A prediction method, system, computer device, and storage medium for predicting sleep deprivation based on LIPG.

Abstract

This invention discloses a method, system, computer device, and storage medium for predicting sleep deprivation based on LIPG (Limited Insulin Progenitor Cell Protein). This invention belongs to the field of intelligent medical technology. Specifically, this invention provides a method, system, computer device, and computer-readable storage medium for predicting sleep deprivation. The method includes: acquiring protein expression data of a serum sample to be tested; extracting expression data of a target protein from the protein expression data, wherein the target protein is a LIPG protein; and performing classification prediction based on the expression data of the target protein to obtain a classification result indicating whether the serum sample to be tested is from a sleep-deprived subject. This invention provides information for early intervention and treatment of sleep deprivation and has broad application prospects.

Description

Technical Field

[0001] This invention belongs to the field of intelligent medical technology, specifically, it relates to a prediction method, system, computer device and storage medium for predicting sleep deprivation based on LIPG, and more specifically, it relates to a prediction method, prediction system, computer device and computer-readable storage medium for predicting sleep deprivation. Background Technology

[0002] In modern society, the fast pace of life, increasing work pressure, and the widespread use of smart electronic devices have led to a growing prevalence of irregular sleep patterns, insufficient sleep duration, and chronic sleep deprivation. Persistent sleep deprivation is a hidden chronic health hazard that can cause multiple damages to the nervous, endocrine, metabolic, and immune systems. It can induce cognitive decline, slowed thinking, low mood, and abnormal mood regulation, disrupt the body's physiological balance, significantly increase the probability of various chronic and mental illnesses, and, over the long term, severely damage physical health and reduce quality of life. Because the harms of sleep deprivation are gradual and cumulative, and often lack typical clinical symptoms in the early stages, they are easily overlooked. By the time obvious discomfort appears, irreversible damage to bodily functions has often already occurred. Therefore, developing a technological solution for the early, objective, and accurate prediction of sleep deprivation is of significant clinical importance and necessity for early health intervention, preventing the chain reaction of health damage caused by sleep deprivation, and protecting the physical and mental health of the population.

[0003] Currently, there is no mature and standardized quantitative detection system for sleep deprivation in clinical practice and daily health management. Existing sleep assessment methods all have significant technical shortcomings and are difficult to adapt to diverse testing scenarios. The mainstream assessment methods at present are divided into two categories: subjective assessment and instrumental testing. Subjective assessment methods such as sleep questionnaires and self-reports are highly dependent on individual memory, subjective perception, and emotional state, resulting in highly random and unobjective assessments that cannot be used as accurate diagnostic criteria. While polysomnography (PSG) is a professional sleep testing method with relatively high accuracy, it suffers from problems such as expensive equipment, cumbersome procedures, stringent testing conditions, and long testing times. It is only suitable for clinical diagnostic testing and cannot be applied to large-scale screening or routine health monitoring in the general population. Meanwhile, conventional smart wearable devices on the market can only indirectly estimate sleep status based on basic data such as body surface activity and heart rate, resulting in significant data errors and lacking clinical diagnostic validity. In summary, the industry currently lacks a sleep deprivation detection and prediction technology that is based on specific biomarkers, easy to operate, provides objective and accurate results, and can be applied on a large scale. There is an urgent need to develop a brand-new intelligent prediction solution to break through existing technological barriers and meet the dual needs of clinical screening and public health monitoring. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a prediction method, system, computer device and storage medium for predicting sleep deprivation based on LIPG.

[0005] This invention aims to address the technical problem in existing technologies that lack biomarkers and intelligent prediction methods for objectively and quantitatively assessing sleep deprivation, leading to large errors in subjective questionnaire assessments and an inability to achieve accurate screening. To this end, this invention provides a sleep deprivation prediction method, prediction system, computer equipment, and computer-readable storage medium based on the serum biomarker LIPG. By detecting the expression level of LIPG protein in a serum sample and using a machine learning model to process and classify the expression data, this invention can objectively and accurately output the classification result of whether the sample is a sleep deprivation sample, providing objective quantitative evidence for clinicians and health management personnel, thereby achieving early warning and personalized intervention treatment for sleep deprivation risk.

[0006] The present invention achieves the above-mentioned objectives by adopting the following technical solution: A first aspect of the present invention provides a method for predicting sleep deprivation, the method comprising: Obtain protein expression data from the serum sample to be tested; Extract the expression data of the target protein from the protein expression data, wherein the target protein is LIPG protein; Based on the expression data of the target protein, classification prediction is performed to obtain the classification result of whether the serum sample to be tested is a serum sample of a sleep-deprived subject; The classification result is obtained based on a prediction model. The method for constructing the prediction model includes: obtaining the target protein expression data and corresponding clinical features of the training set samples, including sleep-deprived subjects and healthy individuals; extracting the target protein expression data from the training set and inputting it into a machine learning model to construct a prediction model; and obtaining the constructed prediction model. If the expression level of the target protein LIPG is below the threshold, the serum sample to be tested is classified as a serum sample from a sleep-deprived subject; if the expression level of the target protein LIPG is above the threshold, the serum sample to be tested is classified as a serum sample from a non-sleep-deprived subject.

[0007] Furthermore, the machine learning models include linear regression models, logistic regression models, random forest models, Lasso regression models, Ridge regression models, linear discriminant analysis models, nearest neighbor models, decision tree models, perceptron models, neural network models, support vector machine models, Naive Bayes models, AdaBoost models, GBDT models, XGBoost models, LightGBM models, and / or CatBoost models.

[0008] Furthermore, the protein expression data includes the expression level data of the target protein LIPG; The protein expression data are obtained by enzyme-linked immunosorbent assay (ELISA), mass spectrometry, chemiluminescence, immunochromatography, Western blotting, immunohistochemistry, or co-immunoprecipitation. Optionally, the sleep deprivation refers to an average of less than 7 hours of actual sleep per night for at least one week.

[0009] A second aspect of the present invention provides a sleep deprivation prediction system, the system comprising: The data acquisition unit is used to acquire protein expression data from the serum sample to be tested. The data extraction unit is used to extract the expression data of the target protein from the protein expression data, wherein the target protein is LIPG protein; The prediction result unit is used to perform classification prediction based on the expression data of the target protein to obtain the classification result of whether the serum sample to be tested is a serum sample of a sleep-deprived subject; The classification result is obtained based on a prediction model. The method for constructing the prediction model includes: obtaining the target protein expression data and corresponding clinical features of the training set samples, including sleep-deprived subjects and healthy individuals; extracting the target protein expression data from the training set and inputting it into a machine learning model to construct a prediction model; and obtaining the constructed prediction model. If the expression level of the target protein LIPG is below the threshold, the serum sample to be tested is classified as a serum sample from a sleep-deprived subject; if the expression level of the target protein LIPG is above the threshold, the serum sample to be tested is classified as a serum sample from a non-sleep-deprived subject.

[0010] Furthermore, the machine learning models include linear regression models, logistic regression models, random forest models, Lasso regression models, Ridge regression models, linear discriminant analysis models, nearest neighbor models, decision tree models, perceptron models, neural network models, support vector machine models, Naive Bayes models, AdaBoost models, GBDT models, XGBoost models, LightGBM models, and / or CatBoost models.

[0011] Furthermore, the protein expression data includes the expression level data of the target protein LIPG; The protein expression data are obtained by enzyme-linked immunosorbent assay (ELISA), mass spectrometry, chemiluminescence, immunochromatography, Western blotting, immunohistochemistry, or co-immunoprecipitation. Optionally, the sleep deprivation refers to an average of less than 7 hours of actual sleep per night for at least one week.

[0012] A third aspect of the present invention provides a computer device comprising: a memory and a processor, the memory being used to store program instructions; the processor being used to invoke the program instructions, wherein when the program instructions are executed, the sleep deprivation prediction method of the first aspect of the present invention is implemented.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the sleep deprivation prediction method described in the first aspect of the present invention.

[0014] A fifth aspect of the invention provides the use of a reagent for detecting the expression level of the serum biomarker LIPG in the preparation of diagnostic products for diagnosing sleep deprivation; Optionally, the reagent for detecting the expression level of the serum biomarker LIPG is selected from: reagents for detecting the protein expression level of LIPG, reagents for detecting the DNA level of LIPG, or reagents for detecting the RNA level of LIPG; Optionally, the reagent for detecting the protein expression level of LIPG is an antibody that specifically binds to the protein encoded by LIPG, an antibody functional fragment, an agglutinant, a receptor, and / or a conjugated antibody; Optionally, the reagent for detecting the protein expression level of LIPG is an antibody that specifically binds to the protein encoded by LIPG; Optionally, the antibody that specifically binds to the LIPG-encoded protein is the antibody against the LIPG protein contained in the ELISA detection kit (Wuhan Yunclone Technology Co., Ltd., SEA469Hu); Optionally, the reagent for detecting the DNA level of LIPG is a reagent for detecting the DNA expression level, DNA methylation level, and / or DNA phosphorylation level of LIPG; Optionally, the detection of LIPG DNA expression level is performed using a reagent that detects the DNA level via sequencing technology; Optionally, the reagent for detecting the RNA level of LIPG is a reagent for detecting the expression levels of LIPG mRNA, lncRNA, and / or miRNA; Optionally, the reagent for detecting the RNA level of LIPG is a primer that specifically amplifies LIPG and / or a probe that specifically recognizes LIPG; Optionally, the diagnostic product is used to diagnose and differentiate between people with sleep deprivation and healthy people; The term "sleep-deprived population" refers to individuals who, on average, sleep less than 7 hours per night for at least one week. The healthy population refers to people whose average actual sleep time per night is ≥7 hours.

[0015] A sixth aspect of the present invention provides a diagnostic product for diagnosing sleep deprivation, the diagnostic product comprising the reagent for detecting the expression level of the serum biomarker LIPG as described in the fifth aspect of the present invention; Optionally, the diagnostic product is selected from test kits, test chips, or test strips.

[0016] Compared with the prior art, the advantages and beneficial effects of the present invention are as follows: This invention is the first to discover that LIPG is significantly differentially expressed in sleep-deprived subjects and has good diagnostic efficacy for this sub-healthy state, with high accuracy, sensitivity, and specificity. Based on this, this invention pioneered an effective method for predicting sleep deprivation and provides a related prediction system, computer equipment, and computer-readable storage medium. This provides information for early intervention and treatment of sleep deprivation, and offers objective quantitative evidence for clinicians and health managers, thereby enabling early warning and personalized intervention for the risk of sleep deprivation, with broad application prospects. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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 effort.

[0018] Figure 1 This invention provides a schematic flowchart of a method for predicting sleep deprivation. Figure 2 : A schematic diagram of a sleep deprivation prediction system provided in an embodiment of the present invention; Figure 3 : A schematic diagram of a computer device provided in an embodiment of the present invention; Figure 4 In the screening set, the serum biomarker LIPG showed significant differences in expression between healthy controls and sleep-deprived subjects, as shown in the corresponding results graph. Figure 5 : Result graph showing the diagnostic efficacy of serum biomarker LIPG in diagnosing sleep deprivation in the screening set; Figure 6 In the validation set, the serum biomarker LIPG showed significant differential expression between healthy controls and sleep-deprived subjects, as shown in the corresponding results graph. Figure 7 The results graph shows the diagnostic efficacy of the serum biomarker LIPG in diagnosing sleep deprivation in the validation set. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0020] In some of the processes described in the specification, claims, and accompanying drawings of this invention, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be performed in the order they appear herein, or may be performed in parallel. The operation numbers, such as S101, S102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be performed sequentially or in parallel.

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Figure 1 This is a schematic flowchart of a method for predicting sleep deprivation provided by an embodiment of the present invention. Specifically, the method includes the following steps: S101: Obtain protein expression data from the serum sample to be tested; In one embodiment, the serum sample to be tested is derived from a subject.

[0023] In one embodiment, the subject includes both mammals and non-mammals. Examples of mammals include, but are not limited to, any member of the class Mammalia: humans, non-human primates such as chimpanzees and other apes and monkeys; farm animals such as cattle, horses, sheep, goats, and pigs; domesticated animals such as rabbits, dogs, and cats; and laboratory animals, including rodents such as rats, mice, and guinea pigs. Examples of non-mammals include, but are not limited to, birds, fish, or other non-mammals.

[0024] In some embodiments, the subject is a human being.

[0025] In one embodiment, the protein expression data includes the expression level data of the target protein LIPG.

[0026] In one embodiment, the protein expression data is obtained by enzyme-linked immunosorbent assay (ELISA), mass spectrometry, chemiluminescence, immunochromatography, Western blotting, immunohistochemistry, or co-immunoprecipitation.

[0027] In one embodiment, the insufficient sleep refers to the actual sleep time at night being less than 7 hours per day on average and lasting for at least 1 week. For the first time through detailed experiments, the present invention discovers that in terms of the serum LIPG expression level, the state of insufficient sleep is characterized by a significantly lower concentration of LIPG protein than that of healthy people (P<0.0001), and it has stable diagnostic efficacy in both the screening set and the validation set (the AUC value is as high as 0.9670, the sensitivity is 100%, and the specificity is 87.5%). Therefore, insufficient sleep is not only a clinical definition based on sleep duration but also a sub-healthy state that can be objectively quantified and evaluated by the serum LIPG expression level.

[0028] In one embodiment, the acquisition of the protein expression data of the serum sample includes: First, collect 3 - 5 mL of fasting peripheral venous blood from the subject and place it in a sterile EP tube without anticoagulant. Let it stand at room temperature for 30 min until the blood completely coagulates, then centrifuge it at 3000 rpm for 15 min at 4°C. Carefully aspirate the upper light yellow clear serum, avoiding touching the lower blood cells and the middle buffy coat. After sub-packaging, store it in an ultra-low temperature freezer at -80°C for later use, avoiding repeated freezing and thawing. When detecting, take out the frozen serum sample and slowly thaw it at 4°C. Use a protein detection kit supporting enzyme-linked immunosorbent assay (ELISA) for detection. Strictly follow the operation instructions of the kit to complete the steps of adding samples, incubation, washing, color development, and terminating the reaction in sequence. Finally, use an enzyme-labeled instrument to read the optical density (OD) value of each well at a wavelength of 450 nm, and substitute it into the standard curve to calculate the expression level of the corresponding protein in the serum sample, thus completing the acquisition of the protein expression data.

[0029] S102: Extract the expression data of the target protein in the protein expression data, and the target protein is the LIPG protein; In one embodiment, the LIPG protein refers to endothelial lipase protein, which is a secreted glycoprotein encoded by the LIPG gene (Gene ID: 9388). It is mainly expressed on the surface of vascular endothelial cells and can participate in lipid metabolism regulation by catalyzing the hydrolysis of lipoproteins such as high-density lipoprotein (HDL). The LIPG protein in the present invention specifically refers to the soluble form derived from serum. The present invention discovers for the first time that its expression level is negatively correlated with the state of insufficient sleep, that is, the concentration of soluble LIPG protein in the serum of people with insufficient sleep is significantly lower than that of normal healthy people, and it can be used as a specific biomarker to distinguish the sample of subjects with insufficient sleep from the sample of non-insufficient sleep subjects.

[0030] In one embodiment, the expression data of the target protein LIPG refers to a quantitative indicator obtained by the above detection method that can be directly used for subsequent classification prediction. Specifically, it refers to the absolute mass concentration of LIPG protein per unit volume of serum to be tested (usually in ng / mL). After standardization, this data can be directly input into the constructed prediction model: if the value is lower than the built-in judgment threshold of the model, the classification result of "serum sample of sleep-deprived subject" is output; if the value is higher than the threshold, the classification result of "serum sample of non-sleep-deprived subject" is output. No additional complex feature engineering processing is required to support the classification decision.

[0031] In one embodiment, the inventors of this invention collected serum samples from individuals with sleep deprivation and healthy individuals for analysis to verify the expression of the target protein LIPG and its diagnostic efficacy for sleep deprivation. The detailed experimental materials, methods, and results are as follows: 1. Filter set and validation set Screening set: 16 serum samples from control subjects and 16 serum samples from subjects with sleep deprivation; Validation set: 24 serum samples from control subjects and 24 serum samples from subjects with sleep deprivation; Among them, the sleep-deprived subjects are those whose actual sleep time per night is less than 7 hours for at least one week (without a history of genetic diseases, color blindness, or color weakness), while the controls are healthy individuals whose average actual sleep time per night is ≥ 7 hours (without a history of genetic diseases, color blindness, or color weakness).

[0032] In one embodiment, the scale test refers to any one of the Pittsburgh Sleep Quality Index (PSQI), Epworth Sleepiness Scale (ESS), or similar standardized sleep assessment questionnaires. These scales allow researchers to systematically collect information on subjects' subjective sleep duration, sleep quality, sleep onset time, and daytime dysfunction, thereby classifying subjects as sleep-deprived (average actual sleep time <7 hours for at least one week) or healthy controls (average actual sleep time ≥7 hours).

[0033] All samples were obtained with the consent of the organization's ethics committee, and all participants or their legal guardians signed written informed consent forms.

[0034] 2. Experimental Materials ELISA test kit: LIPG in human serum was tested using an ELISA test kit (SEA469Hu) provided by Wuhan Yunclone Technology Co., Ltd.

[0035] 3. ELISA experimental procedure (1) Sample addition: Set up standard wells, sample wells, and blank wells respectively. Set up 7 standard wells and add 100 μL of different concentrations of standard in sequence. Add 100 μL of standard diluent to the blank wells and add 100 μL of sample to the remaining wells. Cover the microplate with a membrane and incubate at 37°C for 1 hour.

[0036] (2) Discard the liquid, shake dry, no need to wash.

[0037] (3) Add 100 μL of detection solution A working solution to each well (prepare before use), cover the microplate with a membrane, and incubate at 37°C for 1 hour.

[0038] (4) Discard the liquid in the wells. Wash each well with 350 μL of washing buffer and soak for 1-2 minutes. Gently tap the plate on absorbent paper to remove all liquid from the wells. Repeat the washing process 3 times. After the last wash, aspirate or pour out the remaining washing buffer, invert the plate onto absorbent paper, and blot away all the liquid remaining in the wells.

[0039] (5) Add 100 μL of detection solution B working solution (prepared before use) to each well, cover the microplate with a membrane, and incubate at 37°C for 30 minutes.

[0040] (6) Discard the liquid in the hole, spin dry, and wash the plate 5 times, using the same method as step (4).

[0041] (7) Add 90 μL of TMB substrate solution to each well, cover the microplate with a membrane, and develop the color at 37°C in the dark (control the reaction time to 10-20 minutes, and do not exceed 30 minutes. When there is a clear gradient of blue in the first 3-4 standard wells and no clear gradient in the last 3-4 wells, the reaction can be stopped).

[0042] (8) Add 50 μL of stop solution to each well to stop the reaction. The blue color will immediately turn yellow. The order of adding the stop solution should be as similar as possible to the order of adding the substrate solution. If the color is uneven, gently shake the ELISA plate to mix the solution evenly.

[0043] (9) After ensuring that there are no water droplets at the bottom of the ELISA plate and no air bubbles in the wells, immediately use an ELISA reader to measure the optical density (OD value) of each well at a wavelength of 450 nm.

[0044] 4. Experimental Results In the screening set, the serum biomarker LIPG showed significant differential expression between healthy controls and sleep-deprived subjects, as shown in the corresponding figure. Figure 4 As shown, the results indicated that the serum biomarker LIPG was significantly underexpressed in the serum of sleep-deprived subjects (P < 0.0001).

[0045] The diagnostic efficacy of the serum biomarker LIPG for diagnosing sleep deprivation in the screening set is shown in the following figure. Figure 5As shown, the results indicate that the serum biomarker LIPG can be used to effectively differentiate between healthy individuals and those with sleep deprivation, and it exhibits good diagnostic efficacy, with an AUC of 0.9648, a sensitivity of 93.75%, and a specificity of 87.5%. The results from the screening set preliminarily suggest that the serum biomarker LIPG can be used for the effective diagnosis of sleep deprivation.

[0046] In the validation set, the serum biomarker LIPG showed significant differences in expression between healthy controls and sleep-deprived subjects, as shown in the corresponding graph. Figure 6 As shown, the results indicated that the serum biomarker LIPG was significantly underexpressed in the serum of sleep-deprived subjects (P < 0.0001), consistent with the results in the screening set.

[0047] In the validation set, the diagnostic efficacy of the serum biomarker LIPG for diagnosing sleep deprivation is shown in the following figure. Figure 7 As shown, the results indicate that the serum biomarker LIPG can be used to effectively differentiate between healthy individuals and those with sleep deprivation, exhibiting good diagnostic efficacy (AUC value of 0.9670, sensitivity of 100%, and specificity of 87.5%). The results in the validation set further confirm that the serum biomarker LIPG can be used effectively in the diagnosis of sleep deprivation.

[0048] S103: Based on the expression data of the target protein, perform classification prediction to obtain the classification result of whether the serum sample to be tested is a serum sample of a sleep-deprived subject; In one embodiment, the classification result is obtained based on a prediction model. The method for constructing the prediction model includes: obtaining the target protein expression data and corresponding clinical features of the training set samples, including sleep-deprived subjects and healthy individuals; extracting the target protein expression data from the training set and inputting it into a machine learning model to construct a prediction model, thereby obtaining the constructed prediction model.

[0049] In one embodiment, as described above, this application collects serum samples from individuals with sleep deprivation and healthy individuals for ELISA analysis to obtain expression data of the target protein LIPG corresponding to these two clinical characteristics. The expression data of the target protein LIPG is extracted and input into a machine learning model to construct a prediction model, thus obtaining the constructed prediction model.

[0050] In one embodiment, if the expression level of the target protein LIPG is below a threshold, the serum sample to be tested is classified as a serum sample from a sleep-deprived subject; if the expression level of the target protein LIPG is above a threshold, the serum sample to be tested is classified as a serum sample from a non-sleep-deprived subject.

[0051] In one embodiment, the machine learning model includes linear regression, logistic regression, random forest, Lasso regression, Ridge regression, linear discriminant analysis, nearest neighbor, decision tree, perceptron, neural network, support vector machine, naive Bayes, AdaBoost, GBDT, XGBoost, LightGBM, and / or CatBoost.

[0052] In one embodiment, the training set includes serum samples from sleep-deprived subjects and serum samples from healthy controls. The samples are processed using a standardized fasting venous blood collection and serum separation procedure. The expression level of LIPG protein is detected using an ELISA kit (Wuhan Yunclone Technology Co., Ltd., SEA469Hu). The obtained expression level data is used as input features for model training after missing value verification and standardization. This is used to fit the machine learning model to learn the correlation between LIPG expression level and sleep deprivation, supporting the subsequent determination of thresholds and construction of classification capabilities for the prediction model.

[0053] In one embodiment, the construction of the prediction model refers to iteratively training and optimizing the parameters of a selected machine learning algorithm (at least one of linear regression, logistic regression, random forest, XGBoost, etc.) using the LIPG protein expression level of the training set samples as input features and the corresponding clinical labels of sleep deprivation / health as supervision signals. First, the labeled dataset is divided into a training subset and an internal validation subset according to a preset ratio (e.g., 7:3). Model hyperparameters are adjusted using methods such as grid search and cross-validation. The optimal model structure is selected using AUC, accuracy, sensitivity, and specificity as core evaluation indicators. Finally, a prediction model that can automatically output classification results based on LIPG expression levels is obtained. The model has a built-in, pre-determined judgment threshold. When the LIPG expression level of the sample to be tested is lower than this threshold, it is judged as a serum sample from a sleep-deprived subject; when it is higher than this threshold, it is judged as a serum sample from a non-sleep-deprived subject. After training, an independent validation set (the remaining samples not involved in training) can be used to verify the model's generalization performance, ensuring that its AUC on both the selection set and validation set is consistently above 0.96, and that its sensitivity and specificity are at a high level, meeting the accuracy requirements for sleep deprivation screening. Those skilled in the art can, based on the training set construction rules, feature variables (serum LIPG protein expression level), label definitions (sleep deficiency / health binary classification) and the range of selectable machine learning models disclosed in this invention, combine known model tuning methods such as grid search and cross-validation to adjust the model hyperparameters for adaptability and verify performance, and complete the conventional construction of the prediction model.

[0054] In one embodiment, given the method for constructing the prediction model described above, the prediction model includes the threshold; that is, given the prediction model, the threshold is also determined. Based on the determined threshold, the classification result of whether the test sample is a serum sample from a sleep-deprived subject can be predicted. The specific judgment result based on the prediction model is as follows: if the expression level of the target protein LIPG is lower than the threshold, the serum sample is classified as a serum sample from a sleep-deprived subject; if the expression level of the target protein LIPG is higher than the threshold, the serum sample is classified as a serum sample from a non-sleep-deprived subject.

[0055] In one embodiment, the effectiveness of the constructed prediction model can also be predicted by taking another dataset containing target protein expression data corresponding to sleep-deprived individuals and healthy individuals, and verifying the effectiveness of the constructed prediction model in this dataset.

[0056] In one embodiment, the protein expression data is obtained by enzyme-linked immunosorbent assay (ELISA), mass spectrometry, chemiluminescence immunoassay, immunochromatography, Western blotting, immunohistochemistry, or co-immunoprecipitation. Any method or technique capable of detecting the protein expression level corresponding to LIPG can be used in this invention.

[0057] Figure 2 This invention provides a schematic diagram of a sleep deprivation prediction system. Specifically, the system includes: The data acquisition unit is used to acquire protein expression data from the serum sample to be tested. The data extraction unit is used to extract the expression data of the target protein from the protein expression data, wherein the target protein is LIPG protein; The prediction result unit is used to perform classification prediction based on the expression data of the target protein to obtain the classification result of whether the serum sample to be tested is a serum sample of a sleep-deprived subject; The classification result is obtained based on a prediction model. The method for constructing the prediction model includes: obtaining the target protein expression data and corresponding clinical features of the training set samples, including sleep-deprived subjects and healthy individuals; extracting the target protein expression data from the training set and inputting it into a machine learning model to construct a prediction model; and obtaining the constructed prediction model. If the expression level of the target protein LIPG is below the threshold, the serum sample to be tested is classified as a serum sample from a sleep-deprived subject; if the expression level of the target protein LIPG is above the threshold, the serum sample to be tested is classified as a serum sample from a non-sleep-deprived subject.

[0058] In one embodiment, the machine learning model includes linear regression, logistic regression, random forest, Lasso regression, Ridge regression, linear discriminant analysis, nearest neighbor, decision tree, perceptron, neural network, support vector machine, naive Bayes, AdaBoost, GBDT, XGBoost, LightGBM, and / or CatBoost.

[0059] In one embodiment, the protein expression data includes the expression level data of the target protein LIPG; The protein expression data are obtained by enzyme-linked immunosorbent assay (ELISA), mass spectrometry, chemiluminescence, immunochromatography, Western blotting, immunohistochemistry, or co-immunoprecipitation. Sleep deprivation is defined as an average of less than 7 hours of actual sleep per night for at least one week.

[0060] Figure 3 The computer device provided in this embodiment of the invention specifically includes a memory and a processor. The memory is used to store program instructions, and the processor is used to call the program instructions. When the program instructions are executed, the sleep deprivation prediction method described above is implemented.

[0061] Furthermore, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the sleep deprivation prediction method as described above.

[0062] Furthermore, embodiments of the present invention also provide the application of reagents for detecting the expression level of the serum biomarker LIPG in the preparation of diagnostic products for diagnosing sleep deprivation.

[0063] In one embodiment, the reagent for detecting the expression level of the serum biomarker LIPG is selected from: reagents for detecting the protein expression level of LIPG, reagents for detecting the DNA level of LIPG, or reagents for detecting the RNA level of LIPG.

[0064] In one embodiment, the reagent for detecting the protein expression level of LIPG is an antibody that specifically binds to the protein encoded by LIPG, an antibody functional fragment, an agglutinant, a receptor, and / or a conjugated antibody.

[0065] In one embodiment, the reagent for detecting the protein expression level of LIPG is an antibody that specifically binds to the protein encoded by LIPG.

[0066] In one embodiment, the antibody that specifically binds to the LIPG-encoded protein is an antibody against the LIPG protein contained in an ELISA detection kit (Wuhan Yunclone Technology Co., Ltd., SEA469Hu).

[0067] In one embodiment, the reagent for detecting the DNA level of LIPG is a reagent for detecting the DNA expression level, DNA methylation level, and / or DNA phosphorylation level of LIPG.

[0068] In one embodiment, the detection of LIPG DNA expression level is performed using a reagent that detects the DNA level via sequencing technology.

[0069] In one embodiment, the reagent for detecting the RNA level of LIPG is a reagent for detecting the expression levels of LIPG mRNA, lncRNA, and / or miRNA.

[0070] In one embodiment, the reagent for detecting the RNA level of LIPG is a primer that specifically amplifies LIPG and / or a probe that specifically recognizes LIPG.

[0071] In one embodiment, the diagnostic product is used to diagnose and differentiate between people with sleep deprivation and healthy people.

[0072] In one embodiment, the sleep-deprived population refers to a group of people who have an average nightly sleep duration of less than 7 hours for at least one week. The healthy population refers to people whose average actual sleep time per night is ≥7 hours.

[0073] Furthermore, embodiments of the present invention also provide a diagnostic product for diagnosing sleep deprivation, the diagnostic product comprising the reagents described above for detecting the expression level of the serum biomarker LIPG.

[0074] In one embodiment, the diagnostic product is selected from test kits, test chips, or test strips.

[0075] The verification results of this verification embodiment show that assigning inherent weights to indications can moderately improve the performance of this method compared to the default settings.

[0076] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0078] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0079] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0080] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0081] Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0082] The computer device provided by the present invention has been described in detail above. For those skilled in the art, there will be changes in the specific implementation and application scope based on the ideas of the embodiments of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for predicting sleep deprivation, characterized in that, The method includes: Obtain protein expression data from the serum sample to be tested; Extract the expression data of the target protein from the protein expression data, wherein the target protein is LIPG protein; Based on the expression data of the target protein, classification prediction is performed to obtain the classification result of whether the serum sample to be tested is a serum sample of a sleep-deprived subject; The classification result is obtained based on a prediction model. The method for constructing the prediction model includes: obtaining the target protein expression data and corresponding clinical features of the training set samples, including sleep-deprived subjects and healthy individuals; extracting the target protein expression data from the training set and inputting it into a machine learning model to construct a prediction model; and obtaining the constructed prediction model. If the expression level of the target protein LIPG is below the threshold, the serum sample to be tested is classified as a serum sample from a sleep-deprived subject; if the expression level of the target protein LIPG is above the threshold, the serum sample to be tested is classified as a serum sample from a non-sleep-deprived subject.

2. The method for predicting sleep deprivation according to claim 1, characterized in that, The machine learning models include linear regression models, logistic regression models, random forest models, Lasso regression models, Ridge regression models, linear discriminant analysis models, nearest neighbor models, decision tree models, perceptron models, neural network models, support vector machine models, Naive Bayes models, AdaBoost models, GBDT models, XGBoost models, LightGBM models, and / or CatBoost models.

3. The method for predicting sleep deprivation according to claim 1, characterized in that, The protein expression data includes the expression level data of the target protein LIPG; The protein expression data are obtained by enzyme-linked immunosorbent assay (ELISA), mass spectrometry, chemiluminescence, immunochromatography, Western blotting, immunohistochemistry, or co-immunoprecipitation. Optionally, the sleep deprivation refers to an average of less than 7 hours of actual sleep per night for at least one week.

4. A sleep deprivation prediction system, characterized in that, The system includes: The data acquisition unit is used to acquire protein expression data from the serum sample to be tested. The data extraction unit is used to extract the expression data of the target protein from the protein expression data, wherein the target protein is LIPG protein; The prediction result unit is used to perform classification prediction based on the expression data of the target protein to obtain the classification result of whether the serum sample to be tested is a serum sample of a sleep-deprived subject; The classification result is obtained based on a prediction model. The method for constructing the prediction model includes: obtaining the target protein expression data and corresponding clinical features of the training set samples, including sleep-deprived subjects and healthy individuals; extracting the target protein expression data from the training set and inputting it into a machine learning model to construct a prediction model; and obtaining the constructed prediction model. If the expression level of the target protein LIPG is below the threshold, the serum sample to be tested is classified as a serum sample from a sleep-deprived subject; if the expression level of the target protein LIPG is above the threshold, the serum sample to be tested is classified as a serum sample from a non-sleep-deprived subject.

5. The sleep deprivation prediction system according to claim 4, characterized in that, The machine learning models include linear regression models, logistic regression models, random forest models, Lasso regression models, Ridge regression models, linear discriminant analysis models, nearest neighbor models, decision tree models, perceptron models, neural network models, support vector machine models, Naive Bayes models, AdaBoost models, GBDT models, XGBoost models, LightGBM models, and / or CatBoost models.

6. The sleep deprivation prediction system according to claim 4, characterized in that, The protein expression data includes the expression level data of the target protein LIPG; The protein expression data are obtained by enzyme-linked immunosorbent assay (ELISA), mass spectrometry, chemiluminescence, immunochromatography, Western blotting, immunohistochemistry, or co-immunoprecipitation. Optionally, the sleep deprivation refers to an average of less than 7 hours of actual sleep per night for at least one week.

7. A computer device, characterized in that, The computer device includes: a memory and a processor, the memory being used to store program instructions; the processor being used to invoke the program instructions, and when the program instructions are executed, to implement the sleep deprivation prediction method according to any one of claims 1-3.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the sleep deprivation prediction method according to any one of claims 1-3.

9. Application of reagents for detecting serum marker LIPG expression levels in the preparation of diagnostic products for diagnosing sleep deprivation; Optionally, the reagent for detecting the expression level of the serum biomarker LIPG is selected from: reagents for detecting the protein expression level of LIPG, reagents for detecting the DNA level of LIPG, or reagents for detecting the RNA level of LIPG; Optionally, the reagent for detecting the protein expression level of LIPG is an antibody that specifically binds to the protein encoded by LIPG, an antibody functional fragment, an agglutinant, a receptor, and / or a conjugated antibody; Optionally, the reagent for detecting the protein expression level of LIPG is an antibody that specifically binds to the protein encoded by LIPG; Optionally, the antibody that specifically binds to the LIPG-encoded protein is the antibody against the LIPG protein contained in the ELISA detection kit (Wuhan Yunclone Technology Co., Ltd., SEA469Hu); Optionally, the reagent for detecting the DNA level of LIPG is a reagent for detecting the DNA expression level, DNA methylation level, and / or DNA phosphorylation level of LIPG; Optionally, the detection of LIPG DNA expression level is performed using a reagent that detects the DNA level via sequencing technology; Optionally, the reagent for detecting the RNA level of LIPG is a reagent for detecting the expression levels of LIPG mRNA, lncRNA, and / or miRNA; Optionally, the reagent for detecting the RNA level of LIPG is a primer that specifically amplifies LIPG and / or a probe that specifically recognizes LIPG; Optionally, the diagnostic product is used to diagnose and differentiate between people with sleep deprivation and healthy people; The term "sleep-deprived population" refers to individuals who, on average, sleep less than 7 hours per night for at least one week. The healthy population refers to people whose average actual sleep time per night is ≥7 hours.

10. A diagnostic product for diagnosing sleep deprivation, characterized in that, The diagnostic product comprises the reagent for detecting the expression level of the serum biomarker LIPG as described in claim 9; Optionally, the diagnostic product is selected from test kits, test chips, or test strips.

Images