Pre-eclampsia biomarker and application thereof

Abstract

The invention provides a biomarker for predicting the preeclampsia risk, application of the biomarker and a method for detecting whether an object has the preeclampsia risk or not. According to the method for detecting whether the object has the preeclampsia risk, preeclampsia risk prediction is carried out on all pregnant women in the early and middle stages of pregnancy, whether preeclampsia high-risk groups exist or not is not distinguished, and risk prediction can be carried out before symptoms occur; different types of free RNA in blood are associated with preeclampsia, the biomarker is used for constructing a preeclampsia prediction model in at least 300 samples, and the specificity of the preeclampsia prediction model during pregnancy can be improved to 85% or above.

Description

Preeclampsia biomarkers and their uses Technical Field

[0001] The present disclosure relates to the field of biomedicine technology. Specifically, the present disclosure relates to preeclampsia biomarkers and uses thereof. Background Art

[0002] Preeclampsia (PE) is a pregnancy-related disorder associated with new-onset hypertension, affecting 3–5% of pregnancies. The latest guidelines define preeclampsia as a systolic blood pressure ≥140 mmHg and / or a diastolic blood pressure ≥90 mmHg after 20 weeks of gestation, accompanied by any of the following: urine protein ≥0.3 g / 24 hours, a urine protein / creatinine ratio ≥0.3, or spot urine protein ≥(+) (when protein quantification is unavailable); or the absence of proteinuria accompanied by any of the following organ or system involvement: abnormalities in vital organs such as the heart, lungs, liver, and kidneys, or in the hematologic, digestive, or nervous systems, or placental-fetal involvement. The FIGO guidelines categorize preeclampsia into four types based on the time of diagnosis and delivery: early-onset preeclampsia, late-onset preeclampsia, and preterm / term preeclampsia.

[0003] Up to now, the exact pathogenesis of preeclampsia is still unclear, and there is no effective treatment. Termination of pregnancy and placental delivery are the only effective treatment options for preeclampsia. Among the measures to prevent preeclampsia, taking aspirin for pregnant women at high risk of preeclampsia is a recognized method (Poon LC, Shennan A, Hyett JA, et al. The International Federation of Gynecology and Obstetrics (FIGO) initiative on pre-eclampsia: A pragmatic guide for first-trimester screening and prevention. Int J Gynaecol Obstet. 2019; 145 Suppl 1: 1-33. Obstetricians ACo and Gynecologists. Gestational hypertension and preeclampsia: ACOG Practice Bulletin, number 222. Obstet Gynecol. 2020; 135: e237-e60). Studies have shown that taking aspirin at ≤16 weeks of gestation can significantly reduce the risk of preeclampsia (Villa PM, Kajantie E, K, et al. Aspirin in the prevention of pre-eclampsia in high-risk women: a randomized placebo-controlled PREDO Trial and a meta-analysis of randomized trials. BJOG: An International Journal of Obstetrics & Gynaecology. 2013;120:64-74). How to predict the onset of pre-eclampsia in pregnant women at high risk is an urgent problem that needs to be solved.

[0004] Some known risk factors for preeclampsia, such as advanced age, family history of preeclampsia, pregnancy interval, and obesity, have been investigated for screening high-risk individuals for preeclampsia. However, due to the heterogeneity and complexity of preeclampsia, the absence of known risk factors does not necessarily mean that preeclampsia will not develop. Predicting high-risk individuals for preeclampsia based on maternal risk factors is inaccurate. Placental cell dysfunction can lead to serious pregnancy complications, but invasive placental tissue sampling poses risks to both the mother and the fetus. Analysis of free circulating RNA in maternal blood can noninvasively identify functional abnormalities in the extravillous trophoblastic layer of the placenta in preeclampsia, potentially facilitating early prediction of preeclampsia before symptoms develop.

[0005] In 2019, ILLUMINA INC published a patent titled "Circulating RNA Markers Specific for Preeclampsia" (Publication No. CN110785499A) that proposes a method for detecting preeclampsia and / or determining an increased risk of preeclampsia in pregnant women. The method involves identifying multiple circulating RNA (C-RNA) molecules in a biological sample obtained from the pregnant woman. This patent analyzes a confirmed population and uses these C-RNAs to construct a model for classification, rather than prediction.

[0006] The patent "Means and methods for excluding the onset of preeclampsia within a certain period using the sFlt-1 / PlGF or endoglin / PlGF ratio" (patent CN104412107B) discloses a method for diagnosing whether a pregnant subject is not at risk of preeclampsia within a short time window. This method predicts preeclampsia based on the sFlt-1 / PlGF or endoglin / PlGF ratio, and can only predict whether there is a risk of preeclampsia within a short period of time, which has certain limitations.

[0007] Therefore, to date, there is no widely recognized biomarker that can be used in clinical practice to predict various types of preeclampsia. Therefore, there is an urgent need to develop a non-invasive, clinically applicable, and highly accurate biomarker that can be applied to all stages of pregnancy and can predict preeclampsia early before disease symptoms occur.

[0008] Summary of the Invention

[0009] The present disclosure aims to solve, at least to some extent, one of the technical problems in the related art. To this end, one purpose of the present disclosure is to propose a biomarker combination for predicting the risk of preeclampsia and a method for detecting whether a subject has the risk of preeclampsia. The method provided by the present disclosure for detecting whether a subject has the risk of preeclampsia predicts the risk of preeclampsia for all pregnant women in the early and middle stages of pregnancy, regardless of whether they are at high risk for preeclampsia, and can predict the risk before symptoms occur; discover the association between different types of free RNA in the blood and preeclampsia, construct a preeclampsia prediction model in at least 300 samples, and improve the specificity of the prediction model for preeclampsia during pregnancy to more than 85%.

[0010] In one aspect of the present disclosure, the present disclosure provides a biomarker combination for predicting the risk of preeclampsia. According to one embodiment of the present disclosure, the biomarker combination includes the following genes:

[0011] ALB, FGA, MIR27A, IGFBP5, MIR376C, MIR215, SERPINA1, MIR106A.

[0012] According to one embodiment of the present disclosure, the biomarker combination is an RNA molecular marker combination;

[0013] Preferably, the biomarker combination is a cfRNA molecular marker combination.

[0014] According to one embodiment of the present disclosure, the biomarker combination includes:

[0015] ALB mRNA transcripts, FGA mRNA transcripts, MIR27A miRNA transcripts, IGFBP5 mRNA transcripts, MIR376C miRNA transcripts, MIR215 miRNA transcripts, SERPINA1 mRNA transcripts, MIR106A miRNA transcripts.

[0016] According to one embodiment of the present disclosure, the biomarker combination further includes at least one of the following genes:

[0017] MIR190A, S100A9, APOA1, MALAT1, CGA, LEP, MIR130A, MIR144, MIR19B1, MIR33A.

[0018] According to one embodiment of the present disclosure, the biomarker combination further includes at least one of the following:

[0019] miRNA transcripts of MIR190A, mRNA transcripts of S100A9, mRNA transcripts of APOA1, lcnRNA transcripts of MALAT1, mRNA transcripts of CGA, miRNA transcripts of MIR19B1, miRNA transcripts of MIR144, miRNA transcripts of MIR33A, miRNA transcripts of MIR130A, and mRNA of LEP.

[0020] According to one embodiment of the present disclosure, the biomarker combination includes:

[0021] MIR190A, ALB, FGA, MIR27A, IGFBP5, MIR376C, S100A9, APOA1, MALAT1, MIR215, CGA, SERPINA1, MIR106A.

[0022] According to a preferred embodiment of the present disclosure, the biomarker combination includes: miRNA transcripts of MIR190A, mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA transcripts of IGFBP5, miRNA transcripts of MIR376C, mRNA transcripts of S100A9, mRNA transcripts of APOA1, lcnRNA transcripts of MALAT1, miRNA transcripts of MIR215, mRNA transcripts of CGA, mRNA transcripts of SERPINA1, and miRNA transcripts of MIR106A.

[0023] According to one embodiment of the present disclosure, the biomarker combination includes:

[0024] MIR19B1, ALB, FGA, MIR27A, IGFBP5, MIR376C, MIR144, MIR33A, MIR130A, MIR215, LEP, SERPINA1, MIR106A.

[0025] According to a preferred embodiment of the present disclosure, the biomarker combination includes: miRNA transcripts of MIR19B1, mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA transcripts of IGFBP5, miRNA transcripts of MIR376C, miRNA transcripts of MIR144, miRNA transcripts of MIR33A, miRNA transcripts of MIR130A, miRNA transcripts of MIR215, mRNA of LEP, mRNA transcripts of SERPINA1, and miRNA transcripts of MIR106A.

[0026] According to one embodiment of the present disclosure, the biomarker combination further comprises: a clinical phenotype;

[0027] Preferably, the clinical phenotype includes:

[0028] Whether to undergo IVF, mean arterial pressure, BMI, maternal age, number of births, and at least one of the following: BMI = weight ÷ height 2 .

[0029] In a second aspect, the present disclosure provides a reagent for detecting the biomarker combination described in the first aspect, wherein the reagent includes a biomolecule that specifically hybridizes with any one of the biomarkers in the biomarker combination or their expression products and / or a detection reagent that uses the biomarker as the detection target.

[0030] According to one embodiment of the present disclosure, the biomolecule includes at least one of a primer, a probe, and an antibody.

[0031] A third aspect of the present disclosure provides a detection product, comprising the reagent described in the second aspect.

[0032] According to one embodiment of the present disclosure, the detection product includes a kit and / or a chip.

[0033] A fourth aspect of the present disclosure provides a method for detecting preeclampsia or predicting the risk of preeclampsia. According to one embodiment of the present disclosure, the method comprises:

[0034] (1) Obtaining the biological sample to be tested;

[0035] (2) determining the detection data of the biomarker combination of the first aspect in the biological sample obtained in step (1);

[0036] (3) identifying whether the patient has preeclampsia or is at risk of preeclampsia based on the detection data described in step (2);

[0037] The detection data of the biomarker combination includes the expression level of the gene and optionally the results of clinical characterization.

[0038] When the biomarker combination is a gene, the detection data is the expression level of the gene. When the biomarker combination also includes clinical manifestations, the detection data also includes the results of the clinical manifestations.

[0039] According to one embodiment of the present disclosure, the biological sample includes one or more of plasma, whole blood, amniotic fluid, serum and urine.

[0040] According to one embodiment of the present disclosure, the biological sample is collected before the 33rd gestational week (including the 33rd gestational week) of the pregnant woman, preferably collected between the 12th and 33rd gestational weeks.

[0041] According to one embodiment of the present disclosure, when the biological sample is a blood sample, the biological sample is derived from a peripheral blood sample of a pregnant woman;

[0042] According to one embodiment of the present disclosure, the identification of whether the patient has preeclampsia or the risk of having preeclampsia is achieved by a prediction model;

[0043] According to one embodiment of the present disclosure, the prediction model is obtained by training a machine learning model using the detection data of the biomarker combination of preeclampsia samples and non-preeclampsia samples as input and the risk score of having the risk of preeclampsia as output; preferably, the threshold value of the risk score is 0.5; preferably, the risk score greater than the threshold value indicates that the pregnant woman has preeclampsia or is at risk of preeclampsia.

[0044] According to one embodiment of the present disclosure, the machine learning model includes: one or more of an average neural network model, a gradient boosting machine, a logistic regression model, a neural network model, and a support vector machine, more preferably an average neural network model or a support vector machine.

[0045] According to one embodiment of the present disclosure, the expression level of the gene is the RNA expression level of the gene.

[0046] According to one embodiment of the present disclosure, the RNA expression level of the biomarker is determined by quantitative analysis of cfRNA in the biological sample; preferably, the quantitative analysis method includes high-throughput sequencing or RT-PCR.

[0047] A fifth aspect of the present disclosure provides a device for detecting preeclampsia or predicting the risk of preeclampsia, comprising:

[0048] Acquisition module: used to obtain biological samples to be tested;

[0049] Detection module: used to determine the detection data of the biomarker combination described in the first aspect in the biological sample;

[0050] Identification module: used to identify whether a patient has preeclampsia or is at risk of developing preeclampsia based on the detection data of the biomarker combination.

[0051] When the biomarker combination is a gene, the detection data is the expression level of the gene. When the biomarker combination also includes clinical manifestations, the detection data also includes the results of the clinical manifestations.

[0052] According to one embodiment of the present disclosure, the biological sample includes one or more of plasma, whole blood, amniotic fluid, serum and urine.

[0053] According to one embodiment of the present disclosure, the biological sample is collected before the 33rd gestational week (including the 33rd gestational week) of the pregnant woman, preferably collected between the 12th and 33rd gestational weeks.

[0054] According to one embodiment of the present disclosure, when the biological sample is a blood sample, the biological sample is derived from a peripheral blood sample of a pregnant woman;

[0055] According to one embodiment of the present disclosure, the identification of whether the patient has preeclampsia or has a risk of having preeclampsia is achieved by a prediction model;

[0056] According to one embodiment of the present disclosure, the prediction model is obtained by training a machine learning model using the detection data of the biomarker combination of preeclampsia samples and non-preeclampsia samples as input and the risk score of having the risk of preeclampsia as output; preferably, the threshold value of the risk score is 0.5; preferably, the risk score greater than the threshold value indicates that the pregnant woman has preeclampsia or is at risk of preeclampsia.

[0057] According to one embodiment of the present disclosure, the machine learning model includes: one or more of an average neural network model, a gradient boosting machine, a logistic regression model, a neural network model, and a support vector machine, more preferably an average neural network model or a support vector machine.

[0058] According to one embodiment of the present disclosure, the expression level of the gene is the RNA expression level of the gene.

[0059] According to one embodiment of the present disclosure, the RNA expression level of the biomarker is determined by quantitative analysis of cfRNA in the biological sample; preferably, the quantitative analysis method includes high-throughput sequencing or RT-PCR.

[0060] A sixth aspect of the present disclosure provides an apparatus for detecting preeclampsia or predicting the risk of preeclampsia, comprising a memory for storing a program;

[0061] A processor, configured to implement the method described in the fourth aspect by executing the program stored in the memory.

[0062] A seventh aspect of the present disclosure provides a computer-readable storage medium, on which a program is stored. The program can be executed by a processor to implement the method described in the fourth aspect.

[0063] An eighth aspect of the present disclosure provides use of biomarkers as targets for screening drugs for treating or preventing preeclampsia, wherein the biomarkers include the biomarker combination described in the first aspect.

[0064] A ninth aspect of the present disclosure provides use of biomarkers in diagnosing or predicting the risk of a subject suffering from preeclampsia, wherein the biomarkers include the biomarker combination described in the first aspect.

[0065] The present disclosure provides a combination and method of markers for detecting or predicting the risk of preeclampsia. The combination and method can predict the risk of preeclampsia for all pregnant women in the first and second trimesters, regardless of whether they are at high risk for preeclampsia. Risk prediction can be performed before symptoms occur.

[0066] Discovered the association between different types of free RNA in the blood and preeclampsia, built a preeclampsia prediction model in at least 300 samples, and improved the specificity of the prediction model for preeclampsia during pregnancy to over 85%;

[0067] Based on the constructed preeclampsia prediction model, the prediction model effect was verified using a completely independent sample of at least 150 cases. The area under the receiver operating characteristic curve (AUC) of the validation set was ≥0.75, and the specificity was >80%.

[0068] Additional aspects and advantages of the present disclosure will be given in part in the description that follows and, in part, will be obvious from the description that follows, or will be learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] The above and / or additional aspects and advantages of the present disclosure will become apparent and readily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0070] FIG1 is a diagram showing the effect of the prediction model of molecular markers on the risk of preeclampsia in Example 1 of the present disclosure, and the SVM model validation in 380 samples;

[0071] FIG2 is an independent validation result of the first group of 262 cases for the evaluation of the effect of the preeclampsia risk prediction model in Example 1 of the present disclosure;

[0072] FIG3 is an independent validation of the second group of 288 cases for the evaluation of the effect of the preeclampsia risk prediction model in Example 1 of the present disclosure;

[0073] FIG4 shows the effect of the prediction model of molecular markers on the risk of preeclampsia in Example 2 of the present disclosure, and the validation of the AvNN model in 430 samples;

[0074] FIG5 is an independent validation result of the first group of 288 cases for the evaluation of the effect of the preeclampsia risk prediction model in Example 2 of the present disclosure;

[0075] FIG6 is an independent validation of the second group of 197 cases for the evaluation of the effect of the preeclampsia risk prediction model in Example 2 of the present disclosure;

[0076] FIG7 shows the effect of the prediction model of molecular markers on the risk of preeclampsia in Example 3 of the present disclosure, and the AvNN model validation in 430 samples;

[0077] FIG8 is an independent validation of the second group of 288 cases for the evaluation of the effect of the preeclampsia risk prediction model in Example 3 of the present disclosure;

[0078] FIG9 is a diagram showing the prediction model effect of molecular markers on the risk of preeclampsia in Comparative Example 1 of the present disclosure, and GBM model verification.

[0079] Detailed Description of the Invention

[0080] The embodiments of the present disclosure are described in detail below. The embodiments described below are exemplary and are only used to explain the present disclosure, and should not be understood as limiting the present disclosure.

[0081] It should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, features defined as "first" or "second" may explicitly or implicitly include one or more of the features. Furthermore, in the description of this disclosure, unless otherwise specified, "plurality" means two or more.

[0082] The endpoints of the ranges and any values ​​disclosed herein are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoints of each range, the endpoints of each range and individual point values, and the individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered to be specifically disclosed herein.

[0083] In order to make the present disclosure more easily understood, certain technical and scientific terms are specifically defined below. Unless otherwise clearly defined elsewhere in this document, all other technical and scientific terms used herein have the meaning commonly understood by those skilled in the art to which the present disclosure belongs.

[0084] In this document, the terms “include” or “comprising” are open expressions, that is, including the contents specified in the present disclosure, but not excluding other contents.

[0085] As used herein, the terms "optionally," "optional," or "optionally" generally mean that the subsequently described event or circumstance may but need not occur, and that the description includes instances where the event or circumstance occurs and instances where it does not.

[0086] The present invention uses plasma free RNA to find risk prediction molecular markers before the onset of preeclampsia, uses a non-invasive method to predict the risk of preeclampsia in advance, and constructs a preeclampsia risk prediction model. The biomarker combination obtained by the above screening can be used to accurately predict the risk of preeclampsia. The inventors used maternal peripheral blood plasma to generate the expression spectrum of plasma free RNA, compared the differential characteristics of the plasma free RNA expression spectrum of pregnant women with preeclampsia and non-preeclampsia in the early and middle stages of pregnancy, screened out preeclampsia risk prediction markers, and constructed a preeclampsia risk prediction model using a machine learning algorithm. Finally, the effect of the constructed preeclampsia prediction model was evaluated using completely independent samples.

[0087] Through the above method, the inventors obtained three groups of biomarker combinations with better risk prediction effects for preeclampsia: MIR190A, ALB, FGA, MIR27A, IGFBP5, MIR376C, S100A9, APOA1, MALAT1, MIR215, CGA, SERPINA1, MIR106A; MIR19B1, ALB, FGA, MIR27A, IGFBP5, MIR376C, MIR144, MIR33A, MIR130A, MIR215, LEP, SERPINA1, MIR106A; MIR130A, MIR144, MIR19B1, MIR215, MIR376C, ALB, FGA, MIR27A, LEP, IGFBP5, MIR106A, SERPINA1. The inventors found that these three groups of biomarker combinations all contain 8 genes: ALB, FGA, MIR27A, IGFBP5, MIR376C, MIR215, SERPINA1, MIR106A (nucleic acid sequences are shown in SEQ ID NO: 1-8). Therefore, these 8 gene combinations can be used as basic markers for detecting preeclampsia or predicting the risk of preeclampsia. These 8 gene combinations are supplemented with at least one of the genes MIR190A, S100A9, APOA1, MALAT1, CGA, LEP, MIR130A, MIR144, MIR19B1, and MIR33A (nucleic acid sequences are shown in SEQ ID NO: 9-18). The obtained gene combinations are used as biomarker combinations for detecting preeclampsia or predicting the risk of preeclampsia with higher accuracy. The 8 gene combinations are basic markers, which work synergistically with the additional markers to improve the accuracy of detecting preeclampsia or predicting the risk of preeclampsia.

[0088] According to a specific embodiment of the present disclosure, the biomarker combination further includes a clinical phenotype. According to a preferred embodiment of the present disclosure, the clinical phenotype includes at least one of: whether to undergo in vitro fertilization, mean arterial pressure, BMI, maternal age, parity, and history of miscarriage. The inventors have discovered that combining the above genotypic biomarker combination with these clinical phenotypes can improve the accuracy of detecting preeclampsia or predicting the risk of preeclampsia.

[0089] According to a preferred embodiment of the present disclosure, the biomarker combination includes:

[0090] (1) The following genes:

[0091] ALB, FGA, MIR27A, IGFBP5, MIR376C, MIR215, SERPINA1, MIR106A;

[0092] (2) At least one of the following genes:

[0093] MIR190A, S100A9, APOA1, MALAT1, CGA, LEP, MIR130A, MIR144, MIR19B1, MIR33A;

[0094] (3) At least one of the following factors: whether to undergo in vitro fertilization, mean arterial pressure, BMI, maternal age, parity, and history of miscarriage.

[0095] According to a specific embodiment of the present disclosure, the biomarker combination includes:

[0096] 1) ALB mRNA transcripts, FGA mRNA transcripts, MIR27A miRNA transcripts, IGFBP5 mRNA transcripts, MIR376C miRNA transcripts, MIR215 miRNA transcripts, SERPINA1 mRNA transcripts, MIR106A miRNA transcripts;

[0097] 2) At least one of the following:

[0098] miRNA transcripts of MIR190A, mRNA transcripts of S100A9, mRNA transcripts of APOA1, lcnRNA transcripts of MALAT1, mRNA transcripts of CGA, miRNA transcripts of MIR19B1, miRNA transcripts of MIR144, miRNA transcripts of MIR33A, miRNA transcripts of MIR130A, and mRNA of LEP;

[0099] 3) At least one of the following: whether to undergo in vitro fertilization, mean arterial pressure, BMI, maternal age, parity, and history of miscarriage.

[0100] It should be noted that the "biomarkers" described in the present disclosure refer to biochemical indicators that can be used to distinguish preeclampsia from non-preeclampsia, including not only gene expression levels but also clinical phenotypes.

[0101] According to a specific embodiment of the present disclosure, the following method is used to screen and obtain molecular markers and establish a preeclampsia prediction model:

[0102] (1) Obtaining free RNA from maternal plasma

[0103] Peripheral blood is obtained from pregnant women and immediately stored at 4°C. Plasma separation is performed within 8 hours. After plasma separation, the plasma is immediately stored at -80°C pending further processing. Trizol LS is added to the plasma at a volume ratio of 1:3 and immediately shaken to mix thoroughly for extraction. cfRNA extraction can be performed using conventional RNA extraction methods in the art; this disclosure does not specifically limit the method for cfRNA extraction.

[0104] (2) Sequencing or RT-PCR of plasma free RNA

[0105] Plasma-free RNA (cfRNA) library construction is employed. This method captures both long and short RNA fragments in plasma, providing more features for prediction. cfRNA sequencing utilizes whole transcriptome sequencing, using next-generation sequencing to sequence RNA from peripheral blood plasma samples from pregnant women with and without preeclampsia. RT-PCR analysis can also be used, including but not limited to, to quantitatively analyze cfRNA expression profiles.

[0106] (3) Quantification of expression profile of plasma free RNA

[0107] The original cfRNA sequencing data was quality controlled, including cutting the adapter, removing low-quality reads, and removing reads <17bp in length. The high-quality sequences after quality control were aligned to the human genome and quantified using the TPM method to obtain the expression profile of plasma free RNA.

[0108] (4) Screening of risk prediction markers for preeclampsia

[0109] Using the expression profile of plasma free RNA, we compared the differential expression characteristics of genes up and down regulated in the plasma free RNA expression profiles of pregnant women with preeclampsia and non-preeclampsia in the early and middle stages of pregnancy, and screened out markers for predicting the risk of preeclampsia.

[0110] (5) Screening of molecular markers for preeclampsia based on machine learning models

[0111] Based on the molecular markers screened, a preeclampsia risk prediction model was constructed in the training set using machine learning models (including average neural network model (AvNN), gradient boosting machine (GBM), logistic regression model (LR), neural network model (nnet), or support vector machine (SVM)). A risk score greater than 0.5 was considered a high risk of preeclampsia, while a risk score less than 0.5 was considered a low risk of preeclampsia. The preeclampsia risk score was automatically calculated by the model. The model was then validated in the validation set, and molecular markers and models with better prediction effects were screened based on the AUC value.

[0112] (6) Construction and validation of models based on markers and clinical characteristics

[0113] According to the feature selection in step (5), the optimal cfRNA molecular marker combination is obtained:

[0114] mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA transcripts of IGFBP5, miRNA transcripts of MIR376C, miRNA transcripts of MIR215, mRNA transcripts of SERPINA1, miRNA transcripts of MIR106A, and at least one of the following:

[0115] miRNA transcripts of MIR190A, mRNA transcripts of S100A9, mRNA transcripts of APOA1, lcnRNA transcripts of MALAT1, mRNA transcripts of CGA, miRNA transcripts of MIR19B1, miRNA transcripts of MIR144, miRNA transcripts of MIR33A, miRNA transcripts of MIR130A, and mRNA of LEP.

[0116] Preferably, the molecular marker combination can be miRNA transcripts of MIR190A, mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA transcripts of IGFBP5, miRNA transcripts of MIR376C, mRNA transcripts of S100A9, mRNA transcripts of APOA1, lcnRNA transcripts of MALAT1, miRNA transcripts of MIR215, mRNA transcripts of CGA, mRNA transcripts of SERPINA1, and miRNA transcripts of MIR106A.

[0117] Preferably, the molecular marker combination can also be miRNA transcripts of MIR19B1, mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA transcripts of IGFBP5, miRNA transcripts of MIR376C, miRNA transcripts of MIR144, miRNA transcripts of MIR33A, miRNA transcripts of MIR130A, miRNA transcripts of MIR215, mRNA of LEP, mRNA transcripts of SERPINA1, and miRNA transcripts of MIR106A.

[0118] Preferably, the molecular marker combination can also be miRNA transcripts of MIR130A, miRNA transcripts of MIR144, miRNA transcripts of MIR19B1, miRNA transcripts of MIR215, miRNA transcripts of MIR376C, mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA of LEP, mRNA transcripts of IGFBP5, miRNA transcripts of MIR106A, and mRNA transcripts of SERPINA1.

[0119] By combining plasma cfRNA molecular markers with clinical phenotypic characteristics including IVF, mean arterial pressure, BMI, maternal age, parity, and history of miscarriage, and with a machine learning model, a preeclampsia risk prediction model was developed that can predict the risk of preeclampsia as early as 12 weeks of gestation and up to 20 weeks in advance, with high accuracy.

[0120] (7) Evaluation of the prediction model for preeclampsia risk based on independent samples

[0121] The model constructed in step (6) is verified by independent samples to evaluate the effectiveness of the model.

[0122] The biomarkers screened and the preeclampsia risk prediction model based on these markers are available for asymptomatic pregnant women, regardless of whether they are at high risk or not. They can predict the risk of preeclampsia before symptoms appear, making them more applicable to a wider population and more clinically applicable. This technology can predict the risk of preeclampsia in advance, starting in early pregnancy (12 weeks) and up to 20 weeks in advance, using a non-invasive method that only requires peripheral blood sampling.

[0123] The molecular markers for predicting preeclampsia screened and obtained in the present disclosure are mRNA, miRNA and lncRNA molecules in plasma, revealing the value of different types of RNA molecules in plasma in predicting preeclampsia.

[0124] Cell-free RNA in plasma during pregnancy is derived from various tissues throughout the body, providing a means of monitoring tissue health, including reproductive tissues such as the placenta. By comparing expression levels of cell-free RNA between preeclampsia and non-preeclampsia groups during the first and second trimesters (before disease diagnosis), combined with machine learning algorithms, molecular markers predicting preeclampsia risk can be screened. Models can then be constructed to predict preeclampsia risk, and the effectiveness of the constructed preeclampsia prediction model can be evaluated using completely independent samples.

[0125] The scheme of the present disclosure will be explained below in conjunction with the examples. Those skilled in the art will understand that the following examples are only used to illustrate the present disclosure and should not be considered to limit the scope of the present disclosure. Where specific techniques or conditions are not specified in the examples, they are carried out according to the techniques or conditions described in the literature in this area or according to the product instructions. Where the manufacturer of the reagents or instruments is not specified, they are all conventional products that can be obtained commercially. Example

[0126] Example 1

[0127] (1) Maternal plasma acquisition

[0128] Peripheral blood samples were obtained from 380 singleton pregnancies (50 with preeclampsia and 330 without preeclampsia) at the hospital. The mean gestational age at blood collection was 16.9 weeks (Table 1). The earliest gestational age for blood collection in the preeclampsia group was 12 weeks, and the maximum gestational age difference between blood collection and diagnosis of preeclampsia was 20 weeks. All blood samples were immediately stored at 4°C and plasma separated within 8 hours. Plasma separation was performed using a two-step centrifugation process: centrifugation at 1,600 g for 10 minutes at 4°C and then at 12,000 g for 10 minutes. After plasma separation, the plasma was immediately stored at −80°C pending further processing.

[0129] Table 1. Demographic and clinical characteristics of the training and validation set samples

[0130]

[0131] Note: Mean arterial pressure = [(systolic pressure - diastolic pressure) / 3 + diastolic pressure] / median

[0132] (2) Extraction of cfRNA

[0133] Trizol LS was added to the plasma (plasma:Trizol LS volume ratio = 1:3) and immediately shaken to mix for RNA extraction.

[0134] (3) cfRNA sequencing

[0135] Using PALM-Seq, a library-based approach for extracellular RNA, plasma samples from patients with and without preeclampsia were sequenced. This method can simultaneously capture both long and short RNA fragments in plasma, providing more features for prediction.

[0136] (4) Quantification of cfRNA expression profile

[0137] The raw cfRNA sequencing data was quality controlled, including cutting the adapters and removing low-quality reads using trimmomatic software, including reads <17bp in length. The high-quality sequences after quality control were aligned to the human genome (GRCh38.p13) and the expression profile of cfRNA was quantified using the TPM method. The formula is as follows:

[0138] TPM=(Ni / Li)*1000000 / (sum(N1 / L1+N2 / L2+N3 / L3+…+Nn / Ln))

[0139] Ni is the number of reads aligned to the i-th gene; Li is the length of the i-th gene; sum(N1 / L1+N2 / L2+...+Nn / Ln) is the sum of the values ​​of all (n) genes after normalization by length.

[0140] (5) Molecular marker screening

[0141] Using the expression profile of plasma free RNA, we compared the up- and down-regulated differential expression characteristics of genes in the cfRNA expression profiles of pregnant women with preeclampsia and those without preeclampsia, and screened out markers for predicting the risk of preeclampsia.

[0142] First, all groups were randomly split into a training set and a validation set in a 7:3 ratio. The training set contained 35 preeclampsia samples and 231 non-preeclampsia samples, and the validation set contained 15 preeclampsia samples and 99 non-preeclampsia samples (Figure 1). Candidate molecular marker screening was completed in the training set, and the validation set was used to test the predictive performance of the molecular markers and model. First, molecular markers were preliminarily screened by comparing the expression profiles of the preeclampsia and non-preeclampsia groups. This step was implemented using the DESeq2 package (R software package). For each gene, the difference and stability of the average expression level between the two groups were considered in this step (the absolute value of the average expression level difference was greater than 1.5, the p-value was less than 0.05, and the corrected p-value was less than 0.1). Finally, a generalized linear model was used to screen based on feature importance, and molecules with higher frequency were selected as candidate molecular markers.

[0143] (6) Construction and validation of marker-based models

[0144] The candidate molecular markers and their combinations selected in step (5) are used to construct a model. First, in the training set, based on the selected candidate molecular markers, a support vector machine (SVM) machine learning algorithm is used to assess the risk of preeclampsia. The algorithm uses a ten-fold cross-validation method to select the optimal parameters for predictive model construction. By comparing the results of the model training set and the validation set, RNA molecular combinations with an AUC (Area under the receiver operating characteristic curve) of less than 0.9 in the training set and validation set are removed to obtain an optimized RNA molecular marker combination. The model effect is then verified in the validation set, and the model is sorted according to the size of the AUC value to screen out the model with better prediction effect and the corresponding molecular marker combination.

[0145] (7) Construction of models based on markers and clinical characteristics

[0146] According to step (6), the model with the largest AUC value was selected to obtain the best set of 13 cfRNA molecular markers (Table 5). Using these 13 cfRNA molecular markers combined with the two clinical characteristics of whether to undergo in vitro fertilization and mean arterial pressure, a preeclampsia prediction model was constructed using the SVM model in 380 samples (50 pregnant women with preeclampsia and 330 pregnant women without preeclampsia) (the modeling method was the same as step (6)). The obtained model had an AUC of 0.901 for the training set and an AUC of 0.933 for the validation set (Figure 1), with a specificity of 99% (Table 2).

[0147] (8) Evaluation of the effectiveness of molecular markers in predicting the risk of preeclampsia

[0148] The SVM model preeclampsia prediction model constructed in step (7) was further independently verified using two completely independent groups of samples to evaluate the model effect.

[0149] The model was validated in the first group of 262 independent samples (28 pregnant women with preeclampsia and 234 pregnant women without preeclampsia). The AUC of the independent validation set reached 0.88 (Figure 2), with a specificity of 98.7%, a negative predictive value (NPV) of 92.04%, and a positive predictive value (PPV) of 72.73% (Table 3), which is higher than the existing technology level (PPV = 32%, RNA profiles reveal signatures of future health and disease in pregnancy. Nature (2022)).

[0150] The model was validated in a second group of 288 independent samples (54 pregnant women with preeclampsia and 234 pregnant women without preeclampsia). The AUC of the independent validation set reached 0.848 (Figure 3), with a specificity of 99.14%, a negative predictive value (NPV) of 83.76%, and a positive predictive value (PPV) of 81.82% (Table 4), all of which are higher than the existing technical level (PPV = 32%).

[0151] Table 2. AUC and specificity of the SVM model test set and validation set

[0152]

[0153] Table 3. Evaluation of the effectiveness of the preeclampsia risk prediction model (independent validation of 262 cases in the first group)

[0154]

[0155] Table 4. Evaluation of the effectiveness of the preeclampsia risk prediction model (independent validation of the second group of 288 cases)

[0156]

[0157] Table 5. Gene and transcript information of 13 preeclampsia characteristic genes

[0158] Gene Name Gene ID Gene Sequence Length (bp) RNA Type Sequence Number MIR190A40696585 miRNA SEQ ID NO: 9 ALB21317196 mRNA SEQ ID NO: 1 FGA22437617 mRNA SEQ ID NO: 2 MIR27A40701878 miRNA SEQ ID NO: 3 IGFBP5348823445 mRNA SEQ ID NO: 4 MIR376C44291366 miRNA SEQ ID NO: 5 S100A962803170 mRNA SEQ ID NO: 10 APOA13352200 mRNA SEQ ID NO: 11 MALAT13789388779 lncRNA SEQ ID NO: 12 MIR215406997110 miRNA SEQ ID NO: 6 CGA10819606 mRNA SEQ ID NO:13SERPINA1526513889mRNASEQ ID NO:7MIR106A40689981miRNASEQ ID NO:8

[0159] The gene ID in the table is the gene ID number displayed in the NCBI database

[0160] Example 2

[0161] (1) Maternal plasma acquisition

[0162] Peripheral blood samples were obtained from 430 singleton pregnancies, including 100 with preeclampsia and 330 without preeclampsia. The mean gestational age at blood collection was 16.9 weeks (Table 6). The earliest gestational age for blood collection in the preeclampsia group was 12 weeks, and the maximum gestational age difference between blood collection and diagnosis of preeclampsia was 20 weeks. All blood samples were immediately stored at 4°C and plasma separated within 8 hours. Plasma separation was performed using a two-step centrifugation process: centrifugation at 1,600 g for 10 minutes at 4°C and then at 12,000 g for 10 minutes. After plasma separation, the plasma was immediately stored at −80°C pending further processing.

[0163] Table 6. Demographic and clinical characteristics of training and validation set samples

[0164]

[0165]

[0166] Note: Mean arterial pressure = [(systolic pressure - diastolic pressure) / 3 + diastolic pressure] / median

[0167] (2) Extraction of cfRNA

[0168] Trizol LS was added to the plasma (plasma:Trizol LS volume ratio = 1:3) and immediately shaken to mix for RNA extraction.

[0169] (3) cfRNA sequencing

[0170] Using PALM-Seq, a library-based approach for extracellular RNA, plasma samples from patients with and without preeclampsia were sequenced. This method can simultaneously capture both long and short RNA fragments in plasma, providing more features for prediction.

[0171] (4) Quantification of cfRNA expression profile

[0172] The raw cfRNA sequencing data was quality controlled, including cutting the adapters and removing low-quality reads using trimmomatic software, including reads <17bp in length. The high-quality sequences after quality control were aligned to the human genome (GRCh38.p13) and the expression profile of cfRNA was quantified using the TPM method. The formula is as follows:

[0173] TPM=(Ni / Li)*1000000 / (sum(N1 / L1+N2 / L2+N3 / L3+…+Nn / Ln))

[0174] Ni is the number of reads aligned to the i-th gene; Li is the length of the i-th gene; sum(N1 / L1+N2 / L2+...+Nn / Ln) is the sum of the values ​​of all (n) genes after normalization by length.

[0175] (5) Molecular marker screening

[0176] Using the expression profile of plasma free RNA, we compared the up- and down-regulated differential expression characteristics of genes in the cfRNA expression profiles of pregnant women with preeclampsia and those without preeclampsia, and screened out markers for predicting the risk of preeclampsia.

[0177] First, all groups were randomly split into a training set and a validation set in a 7:3 ratio. The training set contained 70 preeclampsia samples and 231 non-preeclampsia samples, and the validation set contained 30 preeclampsia samples and 99 non-preeclampsia samples (Figure 4). All molecular marker screening was completed in the training set, and the validation set was used to test the predictive effect of the molecular markers and the model. First, candidate molecular markers were preliminarily screened by comparing the expression profile differences between the preeclampsia and non-preeclampsia groups. This step was implemented using the DESeq2 package (R software package). For each gene, the difference and stability of the average expression level between the two groups were considered in this step (the absolute value of the average expression level difference was greater than 1.5, the p-value was less than 0.05, and the corrected p-value was less than 0.1). Finally, the generalized linear model was screened based on feature importance, and molecules with higher occurrence frequency were selected as candidate molecular markers.

[0178] (6) Construction and validation of marker-based models

[0179] Models were constructed based on the selected molecular markers. First, in the training set, the average neural network (AvNN) machine learning algorithm was used to assess the risk of preeclampsia based on the selected molecular markers. The algorithm used a ten-fold cross-validation approach to select the optimal parameters for predictive model construction. The model was trained using the expression levels of the molecular markers as input and the maternal preeclampsia risk score as output. A risk score greater than or equal to 0.5 was considered high risk, while a risk score less than 0.5 was considered low risk. The results of the model training and validation sets were compared. RNA combinations with an area under the receiver operating characteristic curve (AUC) less than 0.9 in both the training and validation sets were removed to obtain the optimized RNA marker combination. The model was then validated in the validation set. Models were ranked according to their AUC values ​​to identify the models with the best predictive performance and the corresponding marker combinations.

[0180] (7) Construction and validation of models based on markers and clinical characteristics

[0181] According to step (6), the model with the largest AUC value was selected to obtain the best combination of 13 cfRNA molecular markers (Table 10). Using these 13 cfRNA molecular markers combined with two clinical characteristics, whether to undergo in vitro fertilization and mean arterial pressure, a preeclampsia prediction model was constructed using the AvNN model in 430 samples (100 pregnant women with preeclampsia and 330 pregnant women without preeclampsia) (the modeling method was the same as step (6)). The model training set AUC = 91.8%, the validation set AUC = 90.9% (Figure 4), and the specificity was 92.9% (Table 7).

[0182] (8) Evaluation of the prediction model for preeclampsia risk based on independent samples

[0183] The AvNN preeclampsia prediction model constructed in step (7) was further independently validated using two completely independent groups of samples to evaluate the model effect.

[0184] The model was first validated in a first group of 288 independent samples (54 pregnant women with preeclampsia and 234 pregnant women without preeclampsia). The AUC of the independent validation set reached 0.828 (Figure 5), with a specificity of 88%, a negative predictive value (NPV) of 90.4%, and a positive predictive value (PPV) of 53.3% (Table 8), which is higher than the existing technical level (PPV = 32%).

[0185] The model was then validated in a second set of 197 independent samples (46 pregnant women with preeclampsia and 151 pregnant women without preeclampsia). The independent validation set achieved an AUC of 0.809 (Figure 6), a specificity of 92.5%, a negative predictive value (NPV) of 87.4%, and a positive predictive value (PPV) of 68.4% (Table 9), which is higher than the state of the art (PPV = 32%). Table 7. AUC and specificity of the AvNN model on the test and validation sets.

[0186]

[0187] Table 8. Evaluation of the effect of the risk prediction model for preeclampsia (independent validation of 288 cases in the first group)

[0188]

[0189] Table 9. Evaluation of the effect of the risk prediction model for preeclampsia (independent validation of the second group of 197 cases)

[0190]

[0191] Table 10. Gene and transcript information of 13 preeclampsia characteristic genes

[0192] Gene Name Gene ID Gene Sequence Length (bp) Transcript Type Sequence Number MIR19B140698087 miRNA SEQ ID NO: 17 ALB21317196 mRNA SEQ ID NO: 1 FGA22437617 mRNA SEQ ID NO: 2 MIR27A40701878 miRNA SEQ ID NO: 3 IGFBP5348823445 mRNA SEQ ID NO: 4 MIR376C44291366 miRNA SEQ ID NO: 5 MIR14440693686 miRNA SEQ ID NO: 16 MIR33A40703969 miRNA SEQ ID NO: 18 MIR130A40691989 miRNA SEQ ID NO: 15 MIR215406997110 miRNA SEQ ID NO: 6

[0193] LEP395216352mRNASEQ ID NO:14SERPINA1526513889mRNASEQ ID NO:7MIR106A40689981miRNASEQ ID NO:8

[0194] Example 3

[0195] This example evaluates the effect of the biomarker prediction model on the risk of preeclampsia according to the method in Example 2, with the only difference being the different molecular markers and clinical manifestations. Specifically, the clinical manifestations of this example are: whether to undergo in vitro fertilization, mean arterial pressure, BMI, maternal age, parity, and history of miscarriage. The 12 molecular marker combinations in this example are: MIR130A, MIR144, MIR19B1, MIR215, MIR376C, ALB, FGA, MIR27A, LEP, IGFBP5, MIR106A, and SERPINA1. The 12 molecular markers and 6 clinical manifestations were modeled using the AvNN model.

[0196] The model training set AUC = 94.3%, the validation set AUC = 90.6% (Figure 7), and the specificity was 90% (Table 11).

[0197] The AvNN preeclampsia prediction model constructed above was further independently validated using a completely independent sample set to evaluate the model's effectiveness. The model was validated in a group of 288 independent samples (54 pregnant women with preeclampsia and 234 pregnant women without preeclampsia).

[0198] The AUC of the independent validation set can reach 0.837 (Figure 8), with a specificity of 89.7%, a negative predictive value (NPV) of 90%, and a positive predictive value (PPV) of 53.9% (Table 12), which are higher than the existing technical level.

[0199] Table 11 AUC and specificity of the AvNN model test set and validation set

[0200]

[0201] Table 12 Evaluation of the effectiveness of the preeclampsia risk prediction model (independent validation of a group of 288 cases)

[0202]

[0203] Comparative Example 1

[0204] In this example, the candidate molecular markers selected and obtained in step (5) of Example 1 were combined, and the GBM model was used to build a model according to the method of step (6). The molecular marker combination was MIR130A, MIR144, MIR19B1, MIR215, and MIR376C.

[0205] The model training set AUC = 92%, the validation set AUC = 83.9% (Figure 9), and the specificity was 96% (Table 13).

[0206] Table 13. Evaluation of the effectiveness of the risk prediction model for preeclampsia

[0207] Disease Control Test Set AUC Validation Set AUC Validation Set Specificity (%) Test Set 352310.92 -- Validation Set 1599 -0.83996%

[0208] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0209] Although the embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are illustrative and are not to be construed as limitations on the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.

Claims

1. A biomarker combination for detecting preeclampsia or predicting the risk of preeclampsia, wherein: These genes include: ALB, FGA, MIR27A, IGFBP5, MIR376C, MIR215, SERPINA1, MIR106A.

2. The biomarker combination according to claim 1, wherein: The biomarker combination is an RNA molecular marker combination; Preferably, the biomarker combination is an extracellular free RNA molecular marker combination.

3. The biomarker combination according to claim 1 or 2, wherein: The biomarker combination includes: ALB mRNA transcripts, FGA mRNA transcripts, MIR27A miRNA transcripts, IGFBP5 mRNA transcripts, MIR376C miRNA transcripts, MIR215 miRNA transcripts, SERPINA1 mRNA transcripts, MIR106A miRNA transcripts.

4. The biomarker combination according to any one of claims 1 to 3, wherein: The biomarker panel further includes at least one of the following genes: MIR190A, S100A9, APOA1, MALAT1, CGA, LEP, MIR130A, MIR144, MIR19B1, MIR33A; Preferably, the biomarker combination further includes at least one of the following: miRNA transcripts of MIR190A, mRNA transcripts of S100A9, mRNA transcripts of APOA1, lcnRNA transcripts of MALAT1, mRNA transcripts of CGA, miRNA transcripts of MIR19B1, miRNA transcripts of MIR144, miRNA transcripts of MIR33A, miRNA transcripts of MIR130A, and mRNA of LEP.

5. The biomarker combination according to any one of claims 1 to 4, wherein: The biomarker combination includes: MIR190A, ALB, FGA, MIR27A, IGFBP5, MIR376C, S100A9, APOA1, MALAT1, MIR215, CGA, SERPINA1, MIR106A; Preferably, the biomarker combination includes: miRNA transcripts of MIR190A, mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA transcripts of IGFBP5, miRNA transcripts of MIR376C, mRNA transcripts of S100A9, mRNA transcripts of APOA1, lcnRNA transcripts of MALAT1, miRNA transcripts of MIR215, mRNA transcripts of CGA, mRNA transcripts of SERPINA1, and miRNA transcripts of MIR106A.

6. The biomarker combination according to any one of claims 1 to 4, wherein: The biomarker combination includes: MIR19B1, ALB, FGA, MIR27A, IGFBP5, MIR376C, MIR144, MIR33A, MIR130A, MIR215, LEP, SERPINA1, MIR106A; Preferably, the biomarker combination includes: miRNA transcripts of MIR19B1, mRNA transcripts of ALB, mRNA transcripts of FGA, miRNA transcripts of MIR27A, mRNA transcripts of IGFBP5, miRNA transcripts of MIR376C, miRNA transcripts of MIR144, miRNA transcripts of MIR33A, miRNA transcripts of MIR130A, miRNA transcripts of MIR215, mRNA of LEP, mRNA transcripts of SERPINA1, and miRNA transcripts of MIR106A.

7. The biomarker combination according to any one of claims 1 to 6, wherein: The biomarker combination also includes: clinical phenotype; Preferably, the clinical phenotype includes: Whether to undergo in vitro fertilization, mean arterial pressure, BMI, maternal age, parity, and at least one of the following:

8. A reagent for detecting the biomarker combination according to any one of claims 1 to 6, wherein The reagents include biomolecules that specifically hybridize with any one of the biomarkers in the biomarker combination or their expression products and / or detection reagents that use the biomarker combination as a detection target.

9. The reagent according to claim 8, wherein The biomolecule includes at least one of a primer, a probe, and an antibody.

10. A detection product, wherein: Comprising the reagent according to claim 8 or 9.

11. The detection product according to claim 10, wherein: The detection product includes a kit and / or a chip.

12. A method for detecting preeclampsia or predicting the risk of preeclampsia, wherein: The method comprises: (1) Obtaining the biological sample to be tested; (2) determining the detection data of the biomarker combination according to any one of claims 1 to 7 in the biological sample obtained in step (1); (3) identifying whether the patient has preeclampsia or is at risk of preeclampsia based on the detection data described in step (2); The detection data includes the expression level of the gene and optionally the results of clinical characterization.

13. The method according to claim 12, wherein: The biological sample includes one or more of plasma, whole blood, amniotic fluid, serum and urine; Preferably, the biological sample is collected from the pregnant woman before the 33rd gestational week; Preferably, the biological sample is derived from a peripheral blood sample of a pregnant woman.

14. The method according to claim 12 or 13, wherein: The identification of whether the patient has preeclampsia or has a risk of preeclampsia is achieved by a prediction model; Preferably, the prediction model is obtained by training a machine learning model using the detection data of the biomarkers of preeclampsia samples and non-preeclampsia samples as input and the risk score of having the risk of preeclampsia as output; Preferably, the machine learning model includes: one or more of an average neural network model, a gradient boosting machine, a logistic regression model, a neural network model, and a support vector machine; Preferably, the expression level of the gene is the RNA expression level of the gene.

15. A device for detecting preeclampsia or predicting the risk of preeclampsia, wherein: include: Acquisition module: used to obtain biological samples to be tested; Detection module: used to determine the detection data of the biomarker combination according to any one of claims 1 to 7 in the biological sample; Identification module: used to identify whether a patient has preeclampsia or is at risk of developing preeclampsia based on the data of the biomarker combination; The detection data of the biomarker combination includes the expression level of the gene and optionally the results of clinical characterization.

16. A device for detecting preeclampsia or predicting the risk of preeclampsia, wherein: including a memory for storing programs; A processor, configured to implement the method according to any one of claims 12 to 14 by executing the program stored in the memory. 17 . A computer-readable storage medium, wherein a program is stored on the medium, and wherein the program can be executed by a processor to implement the method according to claim 12 .

18. Use of a biomarker as a target for screening drugs for treating or preventing preeclampsia, wherein the biomarker comprises the biomarker combination according to any one of claims 1 to 6.

19. Use of a biomarker in detecting preeclampsia or predicting the risk of preeclampsia, wherein the biomarker comprises the biomarker combination according to any one of claims 1 to 7.

Images