Human biological age prediction method and system, computer equipment and storage medium

Abstract

The invention discloses a human biological age prediction method and system, computer equipment and a storage medium, a gradient boosting decision tree model is used for prediction of human biological age, specific CpG related to the human biological age is found in all CpG, and the human biological age is predicted according to the specific CpG. The method has high analysis efficiency and accuracy while significantly reducing the number of CpGs in the identification process, provides a brand new, efficient, accurate and low-cost human biological age prediction method and system, a computer device and a storage medium for the field, and has important application value in the field of human age prediction.

Description

Technical Field

[0001] The present invention belongs to the field of data prediction technology, and specifically, relates to a method for predicting human biological age, and more specifically, relates to a method, system, computer device and storage medium for predicting human biological age. Background Art

[0002] Epigenetic modifications and gene mutations are common in the human body. Common epigenetic modifications include DNA methylation, RNA modification, histone acetylation, and non-coding RNA regulation. In recent years, there have been many research results related to aging biomarkers, but how to determine human biological age is still a challenge. In the field of DNA methylation, the methylation clock first proposed by Professor Steve Horvath selects CpG to establish a simple regression model and uses it to predict human biological age, making the relationship between methylation and age closer.

[0003] At present, the initial model has been continuously improved, and many models such as Lasso, Sparse Group Lasso, RidgeRegression, etc. have been applied to the methylation clock. However, after obtaining the methylation level data to establish the model, these methods have high sequencing costs, misjudgment rates, and calculation time. Therefore, this field urgently needs to develop an efficient, accurate, and low-cost method for predicting human biological age. Summary of the invention

[0004] In order to solve the above technical problems existing in the prior art, the purpose of the present invention is to provide a method, system, computer device and storage medium for predicting human biological age.

[0005] The full name of the gradient boosted decision tree model is Gradient Boosted Decision Trees. It integrates multiple weak decision tree learners, and the final output result is their average. Through the gradient boosting mechanism, the weak learners work sequentially, that is, the next model tries to improve the errors of the previous model, and the weight of the wrong site is increased, and the weight of the correct site is reduced to optimize the model. This is completely different from the concept of parallel learning in the random forest model.

[0006] This application uses a gradient boosting decision tree model to consider all CpGs as a whole, reasonably select those CpGs related to human biological age prediction, and has good interpretability, while significantly reducing the number of CpGs required in the prediction process. This not only reduces costs, but also improves the efficiency of analysis and has a higher accuracy rate.

[0007] The above-mentioned object of the present invention is achieved through the following technical solutions:

[0008] A first aspect of the present invention provides a method for predicting human biological age.

[0009] Further, the method comprises:

[0010] Acquiring data, acquiring methylation level data of the methylation sites of the human sample to be tested;

[0011] Analyzing data, inputting the methylation level data of the methylation site into a constructed prediction model, wherein the prediction model predicts the biological age of the human subject to be tested based on the methylation level data of the methylation site;

[0012] Outputting a prediction result, predicting and outputting a prediction result of the biological age of the human being to be tested based on the output result of the prediction model;

[0013] The methylation site is CpG, and the CpG includes: cg08160331, cg13100137, cg11529819, cg04875128, cg07927379, cg21200841, cg03182374, cg02780919, cg06654079, cg16541931, cg23100720, cg27314761, cg16600634, cg23220814, cg10633463, cg17461003, cg13954457, cg03705912, cg20388732, cg10652277, cg0649399 4. cg10501210, cg16066354, cg10254000, cg14423778, cg03660500, cg167 17122, cg13902645, cg27503921, cg18786171, cg03157588, cg01287391, cg 21186299, cg01543654, cg02546818, cg19281363, cg03521358, cg0250111 6. cg19280121, cg14172603, cg23998119, cg08767286, cg24456846, cg2504 6584, cg11071401, cg20923605, cg21646366, cg16620537, cg17207590, cg 00863397, cg01639712, cg08918057, cg00481951, cg11268834, cg08949408 , cg18576044, cg00751072, cg08976646, cg14122300, cg17140797, cg0657 5035, cg06205172, cg24401044, cg14032732, cg06916591, cg18686747, cg0 5929864, cg14569771, cg09748749, cg22491171, cg23166357, cg21951594 , cg25478614, cg20835708, cg23454205, cg07063171, cg24891660, cg24221 648, cg07505391, cg16145113, cg03404211, cg19214184, cg04892758, cg2 5307778, cg02388276, cg22361181, cg10151953, cg12918357, cg26894354,cg16987524、cg27429080、cg21671476、cg01092293、cg01337508、cg07326413、cg27288268、cg00793935、cg03231647、cg25804860、cg24150153、cg19765038、cg07952813、cg00320138、cg02376524、cg25991761、cg24639902、cg00628697、cg06594186、cg03348196、cg22489583、cg19395089、cg00944163、cg02605232、cg07277624、cg26259381、cg24796644、cg06793562、cg22799132、cg13088417、cg07038098、cg14016806、cg14155388、cg02876258、cg07810366、cg13001142、cg04871807、cg06737250、cg06962177、cg21985251、cg05090150、cg02401978、cg01620309、cg10242901、cg17324339、cg10738648、cg21480605、cg02033788、cg24925741、cg12842219、cg07183637、cg13547577、cg25822326、cg22370920、cg22877428、cg01015395、cg00552235、cg21986027、cg14097619、cg03643998、cg16373229、cg25521439、cg23995914、cg07423651、cg02151408、cg24674302、cg12296007、cg05967403、cg08904363、cg01508045、cg00712289、cg04976479、cg00058515、cg04955333、cg24650267、cg06227171、cg19381811、cg17558214、cg03832839、cg10732215、cg11379315、cg20238412、cg09185773、cg08849574、cg11582100、cg03082830、cg24799830、cg04380669、cg15031299、cg09416096、cg25576248、cg20442599、cg09624466、cg04865110、cg20555778、cg09212058、cg21490980、cg06051250、cg07762234、cg11421509、cg19907331、cg10337027、cg02318784、cg03642066、cg25945412、cg03018489、cg14819399、cg22763680、cg01588224、cg06024113、cg04155348、cg11898347、cg15660573、cg06655623、cg16540704、cg16001713、cg22001782、cg06654537、cg04981556、cg12099357、cg02254261、cg02153893、cg18166915、cg23349790、cg25735922、cg13231700、cg07553761、cg06084312、cg21077300、cg26355573、cg21142272、cg01382414、cg08575537、cg03018478、cg04158367、cg18500368、cg13299548、cg21525022、cg01612220、cg05374654、cg19477084、cg24107848、cg01919377、cg16605942、cg10942775、cg06132502、cg03594515、cg20003983、cg01378090、cg23754392、cg19609502、cg23934731、cg04084157、cg02330214、cg01601746、cg22994272、cg15065631、cg01427815、cg08462478、cg16805188、cg19723775、cg18372207、cg03196745、cg25337514、cg05972216、cg26032101、cg18416022、cg01716527、cg27059389、cg16337574、cg03605463、cg11155865、cg26070134、cg25453664、cg13806070、cg21707131、cg09626984、cg04374321、cg24917945、cg19019804、cg03005261、cg24166694、cg08667024、cg02976054、cg11938672、cg17960615、cg16494530、cg05859099、cg16170380、cg09610735、cg26880297、cg08384155、cg13612317、cg20490724、cg10943458、cg04342223、cg20763254、cg10947146、cg20461538、cg13488284、cg27250180、cg03349582、cg24409876、cg23656386、cg20384941、cg21064080、cg02605178、cg23815043、cg09624953、cg11438134、cg14551562、cg00270789、cg15594750、cg01921437、cg18582581、cg14415160、cg10430205、cg15746415、cg26782825、cg16110704、cg00829753、cg14941172、cg18050295、cg08096786、cg20273670、cg07047131、cg24524735、cg16823105、cg19627034、cg11725689、cg04007931、cg12476754、cg17009433、cg08076018、cg06285925、cg01580681、cg08998953、cg16015712、cg10448759、cg02524035、cg13390332、cg23715407、cg08267245、cg03601797、cg19684151、cg10225865、cg07955752、cg11806594、cg09662747、cg21800400、cg15340644、cg25021259、cg10983013、cg08371947、cg24855943、cg23895439、cg02200491、cg08263226、cg10738003、cg16831440、cg13370746、cg26922511、cg08431893、cg04961189、cg22094845、cg22942169、cg02577651、cg18363918、cg00031722、cg17598358、cg08095377、cg19624775、cg17968943、cg05183226, cg25961432, cg05648225, cg05906092, cg27284034, cg16842187, cg19547459, cg0093353 8. cg04427498, cg05168977, cg24991933, cg01573825, cg08231697, cg14985989, cg24962672, cg000426 57. cg13986215, cg01965874, cg01136191, cg03598412, cg00001583, cg26968804, cg08979194, cg0871 9712, cg06759255, cg12974756, cg24136563, cg23214840, cg15732107, cg11795308, cg15704155, cg252 11348, cg22478179, cg20289913, cg08161546, cg01807241, cg15185717, cg18107144, cg01822570, cg1 7580331, cg20679403, cg00350060, cg02058358, cg20455931, cg06470855, cg25430507, cg11858450, cg 02256410, cg04244183, cg02008727, cg02754763, cg06868622, cg16671365, cg13415371, cg19917720, c g11561665, cg09475262, cg05314124, cg24166520, cg14517502, cg09740598, cg15105660, cg23082620. ,

[0014] Furthermore, the prediction model is a gradient boosting decision tree model, and the method for constructing the prediction model includes:

[0015] Obtain methylation level data and biological age information for human population samples;

[0016] The human population samples are randomly divided into a training set and a test set, and based on the methylation level and biological age information corresponding to the training set, the mlr3verse::tnr function is used to fit N different gradient boosting decision tree models based on N groups of different hyperparameter values ​​learning_rate, num_leaves, and max_depth, and the methylation sites described in the first aspect of the present invention are obtained, and the gradient boosting decision tree model that meets the predetermined requirements is selected as the prediction model, and the test set is used for verification to evaluate the effectiveness of the prediction model, so as to obtain a constructed prediction model;

[0017] Wherein, N is a natural number between 50 and 100;

[0018] Optionally, the sample includes a blood sample, a tissue sample, a saliva sample or a urine sample.

[0019] Furthermore, the gradient boosting decision tree model is constructed based on the mlr3verse software package;

[0020] Optionally, the selecting the gradient boosting decision tree model that meets the predetermined requirements includes: selecting a gradient boosting decision tree model that meets any of the following requirements from among the N different gradient boosting decision tree models constructed as the gradient boosting decision tree model that meets the predetermined requirements:

[0021] i) Gradient boosting decision tree model with the lowest misjudgment rate;

[0022] ii) The RMSE of explanation is between 7.00-7.10 and the gradient boosting decision tree model with the least number of independent variables.

[0023] Furthermore, the parameter values ​​of the gradient boosting decision tree model, learning_rate, are between 0.15 and 0.20, num_leaves, are between 50 and 60, and max_depth, are between 0 and 5.

[0024] Furthermore, the methylation level data and biological age information of the human population sample are obtained by the following method:

[0025] Performing methylation sequencing on the human population samples, obtaining methylation level data of the human population to be tested after preprocessing, and obtaining biological age information of the human population by accessing the GEO database;

[0026] Optionally, the methylation site includes CpG;

[0027] Optionally, the preprocessing includes the following process: converting the red and green value signal data of the methylation sequencing CpG into methylation level data, the CpG sequencing red and green value signal data includes methylated CpG and unmethylated CpG, and the ratio of the number of methylated CpGs to the sum of the total number of CpGs is the Beta value, which is the quantitative value of the CpG methylation level.

[0028] A second aspect of the present invention provides a human biological age prediction system.

[0029] Further, the system comprises:

[0030] A data acquisition unit is used to acquire methylation level data of the methylation site of the human sample to be tested;

[0031] A data analysis unit, used for inputting the methylation level data of the methylation site into a constructed prediction model, wherein the prediction model predicts the biological age of the human being to be tested based on the methylation level data of the methylation site;

[0032] An output result unit, used for predicting and outputting a biological age prediction result of the human being to be tested based on an output result of the prediction model;

[0033] The methylation site is the methylation site described in the first aspect of the present invention;

[0034] The prediction result is obtained based on the prediction model, and the method for constructing the prediction model includes:

[0035] Obtain methylation level data and biological age information for human population samples;

[0036] The human population samples are randomly divided into a training set and a test set, and based on the methylation level and biological age information corresponding to the training set, the mlr3verse::tnr function is used to fit N different gradient boosting decision tree models based on N groups of different hyperparameter values ​​learning_rate, num_leaves, and max_depth, and the methylation sites described in the first aspect of the present invention are obtained, and the gradient boosting decision tree model that meets the predetermined requirements is selected as the prediction model, and the test set is used for verification to evaluate the effectiveness of the prediction model, so as to obtain a constructed prediction model;

[0037] Wherein, N is a natural number between 50 and 100.

[0038] Further, the sample includes a blood sample, a tissue sample, a saliva sample or a urine sample;

[0039] Optionally, the gradient boosting decision tree model is constructed according to the mlr3verse software package;

[0040] Optionally, the selecting the gradient boosting decision tree model that meets the predetermined requirements includes: selecting a gradient boosting decision tree model that meets any of the following requirements from among the N different gradient boosting decision tree models constructed as the gradient boosting decision tree model that meets the predetermined requirements:

[0041] i) Gradient boosting decision tree model with the lowest misjudgment rate;

[0042] ii) The RMSE of explanation is between 7.00 and 7.10, and the gradient boosting decision tree model with the least number of independent variables;

[0043] Optionally, the parameter values ​​of the gradient boosting decision tree model: learning_rate is between 0.15 and 0.20, num_leaves is between 50 and 60, and max_depth is between 0 and 5.

[0044] The third aspect of the present invention provides a computer device, which includes: a memory and a processor, the memory is used to store program instructions; the processor is used to call program instructions, and when the program instructions are executed, the human biological age prediction method described in the first aspect of the present invention is implemented.

[0045] The fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for predicting human biological age as described in the first aspect of the present invention is implemented.

[0046] The fifth aspect of the present invention provides the application of the human biological age prediction method described in the first aspect of the present invention, the human biological age prediction system described in the second aspect of the present invention, the computer device described in the third aspect of the present invention or the computer-readable storage medium described in the fourth aspect of the present invention in human cell regeneration.

[0047] The beneficial effects of the present invention are as follows:

[0048] (1) The present invention has developed for the first time a new, efficient, accurate, and low-cost method, system, computer device, and storage medium for predicting human biological age. The present invention applies a gradient boosting decision tree model to predict human biological age, finds specific CpGs related to human biological age among all CpGs, and significantly reduces the number of CpGs in the identification process while also having high analysis efficiency and accuracy, which has important application value in the field of human age prediction;

[0049] (2) The present invention uses a gradient boosting decision tree model to rationally select methylation sites related to human biological age prediction, and then applies it to human biological age prediction, which greatly reduces the number of methylation sites required in the process of human biological age prediction and thus greatly reduces the cost. At the same time, it also has extremely high analysis speed and accuracy, which is of great significance for human biological age prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 A schematic diagram of a flow chart of a method for predicting biological age of a human provided by an embodiment of the present invention;

[0052] Figure 2 A schematic diagram of a human biological age prediction system provided by an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of a computer device for predicting biological age of a human provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0055] In some of the processes described in the specification and claims of the present invention and the above-mentioned drawings, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as S101, S102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0057] Figure 11 is a flow chart of a method for predicting human biological age provided by an embodiment of the present invention. Specifically, the method comprises the following steps:

[0058] S101: Acquire data, acquire methylation level data of methylation sites of a human sample to be tested;

[0059] In one embodiment, the human sample to be tested refers to a substance derived from the human to be tested and suitable for detecting the methylation level of the methylation site. Specifically, the human sample to be tested includes but is not limited to: blood samples, tissue samples, saliva samples, urine samples, paraffin-embedded samples, pleural effusion samples, alveolar lavage fluid samples, ascites and lavage fluid samples, bile samples, stool samples, cerebrospinal fluid samples, cell smear samples, cervical scraping or brush samples, and / or cell biopsy samples.

[0060] In one embodiment, the methylation site is a CpG, and the CpG includes: cg08160331, cg13100137, cg11529819, cg04875128, cg07927379, cg21200841, cg03182374, cg02780919, cg06654079, cg16541931, cg23100720, cg27314761, cg16600634, cg23220814, cg10633463, cg17461003, cg13954457, cg03705912, cg20388732, cg10652277, cg g06493994, cg10501210, cg16066354, cg10254000, cg14423778, cg036605 00, cg16717122, cg13902645, cg27503921, cg18786171, cg03157588, cg01 287391, cg21186299, cg01543654, cg02546818, cg19281363, cg03521358, cg02501116, cg19280121, cg14172603, cg23998119, cg08767286, cg244568 46. ​​cg25046584, cg11071401, cg20923605, cg21646366, cg16620537, cg17 207590, cg00863397, cg01639712, cg08918057, cg00481951, cg11268834, cg08949408, cg18576044, cg00751072, cg08976646, cg14122300, cg17140 797, cg06575035, cg06205172, cg24401044, cg14032732, cg06916591, cg18 686747, cg05929864, cg14569771, cg09748749, cg22491171, cg23166357, cg21951594, cg25478614, cg20835708, cg23454205, cg07063171, cg24891 660, cg24221648, cg07505391, cg16145113, cg03404211, cg19214184, cg0 4892758, cg25307778, cg02388276, cg22361181, cg10151953, cg12918357,cg26894354、cg16987524、cg27429080、cg21671476、cg01092293、cg01337508、cg07326413、cg27288268、cg00793935、cg03231647、cg25804860、cg24150153、cg19765038、cg07952813、cg00320138、cg02376524、cg25991761、cg24639902、cg00628697、cg06594186、cg03348196、cg22489583、cg19395089、cg00944163、cg02605232、cg07277624、cg26259381、cg24796644、cg06793562、cg22799132、cg13088417、cg07038098、cg14016806、cg14155388、cg02876258、cg07810366、cg13001142、cg04871807、cg06737250、cg06962177、cg21985251、cg05090150、cg02401978、cg01620309、cg10242901、cg17324339、cg10738648、cg21480605、cg02033788、cg24925741、cg12842219、cg07183637、cg13547577、cg25822326、cg22370920、cg22877428、cg01015395、cg00552235、cg21986027、cg14097619、cg03643998、cg16373229、cg25521439、cg23995914、cg07423651、cg02151408、cg24674302、cg12296007、cg05967403、cg08904363、cg01508045、cg00712289、cg04976479、cg00058515、cg04955333、cg24650267、cg06227171、cg19381811、cg17558214、cg03832839、cg10732215、cg11379315、cg20238412、cg09185773、cg08849574、cg11582100、cg03082830、cg24799830、cg04380669、cg15031299、cg09416096、cg25576248、cg20442599、cg09624466、cg04865110、cg20555778、cg09212058、cg21490980、cg06051250、cg07762234、cg11421509、cg19907331、cg10337027、cg02318784、cg03642066、cg25945412、cg03018489、cg14819399、cg22763680、cg01588224、cg06024113、cg04155348、cg11898347、cg15660573、cg06655623、cg16540704、cg16001713、cg22001782、cg06654537、cg04981556、cg12099357、cg02254261、cg02153893、cg18166915、cg23349790、cg25735922、cg13231700、cg07553761、cg06084312、cg21077300、cg26355573、cg21142272、cg01382414、cg08575537、cg03018478、cg04158367、cg18500368、cg13299548、cg21525022、cg01612220、cg05374654、cg19477084、cg24107848、cg01919377、cg16605942、cg10942775、cg06132502、cg03594515、cg20003983、cg01378090、cg23754392、cg19609502、cg23934731、cg04084157、cg02330214、cg01601746、cg22994272、cg15065631、cg01427815、cg08462478、cg16805188、cg19723775、cg18372207、cg03196745、cg25337514、cg05972216、cg26032101、cg18416022、cg01716527、cg27059389、cg16337574、cg03605463、cg11155865、cg26070134、cg25453664、cg13806070、cg21707131、cg09626984、cg04374321、cg24917945、cg19019804、cg03005261、cg24166694、cg08667024、cg02976054、cg11938672、cg17960615、cg16494530、cg05859099、cg16170380、cg09610735、cg26880297、cg08384155、cg13612317、cg20490724、cg10943458、cg04342223、cg20763254、cg10947146、cg20461538、cg13488284、cg27250180、cg03349582、cg24409876、cg23656386、cg20384941、cg21064080、cg02605178、cg23815043、cg09624953、cg11438134、cg14551562、cg00270789、cg15594750、cg01921437、cg18582581、cg14415160、cg10430205、cg15746415、cg26782825、cg16110704、cg00829753、cg14941172、cg18050295、cg08096786、cg20273670、cg07047131、cg24524735、cg16823105、cg19627034、cg11725689、cg04007931、cg12476754、cg17009433、cg08076018、cg06285925、cg01580681、cg08998953、cg16015712、cg10448759、cg02524035、cg13390332、cg23715407、cg08267245、cg03601797、cg19684151、cg10225865、cg07955752、cg11806594、cg09662747、cg21800400、cg15340644、cg25021259、cg10983013、cg08371947、cg24855943、cg23895439、cg02200491、cg08263226、cg10738003、cg16831440、cg13370746、cg26922511、cg08431893、cg04961189、cg22094845、cg22942169、cg02577651、cg18363918、cg00031722、cg17598358、cg08095377、cg19624775、cg17968943, cg05183226, cg25961432, cg05648225, cg05906092, cg27284034, cg16842187, cg19547459, cg00933538, cg04427498, cg05168977, cg24991933, cg01573825, cg08231697, cg14985989, cg24962672, cg00042657, cg13986215, cg01965874, cg01136191, cg03598412, cg00001583, cg26968804, cg08979194, cg08719712, cg06759255, cg12974756, cg24136563, cg23214840, cg15732107, cg11795308, cg15704155, cg25211348, cg22478179, cg20289913, cg08161546, cg01807241, cg15185717, cg18107144, cg01822570, cg17580331, cg20679403, cg00350060, cg02058358, cg20455931, cg06470855, cg25430507, cg11858450, cg02256410, cg04244183, cg02008727, cg02754763, cg06868622, cg16671365, cg13415371, cg19917720, cg11561665, cg09475262, cg05314124, cg24166520, cg14517502, cg09740598, cg15105660, cg23082620. ,

[0061] In one embodiment, the methylation level data of the methylation site is obtained by performing methylation sequencing on the human sample to be tested and then preprocessing the methylation level data.

[0062] In one embodiment, the preprocessing includes the following process: converting the red and green value signal data of the methylation sequencing CpG into methylation level data, the CpG sequencing red and green value signal data includes methylated CpG and unmethylated CpG, and the ratio of the number of methylated CpGs to the sum of the total number of CpGs is the Beta value, which is the quantitative value of the CpG methylation level.

[0063] In one embodiment, any methylation sequencing technology that can perform methylation sequencing on the sample to be tested is within the scope of protection of the present invention. Specifically, the methylation sequencing technology includes but is not limited to: whole genome bisulfite methylation sequencing (WGBS), reduced bisulfite methylation sequencing (RRBS), high-throughput single-cell methylation sequencing (sc-RBS), target gene DNA methylation sequencing, precise DNA methylation / hydroxymethylation sequencing (oxBS-seq), single-cell and micro-sample DNA methylation sequencing (Micro DNA-BS), micro-cfDNA genome methylation sequencing (cfDNA-BS), and / or (h)MeDIP sequencing.

[0064] In one embodiment, the methylation level data of the methylation site refers to the ratio of the number of methylated bases in a specific site or region to the total number of covered bases. This ratio reflects the degree of methylation at a specific position in the DNA sequence and is an important indicator for measuring the state of DNA methylation.

[0065] S102: Analyze data, input the methylation level data of the methylation site into a constructed prediction model, and the prediction model predicts the biological age of the human being to be tested based on the methylation level data of the methylation site;

[0066] In one embodiment, the prediction model is a gradient boosting decision tree model, and the method for constructing the prediction model includes:

[0067] Obtain methylation level data and biological age information for human population samples;

[0068] The human population samples are randomly divided into a training set and a test set, and based on the methylation level and biological age information corresponding to the training set, the mlr3verse::tnr function is used to fit N different gradient boosting decision tree models based on N groups of different hyperparameter values ​​learning_rate, num_leaves, and max_depth, and the methylation sites as described above are obtained, and the gradient boosting decision tree model that meets the predetermined requirements is selected as the prediction model, and the test set is used for verification to evaluate the effectiveness of the prediction model, so as to obtain a constructed prediction model;

[0069] Wherein, N is a natural number between 50 and 100.

[0070] In one embodiment, the gradient boosting decision tree model is constructed based on the mlr3verse software package using the methylation level data for feature selection. The present invention uses mlr3verse using the methylation level data for feature selection to construct a gradient boosting decision tree model to obtain two models (model 1 and model 2, wherein model 2 has a better effect) with high identification accuracy and fewer CpGs, which can achieve efficient and accurate prediction of human biological age.

[0071] In one embodiment, the selecting a gradient boosting decision tree model that meets the predetermined requirements includes: selecting a gradient boosting decision tree model that meets any of the following requirements from among the N different gradient boosting decision tree models constructed as the gradient boosting decision tree model that meets the predetermined requirements:

[0072] i) Gradient boosting decision tree model with the lowest misjudgment rate;

[0073] ii) The RMSE of explanation is between 7.00-7.10 and the gradient boosting decision tree model with the least number of independent variables.

[0074] In one embodiment, the present invention selects as few points as possible within a reasonable range of explanatory power as follows: when the explanatory power meets the requirements, the non-zero CpG is the least, and the explanatory power is a specific value of RMSE, that is, the gap between the predicted value and the actual value is quantified by the root mean square error.

[0075] In one embodiment, the parameter values ​​of the gradient boosting decision tree model, learning_rate, are between 0.15 and 0.20, num_leaves are between 50 and 60, and max_depth is between 0 and 5.

[0076] In one embodiment, the boosted decision tree model obtained by constructing and screening in the present invention has a prediction RMSE between 7.00-7.10 and involves fewer CpGs, which can be used for accurate and efficient prediction of human biological age.

[0077] In one embodiment, the methylation level data and biological age information of the human population sample are obtained by the following method: performing methylation sequencing on the human population sample, obtaining the methylation level data of the human population to be tested after preprocessing, and obtaining the biological age information of the human population by accessing the GEO database;

[0078] The methylation site includes CpG;

[0079] The preprocessing includes the following process: converting the red and green value signal data of the methylation sequencing CpG into methylation level data, the CpG sequencing red and green value signal data includes methylated CpG and unmethylated CpG, and the ratio of the number of methylated CpGs to the sum of the total number of CpGs is the Beta value, which is the quantitative value of the CpG methylation level.

[0080] In one embodiment, the red and green signal data obtained from the methylation sequencing CpG is converted into methylation level data, i.e., Beta value, and the Beta value range for each CpG is 0-1. A value of 0 represents unmethylated, and a value of 1 represents fully methylated.

[0081] In one embodiment, the human population sample refers to a substance derived from humans that is suitable for detecting the methylation level of the methylation site. Specifically, the human population sample includes but is not limited to: blood samples, tissue samples, saliva samples, urine samples, paraffin-embedded samples, pleural effusion samples, alveolar lavage fluid samples, ascites and lavage fluid samples, bile samples, stool samples, cerebrospinal fluid samples, cell smear samples, cervical scraping or brush samples, and / or cell biopsy samples.

[0082] S103: Outputting a prediction result, predicting and outputting a prediction result of the biological age of the human being to be tested based on the output result of the prediction model;

[0083] In one embodiment, the biological age of the human being to be tested is predicted based on the methylation level data of the methylation site of the human being to be tested and the constructed prediction model.

[0084] Figure 2 : is a schematic diagram of a human biological age prediction system provided by an embodiment of the present invention. Specifically, the system includes:

[0085] A data acquisition unit is used to acquire methylation level data of the methylation site of the human sample to be tested;

[0086] A data analysis unit, used for inputting the methylation level data of the methylation site into a constructed prediction model, wherein the prediction model predicts the biological age of the human being to be tested based on the methylation level data of the methylation site;

[0087] The output result unit is used to predict and output the biological age prediction result of the human being to be tested based on the output result of the prediction model.

[0088] In one embodiment, the methylation site is a methylation site as described above.

[0089] In one embodiment, the prediction result is obtained based on the prediction model, and the method for constructing the prediction model includes:

[0090] Obtain methylation level data and biological age information for human population samples;

[0091] The human population samples are randomly divided into a training set and a test set, and based on the methylation level and biological age information corresponding to the training set, the mlr3verse::tnr function is used to fit N different gradient boosting decision tree models based on N groups of different hyperparameter values ​​learning_rate, num_leaves, and max_depth, and the methylation sites as described above are obtained, and the gradient boosting decision tree model that meets the predetermined requirements is selected as the prediction model, and the test set is used for verification to evaluate the effectiveness of the prediction model, so as to obtain a constructed prediction model;

[0092] Wherein, N is a natural number between 50 and 100;

[0093] In one embodiment, the human population sample refers to a substance derived from humans that is suitable for detecting the methylation level of the methylation site. Specifically, the human population sample includes but is not limited to: blood samples, tissue samples, saliva samples, urine samples, paraffin-embedded samples, pleural effusion samples, alveolar lavage fluid samples, ascites and lavage fluid samples, bile samples, stool samples, cerebrospinal fluid samples, cell smear samples, cervical scraping or brush samples, and / or cell biopsy samples.

[0094] In one embodiment, the gradient boosting decision tree model is constructed based on the mlr3verse software package.

[0095] In one embodiment, the selecting a gradient boosting decision tree model that meets the predetermined requirements includes: selecting a gradient boosting decision tree model that meets any of the following requirements from among the N different gradient boosting decision tree models constructed as the gradient boosting decision tree model that meets the predetermined requirements:

[0096] i) Gradient boosting decision tree model with the lowest misjudgment rate;

[0097] ii) The RMSE of explanation is between 7.00-7.10 and the gradient boosting decision tree model with the least number of independent variables.

[0098] In one embodiment, the parameter values ​​of the gradient boosting decision tree model, learning_rate, are between 0.15 and 0.20, num_leaves are between 50 and 60, and max_depth is between 0 and 5.

[0099] Figure 3A schematic diagram of a computer device for predicting human biological age is provided for an embodiment of the present invention. Specifically, the device includes: a memory and a processor, the memory is used to store program instructions; the processor is used to call program instructions, and when the program instructions are executed, the human biological age prediction method provided by the present invention is implemented.

[0100] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the human biological age prediction method provided by the present invention is implemented.

[0101] In one embodiment, the computer-readable storage medium may include, but is not limited to, a floppy disk, an optical disk, a CD-ROM (Compact Disc Read Only Memory), a magneto-optical disk, a ROM (Read Only Memory), a RAM (Random Access Memory), an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Programmable Read Only Memory), a magnetic card or an optical card, a flash memory, or other types of media / machine-readable media suitable for storing machine-executable instructions. The computer-readable storage medium may be a product that is not connected to a computer device, or may be a component that is connected to a computer device for use.

[0102] The verification results of this verification example show that assigning inherent weights to indications can moderately improve the performance of the method relative to the default settings. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0103] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0104] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0105] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0106] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0107] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned storage medium can be a read-only memory, a disk or an optical disk, etc.

[0108] The above is a detailed introduction to a computer device provided by the present invention. For a person skilled in the art, according to the concept of the embodiments of the present invention, there may be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

[0109] Example 1

[0110] This embodiment provides a prediction method for predicting human biological age and a method for constructing a related prediction model, which specifically includes the following steps:

[0111] 1. Methylation sequencing

[0112] The experimental material is methylation level data obtained by methylation sequencing after DNA is extracted from 214 human blood samples, and the biological age information corresponding to the human samples is obtained by accessing the GEO database.

[0113] 2. Modeling

[0114] 2.1 Preprocessing of methylation sequencing data

[0115] Low-quality samples that did not meet quality control criteria were removed by minfi::preprocessIllumina.

[0116] 2.1.1 Generating methylation level data

[0117] The idat file after filtering and quality control is converted into methylation level data. For each human sample to be tested, a red signal file _Red.idat and a green signal file _Grn.idat are generated. The red and green signal files are converted into signal intensities of CpG probes with p-values ​​through minfi::read.metharray, and low-quality CpGs are filtered by p-values. The methylation level data for input into the downstream model is obtained through minfi::getBeta.

[0118] 2.2 Gradient boosting decision tree fitting modeling

[0119] First, 90% of the 214 human individual samples were randomly selected as the training set, and 10% were randomly selected as the test set. The test set does not participate in the modeling process and is only used to predict and evaluate the effect of the model. The mlr3verse and mlr3verse::tnr functions in the package are called to model, and the model is tuned through 9-fold cross validation to find the parameter values ​​of the model with the best performance.

[0120] 2.2.1 Model establishment

[0121] (1) The mlr3verse function uses all the train data to fit the model at one time, which may cause overfitting. The false positive rate may increase in subsequent predictions: 50 sets of hyperparameter values ​​learning_rate, num_leaves, and max_depth are automatically selected through grid search to fit 50 models with different coefficients;

[0122] (2) In the mlr3::rsmp function, nfolds=5 is specified to perform 5-fold cross-validation fitting and then select the model to improve the accuracy of the model to obtain 100 models;

[0123] Finally, the model constructed by mlr3::rsmp function was used for subsequent optimization and screening.

[0124] 2.2.2 Further optimization of model parameters

[0125] There are still many sites retained in the results of the previous step. Considering that the fewer the sites, the lower the cost, a model with as few sites as possible is selected within a reasonable range of explanation for effect evaluation. By using grid search to optimize the hyperparameters, two models that meet the requirements are finally obtained (model 1 and model 2 as described in Table 1 below).

[0126] Table 1 Parameters of different models

[0127]

[0128]

[0129] Note: The explanation in the table is obtained by evaluating the effect using the data of the test set.

[0130] The results show that model 1 retains 872 methylation sites, with a corresponding RMSE of 9.52, and model 2 retains 425 methylation sites (see the methylation sites in the feature selection CpG list of the optimal model described below), with a corresponding RMSE of 7.05. It can be seen that model 2 contains fewer methylation sites (the corresponding detection cost will be lower), and the prediction accuracy is higher and the error is smaller.

[0131] 3. Related formulas for calculation

[0132] The above decision tree model using gradient boosting is a tree model, and the following structure is expected to exist.

[0133] Gradient boosted decision tree (GBDT) framework: The gradient boosting algorithm minimizes the loss function by iteratively fitting the residual tree. At each step, the model update formula is:

[0134] Fm(x)=Fm-1(x)+h(x;θm)

[0135] Among them, Fm(x) is the model after the mth tree, Fm-1(x) is the model of the previous round, h(x;θm) is the newly fitted tree, and θm is the parameter of the tree.

[0136] Gain calculation: When building a decision tree, mlr3verse calculates the gain of the split node to determine the best split point. The gain can be expressed by the following formula:

[0137]

[0138] Among them, Gl and Gr are the gradient sum of the left and right child nodes after splitting, Hl and Hr are the second-order gradient sums, and λ is the regularization parameter.

[0139] Histogram difference acceleration: mlr3verse accelerates the search for the best split point by constructing a feature histogram. The histogram of a leaf can be obtained by subtracting the histogram of its parent node from the histogram of its sibling node:

[0140] Histogramleaf=Histogramparent+Histogramsibling

[0141] Leaf-wise growth strategy: mlr3verse adopts Leaf-wise growth strategy, each time it finds a leaf with the largest split gain from all current leaves for splitting. This method can reduce more errors and improve accuracy.

[0142] Parallel support and optimization: mlr3verse natively supports parallel learning, including feature parallelism and data parallelism. In feature parallelism, different machines search for the optimal split point on different feature sets; in data parallelism, different machines first construct histograms locally and then merge them globally.

[0143] Network communication optimization: mlr3verse reduces network communication costs by using collective communication algorithms, achieving linear acceleration of parallel computing.

[0144] 4. Feature selection CpG list for the optimal model

[0145] [1]"cg08160331""cg13100137""cg11529819""cg04875128""cg07927379"

[0146] [6]"cg21200841""cg03182374""cg02780919""cg06654079""cg16541931"

[0147]

[11] "cg23100720""cg27314761""cg16600634""cg23220814""cg10633463"

[0148]

[16] "cg17461003""cg13954457""cg03705912""cg20388732""cg10652277"

[0149]

[21] "cg06493994""cg10501210""cg16066354""cg10254000""cg14423778"

[0150]

[26] "cg03660500""cg16717122""cg13902645""cg27503921""cg18786171"

[0151]

[31] "cg03157588""cg01287391""cg21186299""cg01543654""cg02546818"

[0152]

[36] "cg19281363""cg03521358""cg02501116""cg19280121""cg14172603"

[0153]

[41] "cg23998119""cg08767286""cg24456846""cg25046584""cg11071401"

[0154]

[46] "cg20923605""cg21646366""cg16620537""cg17207590""cg00863397"

[0155]

[51] "cg01639712""cg08918057""cg00481951""cg11268834""cg08949408"

[0156]

[56] "cg18576044""cg00751072""cg08976646""cg14122300""cg17140797"

[0157]

[61] "cg06575035""cg06205172""cg24401044""cg14032732""cg06916591"

[0158]

[66] "cg18686747""cg05929864""cg14569771""cg09748749""cg22491171"

[0159]

[71] "cg23166357""cg21951594""cg25478614""cg20835708""cg23454205"

[0160]

[76] "cg07063171""cg24891660""cg24221648""cg07505391""cg16145113"

[0161]

[81] "cg03404211""cg19214184""cg04892758""cg25307778""cg02388276"

[0162]

[86] "cg22361181""cg10151953""cg12918357""cg26894354""cg16987524"

[0163]

[91] "cg27429080""cg21671476""cg01092293""cg01337508""cg07326413"

[0164]

[96] "cg27288268""cg00793935""cg03231647""cg25804860""cg24150153"

[0165]

[101] "cg19765038""cg07952813""cg00320138""cg02376524""cg25991761"

[0166]

[106] "cg24639902""cg00628697""cg06594186""cg03348196""cg22489583"

[0167]

[111] "cg19395089""cg00944163""cg02605232""cg07277624""cg26259381"

[0168]

[116] "cg24796644""cg06793562""cg22799132""cg13088417""cg07038098"

[0169]

[121] "cg14016806""cg14155388""cg02876258""cg07810366""cg13001142"

[0170]

[126] "cg04871807""cg06737250""cg06962177""cg21985251""cg05090150"

[0171]

[131] "cg02401978""cg01620309""cg10242901""cg17324339""cg10738648"

[0172]

[136] "cg21480605""cg02033788""cg24925741""cg12842219""cg07183637"

[0173]

[141] "cg13547577""cg25822326""cg22370920""cg22877428""cg01015395"

[0174]

[146] "cg00552235""cg21986027""cg14097619""cg03643998""cg16373229"

[0175]

[151] "cg25521439""cg23995914""cg07423651""cg02151408""cg24674302"

[0176]

[156] "cg12296007""cg05967403""cg08904363""cg01508045""cg00712289"

[0177]

[161] "cg04976479""cg00058515""cg04955333""cg24650267""cg06227171"

[0178]

[166] "cg19381811""cg17558214""cg03832839""cg10732215""cg11379315"

[0179]

[171] "cg20238412""cg09185773""cg08849574""cg11582100""cg03082830"

[0180]

[176] "cg24799830""cg04380669""cg15031299""cg09416096""cg25576248"

[0181]

[181] "cg20442599""cg09624466""cg04865110""cg20555778""cg09212058"

[0182]

[186] "cg21490980""cg06051250""cg07762234""cg11421509""cg19907331"

[0183]

[191] "cg10337027""cg02318784""cg03642066""cg25945412""cg03018489"

[0184]

[196] "cg14819399""cg22763680""cg01588224""cg06024113""cg04155348"

[0185]

[201] "cg11898347""cg15660573""cg06655623""cg16540704""cg16001713"

[0186]

[206] "cg22001782""cg06654537""cg04981556""cg12099357""cg02254261"

[0187]

[211] "cg02153893""cg18166915""cg23349790""cg25735922""cg13231700"

[0188]

[216] "cg07553761""cg06084312""cg21077300""cg26355573""cg21142272"

[0189]

[221] "cg01382414""cg08575537""cg03018478""cg04158367""cg18500368"

[0190]

[226] "cg13299548""cg21525022""cg01612220""cg05374654""cg19477084"

[0191]

[231] "cg24107848""cg01919377""cg16605942""cg10942775""cg06132502"

[0192]

[236] "cg03594515""cg20003983""cg01378090""cg23754392""cg19609502"

[0193]

[241] "cg23934731""cg04084157""cg02330214""cg01601746""cg22994272"

[0194]

[246] "cg15065631""cg01427815""cg08462478""cg16805188""cg19723775"

[0195]

[251] "cg18372207""cg03196745""cg25337514""cg05972216""cg26032101"

[0196]

[256] "cg18416022""cg01716527""cg27059389""cg16337574""cg03605463"

[0197]

[261] "cg11155865""cg26070134""cg25453664""cg13806070""cg21707131"

[0198]

[266] "cg09626984""cg04374321""cg24917945""cg19019804""cg03005261"

[0199]

[271] "cg24166694""cg08667024""cg02976054""cg11938672""cg17960615"

[0200]

[276] "cg16494530""cg05859099""cg16170380""cg09610735""cg26880297"

[0201]

[281] "cg08384155""cg13612317""cg20490724""cg10943458""cg04342223"

[0202]

[286] "cg20763254""cg10947146""cg20461538""cg13488284""cg27250180"

[0203]

[291] "cg03349582""cg24409876""cg23656386""cg20384941""cg21064080"

[0204]

[296] "cg02605178""cg23815043""cg09624953""cg11438134""cg14551562"

[0205]

[301] "cg00270789""cg15594750""cg01921437""cg18582581""cg14415160"

[0206]

[306] "cg10430205""cg15746415""cg26782825""cg16110704""cg00829753"

[0207]

[311] "cg14941172""cg18050295""cg08096786""cg20273670""cg07047131"

[0208]

[316] "cg24524735""cg16823105""cg19627034""cg11725689""cg04007931"

[0209]

[321] "cg12476754""cg17009433""cg08076018""cg06285925""cg01580681"

[0210]

[326] "cg08998953""cg16015712""cg10448759""cg02524035""cg13390332"

[0211]

[331] "cg23715407""cg08267245""cg03601797""cg19684151""cg10225865"

[0212]

[336] "cg07955752""cg11806594""cg09662747""cg21800400""cg15340644"

[0213]

[341] "cg25021259""cg10983013""cg08371947""cg24855943""cg23895439"

[0214]

[346] "cg02200491""cg08263226""cg10738003""cg16831440""cg13370746"

[0215]

[351] "cg26922511""cg08431893""cg04961189""cg22094845""cg22942169"

[0216]

[356] "cg02577651""cg18363918""cg00031722""cg17598358""cg08095377"

[0217]

[361] "cg19624775""cg17968943""cg05183226""cg25961432""cg05648225"

[0218]

[366] "cg05906092""cg27284034""cg16842187""cg19547459""cg00933538"

[0219]

[371] "cg04427498""cg05168977""cg24991933""cg01573825""cg08231697"

[0220]

[376] "cg14985989""cg24962672""cg00042657""cg13986215""cg01965874"

[0221]

[381] "cg01136191""cg03598412""cg00001583""cg26968804""cg08979194"

[0222]

[386] "cg08719712""cg06759255""cg12974756""cg24136563""cg23214840"

[0223]

[391] "cg15732107""cg11795308""cg15704155""cg25211348""cg22478179"

[0224]

[396] "cg20289913""cg08161546""cg01807241""cg15185717""cg18107144"

[0225]

[401] "cg01822570""cg17580331""cg20679403""cg00350060""cg02058358"

[0226]

[406] "cg20455931""cg06470855""cg25430507""cg11858450""cg02256410"

[0227]

[411] "cg04244183""cg02008727""cg02754763""cg06868622""cg16671365"

[0228]

[416] "cg13415371""cg19917720""cg11561665""cg09475262""cg05314124"

[0229]

[421] "cg24166520""cg14517502""cg09740598""cg15105660""cg23082620"。

Claims

1. A method for predicting human biological age, characterized in that: The method comprises: Acquiring data, acquiring methylation level data of the methylation sites of the human sample to be tested; Analyzing data, inputting the methylation level data of the methylation site into a constructed prediction model, wherein the prediction model predicts the biological age of the human subject to be tested based on the methylation level data of the methylation site; Outputting a prediction result, predicting and outputting a prediction result of the biological age of the human being to be tested based on the output result of the prediction model; The methylation site is CpG, and the CpG includes: cg08160331, cg13100137, cg11529819, cg04875128, cg07927379, cg21200841, cg03182374, cg02780919, cg06654079, cg16541931, cg23100720, cg27314761, cg16600634, cg23220814, cg10633463, cg17461003, cg13954457, cg03705912, cg20388732, cg10652277, cg0649399 4. cg10501210, cg16066354, cg10254000, cg14423778, cg03660500, cg167 17122, cg13902645, cg27503921, cg18786171, cg03157588, cg01287391, cg 21186299, cg01543654, cg02546818, cg19281363, cg03521358, cg0250111 6. cg19280121, cg14172603, cg23998119, cg08767286, cg24456846, cg2504 6584, cg11071401, cg20923605, cg21646366, cg16620537, cg17207590, cg 00863397, cg01639712, cg08918057, cg00481951, cg11268834, cg08949408 , cg18576044, cg00751072, cg08976646, cg14122300, cg17140797, cg0657 5035, cg06205172, cg24401044, cg14032732, cg06916591, cg18686747, cg0 5929864, cg14569771, cg09748749, cg22491171, cg23166357, cg21951594 , cg25478614, cg20835708, cg23454205, cg07063171, cg24891660, cg24221 648, cg07505391, cg16145113, cg03404211, cg19214184, cg04892758, cg2 5307778, cg02388276, cg22361181, cg10151953, cg12918357, cg26894354,cg16987524、cg27429080、cg21671476、cg01092293、cg01337508、cg07326413、cg27288268、cg00793935、cg03231647、cg25804860、cg24150153、cg19765038、cg07952813、cg00320138、cg02376524、cg25991761、cg24639902、cg00628697、cg06594186、cg03348196、cg22489583、cg19395089、cg00944163、cg02605232、cg07277624、cg26259381、cg24796644、cg06793562、cg22799132、cg13088417、cg07038098、cg14016806、cg14155388、cg02876258、cg07810366、cg13001142、cg04871807、cg06737250、cg06962177、cg21985251、cg05090150、cg02401978、cg01620309、cg10242901、cg17324339、cg10738648、cg21480605、cg02033788、cg24925741、cg12842219、cg07183637、cg13547577、cg25822326、cg22370920、cg22877428、cg01015395、cg00552235、cg21986027、cg14097619、cg03643998、cg16373229、cg25521439、cg23995914、cg07423651、cg02151408、cg24674302、cg12296007、cg05967403、cg08904363、cg01508045、cg00712289、cg04976479、cg00058515、cg04955333、cg24650267、cg06227171、cg19381811、cg17558214、cg03832839、cg10732215、cg11379315、cg20238412、cg09185773、cg08849574、cg11582100、cg03082830、cg24799830、cg04380669、cg15031299、cg09416096、cg25576248、cg20442599、cg09624466、cg04865110、cg20555778、cg09212058、cg21490980、cg06051250、cg07762234、cg11421509、cg19907331、cg10337027、cg02318784、cg03642066、cg25945412、cg03018489、cg14819399、cg22763680、cg01588224、cg06024113、cg04155348、cg11898347、cg15660573、cg06655623、cg16540704、cg16001713、cg22001782、cg06654537、cg04981556、cg12099357、cg02254261、cg02153893、cg18166915、cg23349790、cg25735922、cg13231700、cg07553761、cg06084312、cg21077300、cg26355573、cg21142272、cg01382414、cg08575537、cg03018478、cg04158367、cg18500368、cg13299548、cg21525022、cg01612220、cg05374654、cg19477084、cg24107848、cg01919377、cg16605942、cg10942775、cg06132502、cg03594515、cg20003983、cg01378090、cg23754392、cg19609502、cg23934731、cg04084157、cg02330214、cg01601746、cg22994272、cg15065631、cg01427815、cg08462478、cg16805188、cg19723775、cg18372207、cg03196745、cg25337514、cg05972216、cg26032101、cg18416022、cg01716527、cg27059389、cg16337574、cg03605463、cg11155865、cg26070134、cg25453664、cg13806070、cg21707131、cg09626984、cg04374321、cg24917945、cg19019804、cg03005261、cg24166694、cg08667024、cg02976054、cg11938672、cg17960615、cg16494530、cg05859099、cg16170380、cg09610735、cg26880297、cg08384155、cg13612317、cg20490724、cg10943458、cg04342223、cg20763254、cg10947146、cg20461538、cg13488284、cg27250180、cg03349582、cg24409876、cg23656386、cg20384941、cg21064080、cg02605178、cg23815043、cg09624953、cg11438134、cg14551562、cg00270789、cg15594750、cg01921437、cg18582581、cg14415160、cg10430205、cg15746415、cg26782825、cg16110704、cg00829753、cg14941172、cg18050295、cg08096786、cg20273670、cg07047131、cg24524735、cg16823105、cg19627034、cg11725689、cg04007931、cg12476754、cg17009433、cg08076018、cg06285925、cg01580681、cg08998953、cg16015712、cg10448759、cg02524035、cg13390332、cg23715407、cg08267245、cg03601797、cg19684151、cg10225865、cg07955752、cg11806594、cg09662747、cg21800400、cg15340644、cg25021259、cg10983013、cg08371947、cg24855943、cg23895439、cg02200491、cg08263226、cg10738003、cg16831440、cg13370746、cg26922511、cg08431893、cg04961189、cg22094845、cg22942169、cg02577651、cg18363918、cg00031722、cg17598358、cg08095377、cg19624775、cg17968943、cg05183226、cg25961432、cg05648225、cg05906092、cg27284034、cg16842187、cg19547459、cg00933538、cg04427498、cg05168977、cg24991933、cg01573825、cg08231697、cg14985989、cg24962672、cg00042657、cg13986215、cg01965874、cg01136191、cg03598412、cg00001583、cg26968804、cg08979194、cg08719712、cg06759255、cg12974756、cg24136563、cg23214840、cg15732107、cg11795308、cg15704155、cg25211348、cg22478179、cg20289913、cg08161546、cg01807241、cg15185717、cg18107144、cg01822570、cg17580331、cg20679403、cg00350060、cg02058358、cg20455931、cg06470855、cg25430507、cg11858450、cg02256410、cg04244183、cg02008727、cg02754763、cg06868622、cg16671365、cg13415371、cg19917720、cg11561665、cg09475262、cg05314124、cg24166520、cg14517502、cg09740598、cg15105660、cg23082620。、 2. The method according to claim 1, characterized in that The prediction model is a gradient boosting decision tree model, and the method for constructing the prediction model includes: Obtain methylation level data and biological age information for human population samples; The human population samples are randomly divided into a training set and a test set, and based on the methylation level and biological age information corresponding to the training set, the mlr3verse::tnr function is used to fit N different gradient boosting decision tree models based on N groups of different hyperparameter values ​​learning_rate, num_leaves, and max_depth, and the methylation sites described in claim 1 are obtained, and the gradient boosting decision tree model that meets the predetermined requirements is selected as the prediction model, and the test set is used for verification to evaluate the effectiveness of the prediction model, so as to obtain a constructed prediction model; Wherein, N is a natural number between 50 and 100; Optionally, the sample includes a blood sample, a tissue sample, a saliva sample or a urine sample.

3. The method according to claim 2, characterized in that The gradient boosting decision tree model is constructed according to the mlr3verse software package; Optionally, the selecting the gradient boosting decision tree model that meets the predetermined requirements includes: selecting a gradient boosting decision tree model that meets any of the following requirements from among the N different gradient boosting decision tree models constructed as the gradient boosting decision tree model that meets the predetermined requirements: i) Gradient boosting decision tree model with the lowest misjudgment rate; ii) The RMSE of explanation is between 7.00-7.10 and the gradient boosting decision tree model with the least number of independent variables.

4. The method according to claim 3, characterized in that The parameter values ​​of the gradient boosting decision tree model are learning_rate between 0.15 and 0.20, num_leaves between 50 and 60, and max_depth between 0 and 5.

5. The method according to claim 2, characterized in that: The methylation level data and biological age information of the human population sample are obtained by the following method: Performing methylation sequencing on the human population samples, obtaining methylation level data of the human population to be tested after preprocessing, and obtaining biological age information of the human population by accessing the GEO database; Optionally, the methylation site includes CpG; Optionally, the preprocessing includes the following process: converting the red and green value signal data of the methylation sequencing CpG into methylation level data, the CpG sequencing red and green value signal data includes methylated CpG and unmethylated CpG, and the ratio of the number of methylated CpGs to the sum of the total number of CpGs is the Beta value, which is the quantitative value of the CpG methylation level.

6. A human biological age prediction system, characterized in that: The system comprises: A data acquisition unit is used to acquire methylation level data of the methylation site of the human sample to be tested; A data analysis unit, used for inputting the methylation level data of the methylation site into a constructed prediction model, wherein the prediction model predicts the biological age of the human being to be tested based on the methylation level data of the methylation site; An output result unit, used for predicting and outputting a prediction result of the biological age of the human being to be tested based on an output result of the prediction model; The methylation site is the methylation site described in claim 1; The prediction result is obtained based on the prediction model, and the method for constructing the prediction model includes: Obtain methylation level data and biological age information for human population samples; The human population samples are randomly divided into a training set and a test set, and based on the methylation level and biological age information corresponding to the training set, the mlr3verse::tnr function is used to fit N different gradient boosting decision tree models based on N groups of different hyperparameter values ​​learning_rate, num_leaves, and max_depth, and the methylation sites described in claim 1 are obtained, and the gradient boosting decision tree model that meets the predetermined requirements is selected as the prediction model, and the test set is used for verification to evaluate the effectiveness of the prediction model, so as to obtain a constructed prediction model; Wherein, N is a natural number between 50 and 100.

7. The human biological age prediction system according to claim 6, characterized in that: The sample includes a blood sample, a tissue sample, a saliva sample or a urine sample; Optionally, the gradient boosting decision tree model is constructed according to the mlr3verse software package; Optionally, the selecting the gradient boosting decision tree model that meets the predetermined requirements includes: selecting a gradient boosting decision tree model that meets any of the following requirements from among the N different gradient boosting decision tree models constructed as the gradient boosting decision tree model that meets the predetermined requirements: i) Gradient boosting decision tree model with the lowest misjudgment rate; ii) The RMSE of explanation is between 7.00-7.10 and the gradient boosting decision tree model with the least number of independent variables; Optionally, the parameter values ​​of the gradient boosting decision tree model: learning_rate is between 0.15 and 0.20, num_leaves is between 50 and 60, and max_depth is between 0 and 5.

8. A computer device, characterized in that: The computer device comprises: a memory and a processor, wherein the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, the human biological age prediction method described in any one of claims 1 to 5 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for predicting human biological age according to any one of claims 1 to 5 is implemented.

10. Application of the human biological age prediction method according to any one of claims 1 to 5, the human biological age prediction system according to claim 6 or 7, the computer device according to claim 8 or the computer-readable storage medium according to claim 9 in human cell regeneration.

Images