MCHC,UA,msizedrate,FBG,Creatinine,TRIG,LDL,MCV,TBIL,HDL,mtCN,LYMR,BMI,Height,RBC,LYM,RDW-SD,PLCR,DHEAS,Albumin,E1,Melatonin,F,Globulin等,详见图8女性40-100岁年龄群中与生物学年龄关联的指标;综上所述,女性不同年龄群与生物学年龄相关的指标种类是不同的,并且相同指标对在不同年龄群中对生物学年龄的贡献度也是不同的。
另外,在男性20-24岁年龄群中,与生物学年龄关联的指标组合有BMI,PDW,Height,PLCR,LDL,Melatonin,MTL,TRIG,DHEAS,mtCN,AST,PCT,CML,RDW-CV, P,Hemoglobin,WBC,UA,FBG,25OHVD3,RDW-SD,GGT,MCHC,Albumin,LYMR等,详见图9男性20-25岁年龄群中与生物学年龄关联的指标;在男性25-29岁的年龄群中,与生物学年龄关联的指标有8-OHdG,Albumin,GGT,Globulin,PCT,AST,LDL,TRIG,BMI,E2,Hemoglobin,HDL,RDW-SD,Melatonin,UA,DHEAS,MTL,IBIL,T,CML,Alkphosph,Creatinine,LYMR,25OHVD3,DHEA等,详见图10男性25-30岁年龄群中与生物学年龄关联的指标;在男性30-34岁的年龄群中,与生物学年龄关联的指标有RDW-SD,WBC,Hemoglobin,Height,Creatinine,RDW-CV,CML,FBG,LYMR,Urea,Albumin,Alkphosph,F,Globulin,UA,BMI,8-OHdG,LDL,P,mtCN,mcDNA-level,MTL,TRIG,GGT等,详见图11男性30-35岁年龄群中与生物学年龄关联的指标;在男性35-39岁的年龄群中,与生物学年龄的指标组合有GGT,PCT,IBIL,FBG,AST,UA,TRIG,BMI,HDL,mtCN,Albumin,LDL,LYMR,MCHC,E1,F,Globulin,Alkphosph,Height,Melatonin,Hemoglobin,DHEA,PDW,Creatinine,PLCR,T,MTL,P,CML,DHEA,mcDNA-level等,详见图12男性35-40岁年龄群中与生物学年龄关联的指标;
在男性大于40岁的年龄群中,与生物学年龄关联的指标有Albumin,T,IBIL,Height,MCHC,GGT,LDL,UA,MTL,BMI,Alkphosph,F,PCT,mtCN,Melatonin,Hemoglobin,PWD,PLCR,AST,Creatinine,Globulin,DHEA,8-OHdG,E1,FBG等,详见图13男性40-100岁年龄群中与生物学年龄关联的指标。综上所述,男性不同年龄群与生物学年龄相关的指标种类是不同的,并且相同指标对在不同年龄群中对生物学年龄的贡献度也是不同的。
综合男性和女性两组与生物学年龄关联的指标比较分析发现,男性和女性在不同年龄段的与生物学年龄关联的指标种类存在较大差异,不同指标在不同年龄段内对衰老的贡献度也是不同。这些与生物学年龄关联的指标即为筛选出的衰老指标,上述结果也证明了我们的衰老评估模型的策略是可行的,有效说明生物学年龄关联的生物学指标会随性别和年龄段改变而不同。
6.9在进行衰老评估基线计算之前,首先,使用生物学年龄计算模型中步骤6.7所选的年龄分组的样本数据。判断各年龄组别中,当样本数大于BM数,去除欧式距离(Euclidean Distance)最高的5%的样本,此时得到性别分组下有效备选BM集;当样本数小于BM数,则去除马氏距离(Mahalanobis Distance)最高的5%的样本,使用上述VIF算法过滤BM。
6.10使用CA对所有BM进行拟合,去除Cook’s Distance>1的样本后,同时去除与CA关联性<0.1的BMs。此时即选出满足确立基线的样本。
6.11使用生物学年龄计算模型计算生物学年龄
Figure PCTCN2017072675-appb-000097
并计算样本生物学年龄标准差。同时,获取中国统计年鉴人群的死亡年龄分布数据,计算先验的生物学年龄标准差。
6.12使用最大后验概率(MAP)计算生理年龄BA。
6.13使用BA对CA进行拟合,判断有无样本大于Cook’s Distance所设阈值。如有,则去除该样本。
6.14重复步骤6.13,直至没有大于Cook’s Distance阈值的样本,然后计算不同年龄段的生理年龄预测的95%置信区间。所述生物学年龄置信区间即为计算的衰老评估基线范围。当个体发生异常衰老时,则会出现在所述衰老基线范围外侧,图3中偏离基线外的离群点。当个体高于置信区间上限时,生物学年龄趋于衰老状态,当个体低于置信区间下限时,生理年龄趋于年轻状态。
七、个体生物学年龄计算与衰老状态评估
以大样本的第167号男性为例,按照前述步骤测定其年代年龄38.13岁,获得各项生物学指标的含量或分值,如下表所示。使用生物学年龄计算模型计算的KD算法计算的BA为38.01岁,处于95%置信区间内;使用MAP算法计算最终BA为37.51岁,处于95%置信区间内。使用衰老评估基线模型建立的男性35-40岁年龄群衰老基线范围观测,该样本未偏离总体范围,属于较好的生理学健康状态(见图3中d图所标记的样本,以及图14)。这意味着样本机体生物学年龄属正常范围,机体状态良好,在人群衰老基线中处于正常范围衰老水平。
第167号样本生物学指标的含量或得分
Figure PCTCN2017072675-appb-000098
Figure PCTCN2017072675-appb-000099
工业实用性
本发明的确定人群样本的生物学年龄预测生物指标集的方法,能够有效地用于确定人群样本的生物学年龄预测生物指标集,进而利用这些生物指标集可以准确地确定个体生物学年龄以及人群样本的衰老评估基线,并进一步,基于获得的个体生物学年龄以及人群样本的衰老评估基线能够有效地评估该个体的相对衰老程度,并且评估结果准确性好,可信度高。
尽管本发明的具体实施方式已经得到详细的描述,本领域技术人员将会理解。根据已经公开的所有教导,可以对那些细节进行各种修改和替换,这些改变均在本发明的保护范围之内。本发明的全部范围由所附权利要求及其任何等同物给出。
在本说明书的描述中,参考术语“一个实施例”、“一些实施例”、“示意性实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本发明的至少一个实施例或示例中。在本说明书中,对上述术语的示意性表述不一定指的是相同的实施例或示例。而且,描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。

Claims (17)

  1. 一种确定人群样本的生物学年龄预测生物指标集的方法,其特征在于,包括以下步骤:
    获得所述人群样本中所有个体的备选生物指标的数据;
    将所述人群样本中所有个体的备选生物指标数据按照性别进行分组,以便获得男性备选生物指标集和女性备选生物指标集;
    分别对所述男性备选生物指标集和所述女性备选生物指标集进行方差膨胀因子算法过滤,以便分别获得男性有效备选生物指标集和女性有效备选生物指标集;
    针对所述男性有效备选生物指标集和所述女性有效备选生物指标集的每一个,分别按照年龄段进行分组,以便获得多个不同年龄段的男性有效备选生物指标集和多个不同年龄段的女性有效备选生物指标集;
    分别确定所述多个不同年龄段的男性有效备选生物指标集和所述多个不同年龄段的女性有效备选生物指标集中每一个的有效生物指标集,以便获得多个不同年龄段的男性有效生物指标集和多个不同年龄段的女性有效生物指标集,其中,针对所述多个不同年龄段的男性有效备选生物指标集和所述多个不同年龄段的女性有效备选生物指标集的每一个,当其样本量大于有效备选生物指标的数目时,对其有效备选生物指标进行方差膨胀因子算法过滤,以便确定有效生物指标集;当其样本量小于有效备选生物指标的数目时,直接以其有效备选生物指标的集合作为有效生物指标集;以及
    将所述多个不同年龄段的男性有效生物指标集和所述多个不同年龄段的女性有效生物指标集,分别进行年代年龄关联性过滤,以便获得多个不同年龄段的男性生物学年龄预测生物指标集和多个不同年龄段的女性生物学年龄预测生物指标集,各个以性别年龄分类的生物学年龄预测生物指标集即为所述人群样本的生物学年龄预测生物指标集。
  2. 根据权利要求1所述的方法,其特征在于,所述备选生物指标为选自长寿基因、线粒体DNA拷贝数、端粒体长度、总体甲基化水平、激素水平的至少之一。
  3. 根据权利要求1所述的方法,其特征在于,所述人群样本中所有个体的所述生物指标的数据符合标准质控。
  4. 根据权利要求1所述的方法,其特征在于,所述男性备选生物指标集和所述女性备选生物指标集中的所有生物指标数据均是同批次检测,或满足批次间检测CV值的要求,或不满足批次间检测CV值要求的已经使用LMM算法修正,并且,各组的样本量均应大于其备选生物指标的数目。
  5. 根据权利要求1所述的方法,其特征在于,通过以下步骤,进行所述方差膨胀因子算法过滤:
    (1)根据多重线性回归,基于下列公式,得到回归系数a0,a1,a2,...,am-1,am
    Figure PCTCN2017072675-appb-100001
    (2)根据所述回归系数a0,a1,a2,...,am-1,am,基于下列公式,求出偏差平方和q:
    Figure PCTCN2017072675-appb-100002
    (3)根据所述偏差平方和q,基于下列公式,求出复相关系数R:
    Figure PCTCN2017072675-appb-100003
    其中,
    Figure PCTCN2017072675-appb-100004
    Figure PCTCN2017072675-appb-100005
    (4)根据所述复相关系数R,基于下列公式,求出每个备选生物指标或有效备选生物指标对应的VIF:
    Figure PCTCN2017072675-appb-100006
    以及
    (5)采用VIF>5或VIF>10两种阈值对各备选生物指标或有效备选生物指标进行过滤,其中,当VIF>5或VIF>10时,认为对应的生物学年龄相关指标与其他生物学年龄相关指标之间存在强共线性,则剔除该备选生物指标或有效备选生物指标。
  6. 根据权利要求1所述的方法,其特征在于,所述按照年龄段进行分组,是按照预定年龄段间隔进行的。
  7. 根据权利要求1所述的方法,其特征在于,利用Pearson相关性分析法,基于以下公式,进行所述年代年龄关联性过滤:
    Figure PCTCN2017072675-appb-100007
    其中y表示年代年龄CA,x表示有效生物指标对应的检测值,
    根据最大化保留有效生物指标的原则,当r<0.1时,过滤对应的有效生物指标。
  8. 一种确定待测个体生物学年龄的方法,其特征在于,包括以下步骤:
    根据权利要求1-7任一项所述的方法,确定待测个体所属人群样本的以性别年龄分类的各生物学年龄预测生物指标集;
    基于待测个体对应的以性别年龄分类的生物学年龄预测生物指标集,计算所述待测个体的生物学年龄初步估计值BASC;以及
    以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值BASC进行最大后验概率计算处理,以便确定所述待测个体的预测生物学年龄BA。
  9. 根据权利要求8所述的方法,其特征在于,以KD模型法,基于以下公式,计算所述待测个体的生物学年龄初步估计值BASC
    Figure PCTCN2017072675-appb-100008
    其中,BASC是预测BA的值,C为年代年龄CA,
    Figure PCTCN2017072675-appb-100009
    为生物学年龄预测生物指标集BMs的方差,j为生物学年龄预测生物指标BM,m为生物学年龄预测生物指标集BMs的种类,kj为每个生物学年龄预测生物指标BM对CA拟合的斜率,qj为每个生物学年龄预测生物指标BM对年代年龄CA拟合的截距,xj为样本x的第j个生物学年龄预测生物指标BM的数值。
  10. 根据权利要求8所述的方法,其特征在于,按照以下步骤确定kj、qj和s:
    (1)根据一元线性回归,基于以下公式,计算回归系数kj,qj(j=0,1,...,m-1):
    y=kx+q,
    Figure PCTCN2017072675-appb-100010
    Figure PCTCN2017072675-appb-100011
    其中
    Figure PCTCN2017072675-appb-100012
    (2)根据所述回归系数kj,qj(j=0,1,...,m-1),基于以下公式,计算偏差平方和q’:
    Figure PCTCN2017072675-appb-100013
    (3)根据偏差平方和q’,基于以下公式,计算平均标准偏差s:
    Figure PCTCN2017072675-appb-100014
  11. 根据权利要求9所述的方法,其特征在于,基于以下公式进行最大后验概率计算,确定预测的生物学年龄BA:
    Figure PCTCN2017072675-appb-100015
    其中,BASC为生理年龄初步估计值,μ0为实际年龄C,σ与σ0分别为似然函数与先验函数标准差。
  12. 一种确定人群样本的衰老评估基线的方法,其特征在于,包括以下步骤:
    根据权利要求1-7任一项所述的方法,确定人群样本的以性别年龄分类的各生物学年龄预测生物指标集;
    对以性别年龄分类的各生物学年龄预测生物指标集分别进行再过滤处理,其中,针对以性别年龄分类的各生物学年龄预测生物指标集,当其样本量大于生物学年龄预测生物指标的数目时,去除欧式距离Euclidean Distance最高的5%的样本;当其样本量小于生物学年龄预测生物指标的数目时,去除马氏距离Mahalanobis Distance最高的5%的样本,并使用方差膨胀因子算法过滤生物学年龄预测生物指标;
    使用年代年龄CA对所有生物学年龄预测生物指标进行线性拟合,去除Cook’s Distance>1的样本,同时去除与CA关联性<0.1的生物学年龄预测生物指标,以便筛选出满足确立基线初始标准的样本和生物学年龄预测生物指标集;
    基于所述满足确立基线初始标准的样本和生物学年龄预测生物指标集,计算所述人群样本的每一个个体的生物学年龄初步估计值BASC
    以不同人群的样本年龄分布数据为参照,对所述生物学年龄初步估计值BASC进行最大后验概率计算处理,以便确定每一个个体的预测生物学年龄BA;
    利用每一个个体的预测生物学年龄BA对年代年龄CA进行线性拟合,去除Cook’s Distance>1的样本,并重复此步骤至没有样本Cook’s Distance>1,以便筛选出满足确立基线要求的样本和生物学年龄预测生物指标集;以及
    基于所述满足确立基线要求的样本和生物学年龄预测生物指标集,计算各性别年龄组的生物学年龄预测的95%置信区间,所述生物学年龄预测的95%置信区间即为各性别年龄组的衰老评估基线。
  13. 根据权利要求12所述的方法,其特征在于,以KD模型法,基于以下公式,计算每一个个体的生物学年龄初步估计值BASC
    Figure PCTCN2017072675-appb-100016
    其中,BASC是预测BA的值,C为年代年龄CA,
    Figure PCTCN2017072675-appb-100017
    为生物学年龄预测生物指标集BMs 的方差,j为生物学年龄预测生物指标BM,m为生物学年龄预测生物指标集BMs的种类,kj为每个生物学年龄预测生物指标BM对CA拟合的斜率,qj为每个生物学年龄预测生物指标BM对年代年龄CA拟合的截距,xj为样本x的第j个生物学年龄预测生物指标BM的数值。
  14. 根据权利要求13所述的方法,其特征在于,按照以下步骤确定kj、qj和s:
    (1)根据一元线性回归,基于以下公式,计算回归系数kj,qj(j=0,1,...,m-1):
    y=kx+q,
    Figure PCTCN2017072675-appb-100018
    其中
    Figure PCTCN2017072675-appb-100019
    (2)根据所述回归系数kj,qj(j=0,1,...,m-1),基于以下公式,计算偏差平方和q’:
    Figure PCTCN2017072675-appb-100020
    (3)根据偏差平方和q’,基于以下公式,计算平均标准偏差s:
    Figure PCTCN2017072675-appb-100021
  15. 根据权利要求12所述的方法,其特征在于,基于以下公式进行最大后验概率计算,确定预测的生物学年龄BA:
    Figure PCTCN2017072675-appb-100022
    其中,BASC为生理年龄初步估计值,μ0为实际年龄C,σ与σ0分别为似然函数标准差与先验函数标准差。
  16. 根据权利要求12所述的方法,其特征在于,基于线性拟合结果对所述人群样本进行筛选包括:去除Cook’s Distance>1的样本。
  17. 一种确定待测个体相对衰老程度的方法,其特征在于,包括以下步骤:
    根据权利要求8-11任一项所述的方法,确定所述待测个体的生物学年龄BA;
    根据权利要求12-16任一项所述的方法,确定所述待测个体所属人群样本的各性别年龄 组的的衰老评估基线;以及
    将所述待测个体的生物学年龄BA与所述待测个体所处性别年龄组的衰老评估基线进行比较,以便确定所述待测个体的相对衰老程度,
    其中,
    当所述待测个体的生物学年龄BA在所述待测个体所处性别年龄组的衰老评估基线的范围内时,判断所述待测个体相对所属人群样本处于正常衰老水平;
    当所述待测个体的生物学年龄BA偏离所述待测个体所处性别年龄组的衰老评估基线的范围时,判断所述待测个体相对所属人群样本处于异常衰老水平,其中,当所述待测个体的生物学年龄BA高于置信区间上限时,所述待测个体相对所属人群样本趋于衰老状态,当所述待测个体的生物学年龄BA低于置信区间下限时,所述待测个体相对所属人群样本趋于年轻状态。
PCT/CN2017/072675 2017-01-25 2017-01-25 确定人群样本生物指标集、预测生物学年龄的方法及其应用 WO2018137203A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201780084324.XA CN110392740A (zh) 2017-01-25 2017-01-25 确定人群样本生物指标集、预测生物学年龄的方法及其应用
PCT/CN2017/072675 WO2018137203A1 (zh) 2017-01-25 2017-01-25 确定人群样本生物指标集、预测生物学年龄的方法及其应用

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/072675 WO2018137203A1 (zh) 2017-01-25 2017-01-25 确定人群样本生物指标集、预测生物学年龄的方法及其应用

Publications (1)

Publication Number Publication Date
WO2018137203A1 true WO2018137203A1 (zh) 2018-08-02

Family

ID=62977920

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/072675 WO2018137203A1 (zh) 2017-01-25 2017-01-25 确定人群样本生物指标集、预测生物学年龄的方法及其应用

Country Status (2)

Country Link
CN (1) CN110392740A (zh)
WO (1) WO2018137203A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110379515A (zh) * 2019-07-12 2019-10-25 上海联影智能医疗科技有限公司 用于评估待检测目标的生理参数指标的方法、计算机设备
CN113257344A (zh) * 2020-02-12 2021-08-13 大江基因医学股份有限公司 细胞状态评估模型的建立方法

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110974177B (zh) * 2019-12-23 2022-09-09 南昌五元生物科技有限公司 一种基于端粒长度的生物学年龄计算方法及系统
CN111340091B (zh) * 2020-02-21 2022-08-23 上海艾瑞德生物科技有限公司 用于分类免疫数据的cnn模型的训练方法及其应用
CN111816307A (zh) * 2020-04-15 2020-10-23 浙江大学 基于临床标志物构建中国人群生物学年龄评价模型的方法及评价方法
CN111477273A (zh) * 2020-05-18 2020-07-31 中国人民解放军国防科技大学 一种基于脑组织基因表达预测个体年龄信息的方法
CN113707316B (zh) * 2020-11-12 2024-03-26 深圳泛因医学有限公司 免疫状态评估方法及应用
CN113254882B (zh) * 2021-06-07 2023-02-10 广州市百果园网络科技有限公司 实验结果的确定方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005048185A1 (en) * 2003-11-17 2005-05-26 Auckland University Of Technology Transductive neuro fuzzy inference method for personalised modelling
CN104357561A (zh) * 2014-10-31 2015-02-18 公安部物证鉴定中心 一种获得中国人群女性个体年龄的方法和系统
CN104694641A (zh) * 2015-02-13 2015-06-10 冯文峰 一种预测基因年龄及疾病易感性的方法和试剂盒
CN104966106A (zh) * 2015-07-13 2015-10-07 哈尔滨工业大学 一种基于支持向量机的生物年龄分步预测方法
CN105765083A (zh) * 2013-09-27 2016-07-13 加利福尼亚大学董事会 基于表观遗传学标记物来估计组织和细胞类型的年龄的方法

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110207128A1 (en) * 2010-02-16 2011-08-25 Cawthon Richard M Methods and kits for determining biological age and longevity based on gene expression profiles
US10087486B2 (en) * 2012-11-09 2018-10-02 The Regents Of The University Of California Methods for predicting age and identifying agents that induce or inhibit premature aging
EP2781602A1 (en) * 2013-03-21 2014-09-24 Universität Konstanz Method for the determination of biological age in human beings
KR101603308B1 (ko) * 2013-11-20 2016-03-14 주식회사 바이오에이지 생체 나이 연산 모델 생성 방법 및 시스템과, 그 생체 나이 연산 방법 및 시스템
KR101669526B1 (ko) * 2015-03-04 2016-10-26 주식회사 바이오에이지 생체나이를 이용한 잔여 수명 예측방법
KR20170000424A (ko) * 2015-06-23 2017-01-03 배철영 텔로미어를 이용한 생체나이 측정 시스템, 텔로미어 생체나이 추정 방법 및 생체나이측정 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005048185A1 (en) * 2003-11-17 2005-05-26 Auckland University Of Technology Transductive neuro fuzzy inference method for personalised modelling
CN105765083A (zh) * 2013-09-27 2016-07-13 加利福尼亚大学董事会 基于表观遗传学标记物来估计组织和细胞类型的年龄的方法
CN104357561A (zh) * 2014-10-31 2015-02-18 公安部物证鉴定中心 一种获得中国人群女性个体年龄的方法和系统
CN104694641A (zh) * 2015-02-13 2015-06-10 冯文峰 一种预测基因年龄及疾病易感性的方法和试剂盒
CN104966106A (zh) * 2015-07-13 2015-10-07 哈尔滨工业大学 一种基于支持向量机的生物年龄分步预测方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110379515A (zh) * 2019-07-12 2019-10-25 上海联影智能医疗科技有限公司 用于评估待检测目标的生理参数指标的方法、计算机设备
CN113257344A (zh) * 2020-02-12 2021-08-13 大江基因医学股份有限公司 细胞状态评估模型的建立方法

Also Published As

Publication number Publication date
CN110392740A (zh) 2019-10-29

Similar Documents

Publication Publication Date Title
WO2018137203A1 (zh) 确定人群样本生物指标集、预测生物学年龄的方法及其应用
Enquobahrie et al. Placental microRNA expression in pregnancies complicated by preeclampsia
JP6161607B2 (ja) サンプルにおける異なる異数性の有無を決定する方法
US10745754B2 (en) Method for predicting congenital heart defect
JP6525894B2 (ja) 超並列rna配列決定による母体血漿のトランスクリプトーム解析
JP6987393B2 (ja) 血液中の無細胞dnaを用いた血液学的障害の検出
WO2018001295A1 (zh) 分子标志物、内参基因及其应用、检测试剂盒以及检测模型的构建方法
CN106834502A (zh) 一种基于基因捕获和二代测序技术的脊髓性肌萎缩症相关基因拷贝数检测试剂盒及方法
CN104812913B (zh) 慢性阻塞性肺疾病(copd)生物标记及其用途
US20080108071A1 (en) Methods and Systems to Determine Fetal Sex and Detect Fetal Abnormalities
CN107541561B (zh) 提高母体外周血中胎儿游离dna浓度的试剂盒、装置及方法
Ng et al. Biomarkers in neonatology: the next generation of tests
WO2016049878A1 (zh) 一种基于snp分型的亲子鉴定方法及应用
CN106755464A (zh) 用于筛选肠癌和/或胃癌的基因标志物的方法、用该方法筛选的基因标志物及其用途
Liu et al. A forward look at noninvasive prenatal testing
CN105821143A (zh) 一种用于检测高血压易感性相关的snp位点的引物及检测方法
WO2016059601A1 (en) Non-invasive methods for detection of genetic abnormalities in an unborn fetus, and primers, probes and kits for uses thereof
GB2559437A (en) Prenatal screening and diagnostic system and method
CN116240273B (zh) 一种基于低深度全基因组测序的判断母源污染比例的方法及其应用
CN112410414B (zh) 孕妇血清外泌体miRNA标志物在制备胎儿先心病早期诊断产品中的应用及试剂盒
JP2020533663A (ja) Dna修復プロファイリング及びそのための方法
CN115011695A (zh) 基于游离环状dna基因的多癌种识别标志物、试剂盒及应用
CN116814781A (zh) 用于检测尿路上皮癌的标志物、试剂盒和装置
US11869630B2 (en) Screening system and method for determining a presence and an assessment score of cell-free DNA fragments
CN116004849A (zh) 一种用于母猪早期妊娠诊断的外泌体miRNA标志物及其应用

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17894044

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17894044

Country of ref document: EP

Kind code of ref document: A1