Immune state and immune age assessment method and system based on TCR immune repertoire

Abstract

The invention discloses an immune state and immune age assessment method and system based on a TCR immune repertoire, and belongs to the technical field of immune monitoring. A TCR sequencing result of a target person is obtained, the TCR sequencing result comprises first sequencing data corresponding to the target person, a multi-dimensional feature data set is determined according to the TCR sequencing result, and the immune state and the immune age of the target person are determined according to the multi-dimensional feature data set and a pre-training deep learning model, the pre-trained deep learning model is a multi-task model, and the beneficial effect of improving the accuracy of the immune state is achieved.

Description

Technical Field

[0001] This invention relates to the field of immune monitoring technology, and in particular to a method and system for assessing immune status and immune age based on a TCR immune repertoire. Background Technology

[0002] In the fields of precision medicine and health management, accurate assessment of an individual's immune system function is increasingly becoming a crucial aspect of disease prevention, health monitoring, and aging research. The state of the immune system is not entirely synchronized with an individual's actual physiological age, and its functional changes profoundly affect the body's health level and disease susceptibility. Traditional immune assessment methods often rely on limited hematological indicators, such as lymphocyte subset counts or cytokine level detection. While these methods have reference value, they often only reflect one aspect of the immune system and struggle to comprehensively capture its high heterogeneity and dynamic complexity. Furthermore, the data from sub-healthy and aging individuals can negatively impact the accuracy of identifying immune status during physical assessments.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a method and system for assessing immune status and immune age based on a TCR immune repertoire, aiming to improve the accuracy of immune status assessment. To achieve the above objective, this invention provides a method for assessing immune status and immune age based on a TCR immune repertoire, which includes the following steps: Obtain the TCR sequencing results of the target personnel, wherein the TCR sequencing results include the first sequencing data corresponding to the target personnel; A multidimensional feature dataset was determined based on the TCR sequencing results; The immune status and immune age of the target individuals are determined based on the multi-dimensional feature dataset and the pre-trained deep learning model, wherein the pre-trained deep learning model is a multi-task model. Optionally, the step of determining the multi-dimensional feature dataset based on the TCR sequencing results includes: The first sequencing data is matched with a pre-constructed TCR data reference library for sub-healthy populations to obtain matching results. The matching results include: first-type sequence data that have been paired and second-type sequence data that have not been paired. The first type of sequence data is updated based on the first type of sequence data and the corresponding target abundance threshold, wherein the target abundance threshold is the average abundance of sequence data that is the same as the first type of sequence data among sub-healthy individuals; The multidimensional feature dataset is determined based on the first sequencing data.

[0005] Optionally, the step of updating the first type of sequence data based on the first type of sequence data and the corresponding target abundance threshold includes: When the first abundance corresponding to the first type of sequence data is greater than or equal to the target abundance threshold, the first type of sequence data in the first sequencing data is deleted; When the first abundance corresponding to the first type of sequence data is less than the target abundance threshold, the first type of sequence data is determined to be in the first sequencing data.

[0006] Optionally, the step of determining the multi-dimensional feature dataset based on the first sequencing data includes: Clean up any abnormal data in the first sequencing data; The target feature data of the first sequencing data is extracted to obtain multiple target feature data. The feature types of the target feature data include: total number of clonal types, relative abundance of each CDR3 clonal type, physicochemical characteristics of each CDR3, and gene usage preference. The multidimensional feature dataset is determined based on multiple target feature data.

[0007] Optionally, before the step of determining the immune status and immune age of the target individual based on the multi-dimensional feature dataset and the pre-trained deep learning model, the method further includes: Normalization processing of the multi-dimensional feature dataset; The multi-dimensional feature dataset is input into the contrastive autoencoder, which outputs an augmentation vector. The contrastive autoencoder is a Siamese network with shared weights. The multidimensional feature dataset is updated based on the enhancement vector.

[0008] Optionally, the step of determining the immune status and immune age of the target individual based on the multi-dimensional feature dataset and the pre-trained deep learning model includes: The multi-dimensional feature dataset is input into the pre-trained deep learning model to obtain the output result of the pre-trained deep learning model; The immune status and immune age are determined based on the output results; The pre-trained deep learning model is obtained by training an MTL model with historical TCR sequencing data. The MTL model includes an encoder and more than one decoder layer, and the encoder data is transmitted to each decoder layer.

[0009] Optionally, the step of inputting the multi-dimensional feature dataset into the pre-trained deep learning model further includes: Obtain historical TCR sequencing data for each population type; The training and test sets were determined based on the historical TCR sequencing data. The MTL model is trained based on the training set to obtain the first model; A pre-trained deep learning model is determined based on the first model and the test set.

[0010] Furthermore, to achieve the above objectives, the present invention also provides an immune status and immune age assessment system based on a TCR immune repertoire, wherein the TCR immune repertoire-based immune status and immune age assessment system includes: The acquisition module is used to acquire the TCR sequencing results of the target personnel, wherein the TCR sequencing results include the first sequencing data corresponding to the target personnel; The analysis module is used to determine a multi-dimensional feature dataset based on the TCR sequencing results; The prediction module is used to determine the immune status and immune age of the target person based on the multi-dimensional feature dataset and the pre-trained deep learning model, wherein the pre-trained deep learning model is a multi-task model.

[0011] Furthermore, to achieve the above objectives, the present invention also provides an immune status and immune age assessment device based on a TCR immune repertoire. The TCR immune repertoire-based immune status and immune age assessment device includes: a memory, a processor, and an immune status and immune age assessment program based on a TCR immune repertoire stored in the memory and executable on the processor. The TCR immune repertoire-based immune status and immune age assessment program is configured to implement the steps of the TCR immune repertoire-based immune status and immune age assessment method described above.

[0012] Furthermore, to achieve the above objectives, the present invention also provides a storage medium, characterized in that the storage medium stores an immune status and immune age assessment program based on a TCR immune repertoire, wherein when the TCR immune repertoire-based immune status and immune age assessment program is executed by a processor, it implements the steps of the TCR immune repertoire-based immune status and immune age assessment method described above.

[0013] This invention proposes a method for assessing immune status and immune age based on a TCR immune repertoire. This method obtains the TCR sequencing results of a target individual, including the first sequencing data corresponding to that individual. A multi-dimensional feature dataset is determined based on the TCR sequencing results. The immune status and immune age of the target individual are then determined using the multi-dimensional feature dataset and a pre-trained deep learning model. This method effectively avoids the influence of data from sub-healthy or aging individuals on the prediction results, thereby improving the accuracy of immune status identification. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the structure of the immune status and immune age assessment device based on the TCR immune repertoire in the hardware operating environment of the embodiment of the present invention; Figure 2 This is a flowchart illustrating the first embodiment of the method for assessing immune status and immune age based on a TCR immune repertoire of the present invention. Figure 3 This is a flowchart illustrating the second embodiment of the method for assessing immune status and immune age based on a TCR immune repertoire according to the present invention.

[0015] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0016] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0017] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of an immune status and immune age assessment device based on a TCR immune repertoire in the hardware operating environment of an embodiment of the present invention.

[0018] like Figure 1As shown, the TCR-based immune repertoire-based immune status and immune age assessment device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, an interactive device 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The interactive device 1003 may include a display screen and an input unit such as a keyboard. Optionally, the interactive device 1003 may also be connected to the communication bus via standard wired or wireless interfaces. The network interface 1004 may optionally include standard wired or wireless interfaces (such as a Wi-Fi interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0019] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on TCR-based immune repertoire assessment devices for assessing immune status and immune age. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0020] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a data storage module, a network communication module, a user interface module, and an immune status and immune age assessment program based on the TCR immune repertoire.

[0021] exist Figure 1 In the TCR-based immune status and immune age assessment device shown, the network interface 1004 is mainly used for data communication with other devices; the interactive device 1003 is mainly used for data interaction with the user; the processor 1001 and memory 1005 in the TCR-based immune status and immune age assessment device of the present invention can be set in the TCR-based immune status and immune age assessment device. The TCR-based immune status and immune age assessment device calls the TCR-based immune status and immune age assessment program stored in the memory 1005 through the processor 1001, and executes the TCR-based immune status and immune age assessment method provided in the embodiments of the present invention.

[0022] This invention provides a method for assessing immune status and immune age based on a TCR immune repertoire, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of a method for assessing immune status and immune age based on a TCR immune repertoire according to the present invention.

[0023] In this embodiment, the method for assessing immune status and immune age based on the TCR immune repertoire includes: Step S1: Obtain the TCR sequencing results of the target personnel, wherein the TCR sequencing results include the first sequencing data corresponding to the target personnel; In this embodiment, TCR sequencing refers to T cell receptor sequencing. T cells refer to T lymphocytes, and the TCR sequencing results include: the CDR3 amino acid sequence of each TCR, the corresponding V, D, and J gene fragment numbers, and the occurrence frequency of each clonal type. Further, optionally, it may also include α-chain and β-chain pairing information, and optionally, it may also include data such as CDR3 amino acid length distribution, average hydrophobicity, and average isoelectric point.

[0024] Step S2: Determine the multi-dimensional feature dataset based on the TCR sequencing results; Specifically, a multi-dimensional feature dataset is extracted based on the TCR sequencing results to obtain the feature dataset. In this embodiment, the number of feature data in the multi-dimensional feature dataset is not limited. In this embodiment, probability data for each CDR3 is constructed by summing the amino acid occurrence frequencies and then normalizing the sums; furthermore, the gene usage frequencies of V, D, J, and C can also be used as feature data. The mean, variance, and peak values ​​are calculated as the feature data.

[0025] Step S3: Determine the immune status and immune age of the target person based on the multi-dimensional feature dataset and the pre-trained deep learning model, wherein the pre-trained deep learning model is a multi-task model.

[0026] Specifically, the multi-dimensional feature dataset is input into the pre-trained deep learning model, whose structure includes a shared underlying encoding network. It should be noted that in this embodiment, the shared underlying encoding network can transmit data to different output layers, which can perform different types of task outputs. Specifically, two task output layers can be set, where the first task output layer can be used to output the immune status of the target individual, and the second task output layer can be used to output the immune age. This enables the acquisition of a user's TCR sequencing results and the prediction of immune status and immune age.

[0027] In this embodiment, by obtaining the TCR sequencing results of the target person, which includes the first sequencing data corresponding to the target person, and determining a multi-dimensional feature dataset based on the TCR sequencing results, the immune status and immune age of the target person are determined based on the multi-dimensional feature dataset and a pre-trained deep learning model. This can effectively avoid the influence of data from sub-healthy and aging populations on the prediction results, thereby improving the accuracy of identifying immune status.

[0028] Furthermore, based on the first embodiment, a second embodiment of the present invention for assessing immune status and immune age based on a TCR immune repertoire is proposed. In this embodiment, referring to... Figure 3 The step of determining the multi-dimensional feature dataset based on the TCR sequencing results includes: Step S21: Match the first sequencing data with the pre-constructed TCR data reference library for sub-healthy populations to obtain matching results. The matching results include: first-type sequence data that have been paired and second-type sequence data that have not been paired. In this embodiment, the TCR data reference library for sub-healthy individuals is a database composed of TCR data collected at the current moment and already labeled as sub-healthy individuals, processed. It should be noted that the data in this TCR data reference library for sub-healthy individuals consists of data that appears frequently in a large number of sub-healthy samples and is significantly low-frequency in healthy samples. The data in this TCR data reference library for sub-healthy individuals is in a labeled state. In this embodiment, the first type of sequence data refers to data that exists simultaneously in both the TCR data reference library for sub-healthy individuals and the first sequencing data; the second type of sequence data refers to other data in the first sequencing data besides the first type of sequence data.

[0029] Step S22: Update the first type of sequence data according to the first type of sequence data and the corresponding target abundance threshold, wherein the target abundance threshold is the average abundance of sequence data that is the same as the first type of sequence data among sub-healthy individuals; It should be noted that the abundance here is used to determine whether the content of the first type of sequence data reaches the target abundance threshold of the TCR data reference library for sub-healthy populations. In this embodiment, data is removed or downweighted based on the comparison results. Furthermore, regardless of how many sequences are removed, the relative abundance of the remaining clones is renormalized and then expanded into a fixed-length feature vector using the same set of V / J gene numbers, CDR3 length buckets, and physicochemical property vectors, thus ensuring that subsequent models receive input of the same size. In addition, the removal ratio can be calculated to determine whether an appropriate amount of data needs to be removed; this can be calculated based on the removed sequence data and the original sequence data. When the removal ratio is higher than 30%, the confidence output is reduced or the first sequencing data is retested. Step S23: Determine the multi-dimensional feature dataset based on the first sequencing data.

[0030] Multiple features of the first sequencing data are extracted as the multi-dimensional feature dataset.

[0031] In this embodiment, the first sequencing data is matched with a pre-constructed TCR data reference library for sub-healthy individuals to obtain matching results. The first type of sequence data is updated according to the first type of sequence data and the corresponding target abundance threshold. Transitional clones that are "high frequency in sub-health and low frequency in health" are removed. These clones are neither health features nor typical disease features and are located in the middle gray area. For healthy individuals, they are almost not matched, thus preserving complete diversity. For sub-healthy individuals, they are matched but weighted down, causing the diversity calculation to be corrected downward. For confirmed patients, the number of matches is between the two, which plays a soft correction role. This can improve the difference between sub-health and health and enhance the accuracy of the model.

[0032] Furthermore, the step of updating the first type of sequence data based on the first type of sequence data and the corresponding target abundance threshold includes: When the first abundance corresponding to the first type of sequence data is greater than or equal to the target abundance threshold, the first type of sequence data in the first sequencing data is deleted; When the first abundance corresponding to the first type of sequence data is less than the target abundance threshold, the first type of sequence data is determined to be in the first sequencing data.

[0033] In this embodiment, for example, if the content of a specific CDR3 amino acid sequence is extremely low, even if it is a Class I sequence, the data will not be removed. The target abundance threshold here can be the average abundance in the TCR data reference library for sub-healthy individuals plus the corresponding two standard deviations. By using different comparison results, the Class I sequence data can be appropriately adjusted. This effectively improves the accuracy of subsequent model classification.

[0034] Furthermore, based on the first or second embodiment, a third embodiment of the present invention for assessing immune status and immune age based on a TCR immune repertoire is proposed. In this embodiment, the step of determining the multi-dimensional feature dataset based on the first sequencing data includes: Clean up any abnormal data in the first sequencing data; The target feature data of the first sequencing data is extracted to obtain multiple target feature data. The feature types of the target feature data include: total number of clonal types, relative abundance of each CDR3 clonal type, physicochemical characteristics of each CDR3, and gene usage preference. The multidimensional feature dataset is determined based on multiple target feature data.

[0035] In this embodiment, abnormal data in the first sequencing data is identified and removed. Specifically, peptide chains corresponding to the CDR3 region that are less than 8 amino acids or more than 20 amino acids in length need to be removed. This is because if the length is less than 8 amino acids, the TCR structure may be incomplete, while if the length is more than 20 amino acids, it is physiologically very rare and may be due to sequencing errors.

[0036] Furthermore, before the step of determining the immune status and immune age of the target individual based on the multi-dimensional feature dataset and the pre-trained deep learning model, the method further includes: Normalization processing of the multi-dimensional feature dataset; The multi-dimensional feature dataset is input into the contrastive autoencoder, which outputs an augmentation vector. The contrastive autoencoder is a Siamese network with shared weights. The multidimensional feature dataset is updated based on the enhancement vector.

[0037] Specifically, the primary multi-dimensional feature set is Z-score standardized along each dimension, making the mean 0 and variance 1 for each dimension, resulting in a standardized primary feature vector. A Siamese network architecture with shared weights is employed, with the following structure and purpose: Encoder E: two fully connected layers with ReLU activation function; Decoder D: used to output the reconstructed vector; Extra branch: a 2-node Softmax discriminator C following the latent space for label supervision. In this embodiment, the augmented vector is used to improve the discriminative power of the vectors. The augmented vector is concatenated with the features of the multi-dimensional feature dataset, and the resulting concatenated vector is used as the new feature for that row of samples, replacing the multi-dimensional feature dataset for subsequent input to the multi-task deep learning model.

[0038] Furthermore, based on any of the above embodiments, a fourth embodiment of the present invention for the method of assessing immune status and immune age based on a TCR immune repertoire is proposed. In this embodiment, the step of determining the immune status and immune age of the target individual based on the multi-dimensional feature dataset and the pre-trained deep learning model includes: The multi-dimensional feature dataset is input into the pre-trained deep learning model to obtain the output result of the pre-trained deep learning model; The immune status and immune age are determined based on the output results; The pre-trained deep learning model is obtained by training an MTL model with historical TCR sequencing data. The MTL model includes an encoder and more than one decoder layer, and the encoder data is transmitted to each decoder layer.

[0039] In this embodiment, specifically, the MTL model refers to a multi-task learning model, which can generally handle multiple different task types. Different decoding layers are set according to different tasks, and optionally, they may include an immune age decoder and a disease risk decoder. The disease risk decoder outputs prediction results in seven dimensions, corresponding to the risk of seven common tumors, thereby determining whether the patient is in a disease state. The immune age decoder outputs a one-dimensional linear regression to predict immune age, which is compared with the actual age to indicate whether premature immune aging has occurred.

[0040] In this embodiment, by inputting the multi-dimensional feature dataset into the pre-trained deep learning model, the output of the pre-trained deep learning model is obtained, and the immune status and immune age are determined based on the output. This allows the determination of the immune status and immune age within a single model through data sharing, thereby improving the accuracy of prediction.

[0041] Furthermore, the step of inputting the multi-dimensional feature dataset into the pre-trained deep learning model includes: Obtain historical TCR sequencing data for each population type; The training and test sets were determined based on the historical TCR sequencing data. The MTL model is trained based on the training set to obtain the first model; A pre-trained deep learning model is determined based on the first model and the test set.

[0042] In this embodiment, historical TCR sequencing data for each population type is acquired. This historical TCR sequencing data is then divided into a training set and a test set according to a preset ratio. Historical features are extracted from both the training and test sets. These historical features are identical to the features in the multi-dimensional feature dataset input to the pre-trained deep learning model. The MTL model is trained using the training set to obtain a first model. A determination is made based on the first model and the test set to see if a corresponding model metric is achieved. If the corresponding model metric is achieved, the first model and the test set are used to determine the pre-trained deep learning model. If the corresponding model metric is not achieved, the model training parameters are adjusted, and the process returns to the step of training the MTL model using the training set to obtain the first model.

[0043] Furthermore, embodiments of the present invention also propose an immune status and immune age assessment system based on a TCR immune repertoire, wherein the immune status and immune age assessment system based on a TCR immune repertoire includes: The acquisition module is used to acquire the TCR sequencing results of the target personnel, wherein the TCR sequencing results include the first sequencing data corresponding to the target personnel; The analysis module is used to determine a multi-dimensional feature dataset based on the TCR sequencing results; The prediction module is used to determine the immune status and immune age of the target person based on the multi-dimensional feature dataset and the pre-trained deep learning model, wherein the pre-trained deep learning model is a multi-task model.

[0044] Furthermore, embodiments of the present invention also propose an immune status and immune age assessment device based on a TCR immune repertoire. The TCR immune repertoire-based immune status and immune age assessment device includes: a memory, a processor, and an immune status and immune age assessment program based on a TCR immune repertoire stored in the memory and executable on the processor. The immune status and immune age assessment program based on a TCR immune repertoire is configured to implement the steps of the TCR immune repertoire-based immune status and immune age assessment method described above.

[0045] Furthermore, embodiments of the present invention also propose a storage medium storing an immune status and immune age assessment program based on a TCR immune repertoire. When the TCR immune repertoire-based immune status and immune age assessment program is executed by a processor, it implements the steps of the TCR immune repertoire-based immune status and immune age assessment method described above.

[0046] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0047] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0048] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0049] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for assessing immune status and immune age based on a TCR immune repertoire, characterized in that, The method for assessing immune status and immune age based on the TCR immune repertoire includes the following steps: Obtain the TCR sequencing results of the target personnel, wherein the TCR sequencing results include the first sequencing data corresponding to the target personnel; A multidimensional feature dataset was determined based on the TCR sequencing results; The immune status and immune age of the target individuals are determined based on the multi-dimensional feature dataset and the pre-trained deep learning model, wherein the pre-trained deep learning model is a multi-task model.

2. The method for assessing immune status and immune age based on a TCR immune repertoire as described in claim 1, characterized in that, The step of determining the multidimensional feature dataset based on the TCR sequencing results includes: The first sequencing data is matched with a pre-constructed TCR data reference library for sub-healthy populations to obtain matching results. The matching results include: first-type sequence data that have been paired and second-type sequence data that have not been paired. The first type of sequence data is updated based on the first type of sequence data and the corresponding target abundance threshold, wherein the target abundance threshold is the average abundance of sequence data that is the same as the first type of sequence data among sub-healthy individuals; The multidimensional feature dataset is determined based on the first sequencing data.

3. The method for assessing immune status and immune age based on a TCR immune repertoire as described in claim 2, characterized in that, The step of updating the first type of sequence data based on the first type of sequence data and the corresponding target abundance threshold includes: When the first abundance corresponding to the first type of sequence data is greater than or equal to the target abundance threshold, the first type of sequence data in the first sequencing data is deleted; When the first abundance corresponding to the first type of sequence data is less than the target abundance threshold, the first type of sequence data is determined to be in the first sequencing data.

4. The method for assessing immune status and immune age based on a TCR immune repertoire as described in claim 2, characterized in that, The step of determining the multi-dimensional feature dataset based on the first sequencing data includes: Clean up any abnormal data in the first sequencing data; The target feature data of the first sequencing data is extracted to obtain multiple target feature data. The feature types of the target feature data include: total number of clonal types, relative abundance of each CDR3 clonal type, physicochemical characteristics of each CDR3, and gene usage preference. The multidimensional feature dataset is determined based on multiple target feature data.

5. The method for assessing immune status and immune age based on a TCR immune repertoire as described in claim 4, characterized in that, Before the step of determining the immune status and immune age of the target individual based on the multi-dimensional feature dataset and the pre-trained deep learning model, the method further includes: Normalization processing of the multi-dimensional feature dataset; The multi-dimensional feature dataset is input into the contrastive autoencoder, which outputs an augmentation vector. The contrastive autoencoder is a Siamese network with shared weights. The multidimensional feature dataset is updated based on the enhancement vector.

6. The method for assessing immune status and immune age based on a TCR immune repertoire as described in any one of claims 1 to 5, characterized in that, The steps of determining the immune status and immune age of the target individual based on the multi-dimensional feature dataset and the pre-trained deep learning model include: The multi-dimensional feature dataset is input into the pre-trained deep learning model to obtain the output result of the pre-trained deep learning model; The immune status and immune age are determined based on the output results; The pre-trained deep learning model is obtained by training an MTL model with historical TCR sequencing data. The MTL model includes an encoder and more than one decoder layer, and the encoder data is transmitted to each decoder layer.

7. The method for assessing immune status and immune age based on a TCR immune repertoire as described in claim 6, characterized in that, Before the step of inputting the multi-dimensional feature dataset into the pre-trained deep learning model, the following steps are also included: Obtain historical TCR sequencing data for each population type; The training and test sets were determined based on the historical TCR sequencing data. The MTL model is trained based on the training set to obtain the first model; A pre-trained deep learning model is determined based on the first model and the test set.

8. A system for assessing immune status and immune age based on a TCR immune repertoire, characterized in that, The TCR-based immune repertoire-based immune status and immune age assessment system includes: The acquisition module is used to acquire the TCR sequencing results of the target personnel, wherein the TCR sequencing results include the first sequencing data corresponding to the target personnel; The analysis module is used to determine a multi-dimensional feature dataset based on the TCR sequencing results; The prediction module is used to determine the immune status and immune age of the target person based on the multi-dimensional feature dataset and the pre-trained deep learning model, wherein the pre-trained deep learning model is a multi-task model.

9. A device for assessing immune status and immune age based on a TCR immune repertoire, characterized in that, The TCR-based immune repertoire-based immune status and immune age assessment device includes: a memory, a processor, and a TCR-based immune repertoire-based immune status and immune age assessment program stored in the memory and executable on the processor, wherein the TCR-based immune repertoire-based immune status and immune age assessment program is configured to implement the steps of the TCR-based immune repertoire-based immune status and immune age assessment method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores an immune status and immune age assessment program based on a TCR immune repertoire. When the TCR immune repertoire-based immune status and immune age assessment program is executed by a processor, it implements the steps of the immune status and immune age assessment method based on a TCR immune repertoire as described in any one of claims 1 to 7.

Images