Prediction method for assisting reproductive development outcome and electronic device

Abstract

The invention provides an auxiliary reproductive development outcome prediction method and an electronic device. The RNA marker disclosed by the invention can be applied to the preparation of related products for predicting and assisting the reproductive development outcome, and the RNA marker comprises one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG0000023633, ENSG00000247311 or LURAP1L-AS1. The method can solve the problem that in the prior art, it is difficult to effectively predict the development outcome of assisted reproductive embryos before treatment, and is suitable for the technical field of prediction of the development outcome of reproductive embryos.

Description

Technical Field

[0001] This invention relates to the field of predicting the developmental outcome of reproductive embryos, and more specifically, to a method and electronic device for predicting the developmental outcome of assisted reproductive technology. Background Technology

[0002] Approximately 9% of the global population of reproductive age suffers from infertility, making it the third leading cause of infertility after cancer and cardiovascular disease. Male factors are a significant contributing factor. Male infertility negatively impacts the psychological well-being and quality of life of couples and their families; therefore, early diagnosis and treatment are crucial to finding the most suitable solutions to improve the chances of conception.

[0003] Assisted reproductive technology (ART) is a routine treatment for infertility; however, its success rate is currently very low. Many couples experience abnormal embryo development in vitro during treatment cycles, leading to treatment failure due to the inability to obtain embryos for implantation. Clinically, some couples only choose donor sperm after multiple attempts and eventually achieve pregnancy. The repeated failures place a heavy physical, psychological, and financial burden on patients.

[0004] In existing technologies, specific proteins in serum and follicular fluid, specific RNAs in sperm, and multi-omics characteristics in embryo culture medium are often used to predict embryonic development outcomes. However, when these technologies predict embryonic development outcomes, patients have already entered the treatment cycle, resulting in insufficient prediction time windows and imposing corresponding physical, mental, and economic burdens on patients.

[0005] Therefore, current technologies still lack a solution that can effectively predict the developmental outcome of assisted reproductive embryos before treatment. Summary of the Invention

[0006] The main objective of this invention is to provide a method and electronic device for predicting the developmental outcome of assisted reproductive technology, in order to solve the problem in the prior art that it is difficult to effectively predict the developmental outcome of assisted reproductive embryos before treatment.

[0007] To achieve the above objectives, according to a first aspect of the present invention, an application of an RNA biomarker in the preparation of products related to predicting developmental outcomes in assisted reproductive technologies is provided, the RNA biomarker comprising one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

[0008] Furthermore, the developmental outcome includes the probability of successful development of the assisted reproductive embryo and / or the probability of successful clinical pregnancy of the reproductive embryo; preferably, the related products include kits, chips and / or electronic devices.

[0009] To achieve the above objectives, according to a second aspect of the present invention, a kit for predicting developmental outcomes in assisted reproductive technology is provided, the kit comprising a detection reagent for detecting the expression levels of RNA biomarkers; the RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

[0010] Furthermore, the kit includes probes and / or primers for detecting RNA markers.

[0011] According to a third aspect of the present invention, a method for predicting the developmental outcome of assisted reproductive technology is provided. The method includes: acquiring a first feature dataset of cell-free RNA from a seminal plasma sample of a test subject, the first feature dataset including expression levels of RNA markers; inputting the first feature dataset into a prediction model for the developmental outcome of assisted reproductive embryos to obtain a prediction score, the prediction score being used to indicate the success rate of assisted reproductive embryo development in the test subject; the RNA markers include one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

[0012] Furthermore, the seminal plasma sample of the subject is obtained before the subject undergoes assisted reproduction.

[0013] Furthermore, the method for obtaining the first feature dataset includes: performing quantitative analysis on free RNA to obtain expression level data of RNA markers, thereby obtaining the first feature dataset.

[0014] According to a fourth aspect of the present invention, an electronic device for predicting the developmental outcome of assisted reproductive technology is provided. The electronic device includes: an acquisition unit for acquiring a first feature dataset of cell-free RNA from a seminal plasma sample of a test subject, the first feature dataset including expression level data of RNA markers; and a prediction unit for inputting the first feature dataset into a prediction model for the developmental outcome of assisted reproductive embryos and outputting a prediction score, the prediction score indicating the success rate of assisted reproductive embryo development in the test subject; the RNA markers include one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

[0015] According to a fifth aspect of the present invention, a method for screening RNA biomarkers for predicting assisted reproductive outcomes is provided, the method comprising:

[0016] I) Obtain the expression levels of all free RNAs in the training set samples to obtain the second feature dataset, with the expression level of each free RNA serving as a candidate feature; the training set samples include seminal plasma samples from successfully developed in vitro embryos and seminal plasma samples from failed in vitro embryos;

[0017] II) Input each candidate feature into Formula I as shown below, and perform the chi-square test one by one to calculate the P-value of each candidate feature; Formula I represents the formula of the prediction model of assisted reproductive development outcome constructed with different candidate features.

[0018]

[0019] Where e represents the exponential score; b represents the bias of the prediction model; f1, f2, ..., f n Represents the values ​​of candidate features in the prediction model; β1, β2, ..., β n y' represents the feature vector corresponding to the value of the candidate feature in the prediction model; y' represents the linear combination part of the prediction model, which represents the sum of the product of all candidate features and all feature vectors.

[0020] III) Screen candidate features whose P-values ​​meet the preset threshold as RNA biomarkers for predicting the developmental outcome of assisted reproduction; RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311 or LURAP1L-AS1.

[0021] Furthermore, the chi-square test includes the likelihood ratio chi-square test; preferably, the method for obtaining the expression level of free RNA includes: performing quantitative analysis on free RNA to obtain expression level data of RNA markers.

[0022] According to a sixth aspect of the present invention, a method for constructing a predictive model for the developmental outcome of assisted reproductive technology is provided. The method includes: S1) detecting data on the expression levels of RNA markers in training set samples; the training set samples include seminal plasma samples of successfully developed in vitro embryos and seminal plasma samples of failed in vitro embryos; S2) constructing a predictive model for the developmental outcome of assisted reproductive embryos based on the data on the expression levels of RNA markers and a machine learning algorithm; the RNA markers include one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311 or LURAP1L-AS1.

[0023] According to a seventh aspect of the present invention, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the above-described method for predicting the developmental outcome of assisted reproductive technology, the above-described method for screening RNA markers for predicting the developmental outcome of assisted reproductive technology, or the above-described method for constructing a predictive model for the developmental outcome of assisted reproductive technology via executing the executable instructions.

[0024] According to an eighth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, the device on which the computer-readable storage medium is located executes the above-described method for predicting the developmental outcome of assisted reproduction, the above-described method for screening RNA markers for predicting the developmental outcome of assisted reproduction, or the above-described method for constructing a predictive model for the developmental outcome of assisted reproduction.

[0025] By applying the technical solution of this invention, and utilizing any one or more RNA markers in this application, including one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1, a product for predicting assisted reproductive outcomes can be prepared. This product can predict the in vitro embryo development outcome before the patient enters the assisted reproductive treatment cycle, determine whether there is a transplantable embryo, provide timely reference and basis for the patient to choose the donor sperm treatment plan, rationally select the treatment plan to improve the treatment outcome, and promptly reduce the burden on the patient. Attached Figure Description

[0026] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0027] Figure 1 A hardware block diagram of an electronic device (or mobile device) for a method of predicting developmental outcomes in assisted reproduction according to an embodiment of the present invention is shown.

[0028] Figure 2 A schematic diagram illustrating the contribution of RNA biomarkers to the prediction of embryonic development outcomes according to Embodiment 1 of the present invention is shown.

[0029] Figure 3 A schematic diagram of the ROC curve of a predictive model for the developmental outcome of assisted reproductive embryos constructed using RNA markers according to Example 1 of the present invention is shown. Detailed Implementation

[0030] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present invention will now be described in detail with reference to the embodiments.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0032] Terminology Explanation:

[0033] Cell-free RNA in seminal plasma (SRP) refers to RNA molecules present in seminal plasma, primarily derived from sperm cells and testicular tissue cells. It is a type of RNA that exists freely in seminal fluid. SRP plays an important role in the reproductive system and may be involved in the formation and regulation of sperm function.

[0034] lncRNA (Long non-coding RNA): refers to non-coding RNA with a length greater than 200 nucleotides.

[0035] Forward Feature Selection (FROM) is a feature subset selection method used to identify the features that most significantly improve the model's predictive performance from a large pool of original features. This method is incremental; it starts with an empty set or a model containing only one feature, then adds features one by one until a stopping condition is met, selecting the feature that maximizes the improvement in model performance. The feature subset X starts as an empty set, and each time a feature x is selected and added to the feature subset X, the feature function J(X) is optimized. Simply put, it selects a feature that optimizes the evaluation function each time, making it a simple greedy algorithm.

[0036] As mentioned in the background section, existing methods for predicting the developmental outcome of assisted reproductive embryos typically utilize specific proteins in serum and follicular fluid, specific RNAs in sperm, and multi-omics characteristics in embryo culture medium. However, these methods often rely on invasive sampling, and predicting from a single embryo in the culture medium has limitations. Furthermore, by the time of detection and prediction, the patient is already in the treatment phase, resulting in insufficient prediction time windows and imposing corresponding physical, psychological, and economic burdens on the patient. Therefore, in this application, the inventors attempt to develop a new method for predicting the developmental outcome of assisted reproductive technology, capable of effectively predicting the developmental outcome of the patient's assisted reproductive embryos before treatment, and thus propose a series of protection schemes in this application.

[0037] In a first typical embodiment of this application, an application of an RNA biomarker in the preparation of products related to predicting developmental outcomes in assisted reproductive technology is provided. The RNA biomarker includes one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

[0038] The RNA markers in this application include mRNA markers and lncRNA markers, wherein the mRNA markers include one or more of the following mRNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, and PLAAT4; and the lncRNA markers include one or more of the following lnRNA markers: ENSG00000258422, ENSG00000233633, ENSG00000247311, and LURAP1L-AS1.

[0039] Current technologies for predicting and detecting developmental outcomes in assisted reproductive technologies primarily focus on the female perspective. However, invasive sampling methods for female samples are relatively difficult, hindering rapid clinical translation. Furthermore, there are few methods that study male infertility. While some reports suggest predicting embryonic development outcomes using specific RNAs in sperm and multi-omics characteristics in embryo culture media, these methods only predict the outcome of a single embryo and cannot provide a holistic assessment of efficacy and outcomes over multiple treatment cycles. Moreover, current technologies often require patients to be in the treatment phase before detection and prediction, resulting in a limited predictive window and imposing physical, psychological, and financial burdens on patients, thus hindering widespread clinical application.

[0040] The RNA biomarker in this application is obtained by differential expression screening in seminal plasma samples from successfully developed in vitro embryos and those from failed in vitro embryos. It can serve as a biomarker for predicting assisted reproductive outcomes. Using this RNA biomarker to predict the developmental outcome of assisted reproductive embryos allows for non-invasive sampling, requiring only a seminal plasma sample from the patient. Furthermore, this application enables the prediction of in vitro embryo development outcomes before entering the assisted reproductive treatment cycle, determining the availability of transplantable embryos and providing timely reference and basis for patients to choose donor sperm treatment, thus improving treatment outcomes and reducing the burden on patients. The predictive model for assisted reproductive outcomes obtained by training machine algorithms using the aforementioned RNA biomarker in this application has been tested and verified, demonstrating high accuracy, low cost, and simple operation, making it more suitable for clinical application.

[0041] In a preferred embodiment, the developmental outcome includes the probability of successful development of an assisted reproductive embryo and / or the probability of successful clinical pregnancy of the assisted reproductive embryo; preferably, the related products include kits, chips, and / or electronic devices.

[0042] In a second typical embodiment of this application, a kit for predicting assisted reproductive outcomes is provided. The kit includes a detection reagent for detecting the expression levels of RNA biomarkers. The RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

[0043] In a preferred embodiment, the kit includes probes and / or primers for detecting RNA markers.

[0044] Based on the existing kit preparation principles, a detection kit for the content of the RNA biomarker of the present invention can be prepared, or a detection probe and chip for detecting and predicting assisted reproductive outcomes can be prepared using the RNA biomarker of the present application, or a corresponding prediction model can be obtained by training a machine learning algorithm.

[0045] Machine learning can be implemented through machine learning methods. Machine learning methods are selected from regression, classification, or combinations thereof. "Machine learning" generally refers to algorithms that give computers the ability to learn without explicit programming, including algorithms that learn from data and make predictions about that data. The machine learning methods used in this invention may include random forests, minimum absolute shrinkage and selection operator logistic regression, regularized logistic regression, XGBoost, decision tree learning, artificial neural networks, deep neural networks, support vector machines, rule-based machine learning, generalized linear models, gradient boosting machines, etc.

[0046] In a third typical embodiment of this application, a method for predicting the developmental outcome of assisted reproductive technology is provided. The method includes: obtaining a first feature dataset of cell-free RNA from a seminal plasma sample of a test subject, the first feature dataset including expression level data of RNA markers; inputting the first feature dataset into a prediction model for the developmental outcome of assisted reproductive embryos to obtain a prediction score, the prediction score being used to indicate the success rate of assisted reproductive embryo development in the test subject; the RNA markers include one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

[0047] In a preferred embodiment, the seminal plasma sample of the subject is obtained before the subject undergoes assisted reproduction. By obtaining a seminal plasma sample from the patient (subject) before undergoing assisted reproduction, the prediction method of this application can predict the outcome of assisted reproductive embryo development before treatment. Furthermore, the seminal plasma sample of this application can be obtained through non-invasive sampling, avoiding the drawbacks of the prior art and reducing the physical, mental, and economic burden on the patient.

[0048] In a preferred embodiment, the method for obtaining the first feature dataset includes: performing quantitative analysis on free RNA to obtain expression level data of RNA markers, thereby obtaining the first feature dataset.

[0049] The quantitative analysis described above includes: quality control of the sequencing files of seminal plasma free RNA from the test subject (including splice adapter sequences, low-quality reads, and reads <17 bp in length), and removal of rRNA, vtRNA, and γRNA sequences to obtain a first-filtered seminal plasma free RNA sequencing file; alignment of the first-filtered seminal plasma free RNA sequencing file with a reference transcriptome, and quantification of mRNA and lncRNA, with expression levels standardized using TPM (Transcripts Per Million); after standardization, a seminal plasma free RNA expression profile is obtained, and expression data of the aforementioned RNA markers are screened from the seminal plasma free RNA expression profile of the test subject. The quantitative analysis method of this application can also utilize other known quantitative methods in the art to quantify mRNA and lncRNA; the method for extracting seminal plasma free RNA in this application can utilize commonly used extraction methods or kits in the art to extract seminal plasma free RNA from the test subject.

[0050] In a preferred embodiment, the predictive model for the developmental outcome of assisted reproductive embryos is trained using a machine learning algorithm based on data on the expression levels of RNA markers in seminal plasma samples from successfully developed in vitro embryos and seminal plasma samples from failed in vitro embryos.

[0051] The method for predicting assisted reproductive outcomes in this application involves non-invasively collecting and extracting cell-free RNA from seminal plasma from patients with low male fertility undergoing assisted reproductive treatment. The method tracks and records the developmental outcomes of in vitro embryos during the assisted reproductive treatment cycle. By comparing the expression levels of cell-free RNA in seminal plasma samples from successfully developed and unsuccessfully developed embryos, and using machine learning algorithms, RNA biomarkers for predicting successful in vitro embryo development are selected. A corresponding prediction model is then trained and constructed to predict the outcome of in vitro embryo development before the start of the assisted reproductive treatment cycle. The prediction model automatically calculates a prediction score to determine the probability of successful development in the tested subject. In this model, a prediction score greater than 0.5 indicates a high probability of successful development, while a score less than 0.5 indicates a low probability. Appropriate treatment plans can be formulated based on this prediction score.

[0052] In a preferred embodiment, the machine learning algorithm includes one or more of the following: a generalized linear model, a gradient boosting machine, a random forest, or a support vector machine.

[0053] In a fourth typical embodiment of this application, a method for screening RNA biomarkers to predict the developmental outcome of assisted reproduction is provided, the method comprising:

[0054] I) Obtain the expression levels of all free RNAs in the training set samples to obtain the second feature dataset. The expression level of each free RNA is used as a candidate feature. The training set samples include seminal plasma samples from successfully developed in vitro embryos and seminal plasma samples from failed in vitro embryos. II) Input each candidate feature into Formula I as shown below, and perform a chi-square test successively to calculate the P-value of each candidate feature. Formula I represents the formula for predicting the developmental outcome of assisted reproduction constructed using different candidate features.

[0055]

[0056] Where e represents the exponential score; b represents the bias of the prediction model; f1, f2, ..., f n Represents the values ​​of candidate features in the prediction model; β1, β2, ..., β n y' represents the feature vector corresponding to the value of the candidate feature in the prediction model; y' represents the linear combination part of the prediction model, which represents the sum of the product of all candidate features and all feature vectors.

[0057] III) Screen candidate features whose P-values ​​meet the preset threshold as RNA biomarkers for predicting the developmental outcome of assisted reproduction (the preset threshold means that candidate features with the smallest P-value in each chi-square test must meet the condition P<0.05 to be considered as RNA biomarkers for predicting the developmental outcome of assisted reproduction).

[0058] IV) When the p-value of all candidate features in the prediction model is no longer less than 0.05, stop the screening; RNA markers include one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311 or LURAP1L-AS1.

[0059] In a preferred embodiment, the chi-square test includes a likelihood ratio chi-square test; preferably, the method for obtaining the expression level of free RNA includes: performing quantitative analysis on the free RNA to obtain expression level data of RNA markers.

[0060] The method for screening RNA biomarkers for predicting assisted reproductive development outcomes in this application includes feature selection using a forward feature selection method. Specifically, this application obtained 19,655 mRNA biomarkers and 19,059 lncRNA biomarkers, totaling 38,714 features, through preliminary experimental screening. The forward feature selection method was then used for feature optimization, inputting each of the 38,714 features into Formula I, which represents a logistic linear model (prediction model).

[0061] In the initial stage of forward feature selection, both the feature vector β and the feature set f are empty. Each candidate feature x is then selected... i The added feature x is added to the current model (substituting it into the above formula to calculate the addition to the feature set f), and the likelihood ratio chi-square test is used to evaluate the improvement in model performance. For the added feature x... i Calculate the corresponding chi-square test value X. 2 The p-value test was used to select the candidate feature with the smallest p-value (p<0.05) as the optimal feature, which was then recorded and retained in the final prediction model. After each optimal feature selection, new features were added, and the model was updated, adding the corresponding feature vector β (automatically learned through a computer algorithm). This process continued until no more candidate features could significantly improve model performance; that is, when the p-values ​​of all candidate features in the chi-square test were no longer less than 0.05, the screening stopped. Using the above method, 13 RNA biomarkers for predicting assisted reproductive development outcomes were obtained in this application, including 9 mRNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, and PLAAT4; and 4 lnRNA biomarkers: ENSG00000258422, ENSG00000233633, ENSG00000247311, and LURAP1L-AS1.

[0062] In a fifth typical embodiment of this application, a method for constructing a predictive model for the developmental outcome of assisted reproductive technology is provided. The method includes: S1) detecting the expression levels of RNA markers in training set samples; the training set samples include seminal plasma samples from successfully developed in vitro embryos and seminal plasma samples from failed in vitro embryos; S2) constructing a predictive model for the developmental outcome of assisted reproductive embryos based on the expression levels of RNA markers and a machine learning algorithm; the RNA markers include NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, and LURAP1L-AS1.

[0063] This application uses RNA biomarkers obtained through the aforementioned forward feature selection method. By constructing a logistic regression model, a predictive model for the developmental outcome of assisted reproductive embryos is obtained. The receiver operating characteristic (ROC) curve shows an AUC of 0.95, indicating that the predictive model obtained in this application has high goodness of fit, strong specificity, and high sensitivity. The predictive model for the developmental outcome of assisted reproductive embryos constructed using RNA biomarkers from seminal plasma can input the expression levels of the aforementioned RNA biomarkers from the seminal plasma samples of the test subjects into the model to predict the developmental outcome of the assisted reproductive embryos. Compared with existing technologies, this avoids the sampling difficulties of invasive detection methods and allows for the prediction of in vitro embryo developmental outcomes before entering the assisted reproductive treatment cycle, which is more conducive to further clinical application.

[0064] In a preferred embodiment, in S2), the machine learning algorithm includes one or more of the following: a generalized linear model, a gradient boosting machine, a random forest, or a support vector machine.

[0065] In a sixth typical embodiment of this application, an electronic device for predicting the developmental outcome of assisted reproductive technology is provided. The electronic device includes: an acquisition unit for acquiring a first feature dataset of cell-free RNA from the seminal plasma of a test subject, the first feature dataset including expression level data of one or more RNA markers; and a prediction unit for inputting the first feature dataset into a prediction model for the developmental outcome of assisted reproductive embryos and outputting a prediction score, the prediction score representing the success rate of assisted reproductive embryo development of the test subject; the RNA markers include one or more of the following RNA markers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, and LURAP1L-AS1.

[0066] In a preferred embodiment, the device incorporates a predictive model for the developmental outcome of assisted reproductive embryos. The predictive model is trained using a machine learning algorithm based on data on the expression levels of RNA markers in seminal plasma samples from successfully developed in vitro embryos and seminal plasma samples from failed in vitro embryos.

[0067] In a seventh typical embodiment of this application, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the above-mentioned method for predicting the developmental outcome of assisted reproductive technology, the above-mentioned method for screening RNA markers for predicting the developmental outcome of assisted reproductive technology, or the above-mentioned method for constructing a predictive model for the developmental outcome of assisted reproductive technology by executing the executable instructions.

[0068] In an eighth typical embodiment of this application, a computer-readable storage medium is provided, which includes a stored computer program, wherein, when the computer program is running, it controls the device where the computer-readable storage medium is located to execute the above-mentioned method for predicting the developmental outcome of assisted reproduction, the above-mentioned method for screening RNA markers for predicting the developmental outcome of assisted reproduction, or the above-mentioned method for constructing a predictive model for the developmental outcome of assisted reproduction.

[0069] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the present invention.

[0070] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus hardware devices such as detection devices. Based on this understanding, the data processing part of the technical solution of this application can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of the embodiments of this application.

[0071] This application can be used in a wide range of general-purpose or special-purpose computing system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices.

[0072] The method provided in this application can be executed on a terminal, computer terminal, or similar computing device. Taking running on a terminal as an example, Figure 1 This is a hardware structure block diagram of a predictive model for the developmental outcome of assisted reproductive embryos according to an embodiment of the present invention. Figure 1 As shown, a terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor A1 (processor A1 may include, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA), etc.) and a memory B1 for storing data are also shown. Optionally, the terminal may further include a transmission device C1 for communication functions and an input / output device D1. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0073] Memory B1 can be used to store computer programs, such as application software programs and modules, like the computer programs corresponding to the methods of segment concatenation, clustering, and consistency processing in this embodiment of the invention. Processor A1 executes various functional applications and data processing by running the computer programs stored in memory B1, thereby implementing the methods described above. Memory B1 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, memory B1 may further include memory remotely located relative to processor A1, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0074] Transmission device C1 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the terminal's communication provider. In one example, transmission device C1 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, transmission device C1 may be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0075] Obviously, those skilled in the art should understand that some modules or steps of this application described above can be implemented on general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device, or fabricating them separately as individual integrated circuit modules, or fabricating multiple modules or steps as a single integrated circuit module. Thus, this application is not limited to any specific hardware and software combination. The beneficial effects of this application will be further explained in detail below with reference to specific embodiments.

[0076] The beneficial effects of this application will be explained in more detail below with reference to specific embodiments.

[0077] Example 1

[0078] Data was collected and used as training and validation sets for the model based on seminal plasma samples from 113 male infertility patients, including those with oligospermia, asthenospermia, and azoospermia. Of these, 33 embryos were successfully implanted during assisted reproductive technology (ART), while 80 were not.

[0079] 1. Acquisition of cell-free RNA from seminal plasma

[0080] All seminal plasma samples were centrifuged at 10,000 g for 30 min at 4 °C to remove sperm. The supernatant was collected, and TRIzol LS was added at a ratio of 1:3 and immediately vortexed to mix. Free RNA was then extracted to obtain free RNA from all seminal plasma samples.

[0081] 2. Sequencing

[0082] Library construction of seminal plasma free RNA from the above samples was performed using PALM-Seq (Polyadenylation ligation-mediated sequencing), and the seminal plasma free RNA was sequenced using next-generation sequencing to obtain a complete transcription map of the seminal plasma free RNA.

[0083] 3. Quantitative analysis of free RNA expression profiles

[0084] Quality control was performed on the sequencing data of cell-free RNA from seminal plasma, including cut adapter sequences, low-quality reads, and reads <17 bp in length.

[0085] The rRNA, vtRNA, and YRNA sequences were removed using the alignment software bowtie.

[0086] The remaining reads were aligned to the human transcriptome (in the order of miRNA, tRNA, and piRNA; mRNA and lncRNA; and finally, other RNAs). mRNA and lncRNA were quantified using standard quantitative software RSEM, and expression levels were normalized using TPM (Transcripts Per Million).

[0087] TPM=(Ni / Li)*1000000 / (sum(N1 / L1+N2 / L2+N3 / L3+…+Nn / Ln));

[0088] 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 standardization by length.

[0089] 4. Screening of RNA biomarkers for predicting embryonic developmental outcomes in assisted reproductive technology

[0090] Through preliminary experiments, 19,655 mRNA biomarkers and 19,059 lncRNA biomarkers were identified, totaling 38,714 features. To simplify the model and select features that play a crucial role in predicting the developmental outcomes of assisted reproductive embryos, a forward feature selection method was used for feature optimization. Each of the 38,714 features was individually input into a logistic linear model (a predictive model for assisted reproductive developmental outcomes constructed from different candidate features), as shown in Equation I.

[0091]

[0092] e represents the value of the exponential fraction output by logical transformation, which conforms to a distribution from 0 to 1;

[0093] f1, f2, ..., f n This represents the value of a candidate feature in the model, which is usually a feature column in the input data;

[0094] β1, β2, ..., β n This represents the feature vector corresponding to the candidate features in the model, indicating the magnitude of the influence of each candidate feature on y′;

[0095] b represents the bias value of the prediction model;

[0096] y′ represents the linear combination part of the model, which is the sum of the product of all candidate features and all feature vectors.

[0097] In the initial stage of forward feature selection, the feature vector β and the feature set f are both empty. Each candidate feature is added to the current model (substituting into the above formula means adding it to the feature set f for calculation), and the likelihood ratio chi-square test is used to evaluate the degree of improvement in model performance.

[0098] For the added feature x i Calculate the corresponding chi-square test value X. 2 And the corresponding p-value, select the candidate feature with the smallest p-value (i.e., p < 0.05) as the optimal feature, and record it and keep it in the final prediction model to be generated.

[0099] After each optimal feature is selected, new features are added, and the model is updated by adding the corresponding feature vector β (automatically learned through a computer algorithm). This process continues until no more candidate features can significantly improve model performance, i.e., when the p-value of all candidate features in the chi-square test is no less than 0.05, the selection process stops.

[0100] Following the above method, the final selected features include 9 mRNA features (mRNA biomarkers): NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, and 4 lncRNA features (lncRNA biomarkers): ENSG00000258422, ENSG00000233633, ENSG00000247311, and LURAP1L-AS1, for a total of 13 features. These selected features will be used to construct a predictive model for assisted reproductive development outcomes. The contribution of the 13 final selected features (RNA biomarkers) to the predictive effect of embryonic development outcomes is as follows: Figure 2 As shown.

[0101] 5. Construct a predictive model for developmental outcomes in assisted reproductive technology.

[0102] The method for constructing a predictive model for assisted reproductive development outcomes based on the logistic regression algorithm is as follows: assuming an empty model... By adding the first feature x1, the model is updated to M1 = β1x1. Adding the second feature f2 updates the model to M2 = β1x1 + β2x2. This process continues until the k-th feature is added, at which point the model is represented as... (x1, x2, ..., x n Represents the feature values ​​in the model; β1, β2, ..., β n (This represents the feature vector in the model, which is automatically learned from the training set data). Using the 13 RNA biomarker features selected through step 4 above, a predictive model for the developmental outcome of assisted reproduction in this application is constructed:

[0103] This embodiment employs a ten-fold cross-validation method to evaluate the model's predictive performance. Each time, all samples (seminal plasma samples from the aforementioned 113 male infertility patients) are divided into a training set and a validation set at a 9:1 ratio. The RNA biomarker expression levels of the training set samples are used for training and building the predictive model, while the RNA biomarker expression levels of the validation set samples are used to validate the predictive efficacy of the model built from the training set samples. The area under the curve (AUC) is calculated as the primary evaluation metric for the model's predictive performance. This division is repeated ten times, ensuring that each sample is included in the validation set at least once to validate the predictive model built from the training set samples.

[0104] The final ROC curve of the predictive model for the developmental outcome of assisted reproductive technology obtained in this application has an average AUC of 0.95, proving that the predictive model can accurately predict the developmental outcome of embryos treated with assisted reproductive technology. The ROC curve of the predictive model for the developmental outcome of assisted reproductive technology in this application is shown below. Figure 3 As shown, Figure 3 In Chinese, "ROC cruve" refers to the ROC curve.

[0105] As can be seen from the above description, the embodiments of the present invention achieve the following technical effects: This application uses the cell-free RNA expression profile data of seminal plasma of male infertile patients to predict the effect of assisted reproductive treatment, which can achieve non-invasive sampling. The constructed prediction model of assisted reproductive outcome can judge the development outcome of assisted reproductive embryos. It can predict the development outcome of in vitro embryos in advance before entering the patient's assisted reproductive treatment cycle. After being translated into clinical applications, it can predict the treatment effect before infertility treatment, provide timely reference and basis for patients to choose the treatment plan of donor sperm, rationally select the treatment plan to improve the treatment outcome, and timely reduce the burden on patients. Compared with the prior art, it is more suitable for clinical promotion.

[0106] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. The application of RNA biomarkers in the preparation of products related to predicting developmental outcomes in assisted reproductive technologies, characterized in that, The RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

2. The application according to claim 1, characterized in that, The developmental outcomes include the probability of successful development of assisted reproductive embryos and / or the probability of successful clinical pregnancy with assisted reproductive embryos. Preferably, the related products include reagent kits, chips, and / or electronic devices.

3. A kit for predicting developmental outcomes in assisted reproductive technology, characterized in that, The kit includes detection reagents for detecting the expression levels of RNA biomarkers; The RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

4. The reagent kit according to claim 3, characterized in that, The kit also includes probes and / or primers for detecting the RNA markers.

5. A method for predicting developmental outcomes in assisted reproductive technology, characterized in that, The prediction method includes: A first feature dataset of free RNA from a seminal plasma sample of the subject to be tested is obtained, the first feature dataset including expression level data of RNA markers; The first feature dataset is input into the prediction model for the developmental outcome of assisted reproductive embryos to obtain a prediction score, which is used to indicate the success rate of assisted reproductive embryo development of the subject. The RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

6. The prediction method according to claim 5, characterized in that, The seminal plasma sample of the subject to be tested was obtained before the subject underwent assisted reproduction.

7. The prediction method according to claim 5, characterized in that, The methods for obtaining the first feature dataset include: The free RNA is quantitatively analyzed to obtain the expression level data of the RNA marker, thereby obtaining the first feature dataset.

8. An electronic device for predicting developmental outcomes in assisted reproductive technology, characterized in that, The electronic device includes: The acquisition unit is used to acquire a first feature dataset of cell-free RNA in the seminal plasma sample of the test subject, the first feature dataset including expression level data of RNA markers; The prediction unit is used to input the first feature dataset into the prediction model of assisted reproductive embryo development outcome and output a prediction score, wherein the prediction score is used to indicate the success rate of assisted reproductive embryo development of the subject. The RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

9. A method for screening RNA biomarkers to predict assisted reproductive outcomes, characterized in that, The screening method includes: I) Obtain the expression level data of all free RNAs in the training set samples to obtain the second feature dataset, with the expression level data of each free RNA as a candidate feature; The training set samples include seminal plasma samples from successfully developed in vitro embryos and seminal plasma samples from failed in vitro embryos. II) Input each of the candidate features into Formula I as shown below, and perform a chi-square test one by one to calculate the P-value of each candidate feature. Formula I represents a formula for a predictive model of developmental outcomes in assisted reproduction constructed using different candidate features; Where e represents the exponential fraction; b represents the bias value of the prediction model; f1, f2, ..., f n This represents the value of the candidate feature in the prediction model; β1, β2, ..., β n This represents the feature vector corresponding to the value of the candidate feature in the prediction model; y′ represents the linear combination part of the prediction model, which is the sum of the product of all the candidate features and all the feature vectors; III) Screen the candidate features whose P values ​​meet a preset threshold as RNA biomarkers for predicting the developmental outcome of assisted reproduction; The RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

10. The screening method according to claim 9, characterized in that, The chi-square test includes the likelihood ratio chi-square test; Preferably, the method for obtaining the expression level of the free RNA includes: performing quantitative analysis on the free RNA to obtain expression level data of the RNA marker.

11. A method for constructing a predictive model for developmental outcomes in assisted reproductive technology, characterized in that, The construction method includes: S1) Data on the expression levels of RNA biomarkers in the training set samples; The training set samples include seminal plasma samples from successfully developed in vitro embryos and seminal plasma samples from failed in vitro embryos. S2) Based on the expression levels of the RNA markers and machine learning algorithms, construct a predictive model for the developmental outcome of the assisted reproductive embryos; The RNA biomarkers include one or more of the following RNA biomarkers: NECTIN2, TUBB3, SAXO1, FBLN2, CLIP2, BCL2L15, ASB12, CRYAB, PLAAT4, ENSG00000258422, ENSG00000233633, ENSG00000247311, or LURAP1L-AS1.

12. An electronic device, characterized in that, The electronic device includes: Processor; and Memory for storing the executable instructions of the processor; The processor is configured to execute, via executing the executable instructions, the method for predicting the developmental outcome of assisted reproduction as described in any one of claims 5 to 7, the method for screening RNA markers for predicting the developmental outcome of assisted reproduction as described in any one of claims 9 or 10, or the method for constructing a predictive model for the developmental outcome of assisted reproduction as described in claim 11.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the prediction method for assisted reproductive development outcomes as described in any one of claims 5 to 7, the screening method for RNA markers for predicting assisted reproductive development outcomes as described in any one of claims 9 or 10, or the method for constructing a prediction model for assisted reproductive development outcomes as described in claim 11.

Images