Molecular detection method, device, computer equipment, storage medium and program product

Abstract

The application discloses a molecular detection method and device, computer equipment, a storage medium and a program product. The method comprises the following steps: acquiring a plurality of omics molecular omics data sets; for each omics data set, performing correlation analysis on the omics data in the omics data set to obtain intra-group correlation characteristics of the omics data set; performing correlation analysis on a plurality of intra-group correlation characteristics to obtain inter-group correlation characteristics of the plurality of intra-group correlation characteristics; determining the influence value of each feature in the inter-group correlation characteristics on a preset omics model, and determining the molecules corresponding to the features with the influence values meeting a preset condition as target molecules; wherein the preset omics model is obtained by training the inter-group correlation characteristics of the plurality of omics data sets. The method can improve the accuracy of molecular detection of the to-be-detected object.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of biological information, and in particular to a molecule detection method and device, computer equipment, storage medium and program product. BACKGROUND

[0002] With the development of personalized medical technology, a technology of detecting user biological information based on a model has appeared. The model can detect important omics molecules from omics data of a user, and provide services such as disease prediction and health assessment for the user.

[0003] At present, features are extracted from multi-omics data and models are trained by means such as deep learning. However, the accuracy of the model obtained based on deep learning is low. SUMMARY

[0004] In view of the above defects or deficiencies in the prior art, it is desirable to provide a molecule detection method and device, computer equipment, storage medium and program product, to improve the accuracy of molecule detection of a to-be-detected object.

[0005] In a first aspect, the present application provides a molecule detection method. The method comprises:

[0006] obtaining a set of omics data of a plurality of omics molecules;

[0007] performing correlation analysis on the omics data in each set of omics data to obtain intra-group correlation features of the set of omics data; performing correlation analysis on the plurality of intra-group correlation features to obtain inter-group correlation features of the plurality of intra-group correlation features;

[0008] determining an influence value of each feature in the inter-group correlation features on a preset omics model, and determining a molecule corresponding to a feature whose influence value meets a preset condition as a target molecule; wherein the preset omics model is trained based on the inter-group correlation features of the plurality of sets of omics data.

[0009] In combination with the first aspect, in a possible implementation manner, the performing correlation analysis on the plurality of intra-group correlation features to obtain the inter-group correlation features of the plurality of intra-group correlation features comprises: performing fusion processing on the plurality of intra-group correlation features to obtain fusion features; and performing correlation analysis on feature data in the fusion features to obtain the inter-group correlation features.

[0010] In combination with the first aspect, in a possible implementation manner, the performing fusion processing on the plurality of intra-group correlation features to obtain the fusion features comprises: performing dimension reduction processing on each intra-group correlation feature to obtain a dimension-reduced feature corresponding to the intra-group correlation feature; and performing fusion processing on the dimension-reduced features to obtain the fusion features.

[0011] With reference to the first aspect, in a possible implementation manner, the molecular detection method further includes: determining a total loss value of the plurality of omics data sets according to the dimension reduction features and the omics data sets corresponding to the dimension reduction features; adjusting the dimension reduction parameters according to the total loss value, and performing dimension reduction processing on the inter-group correlation features according to the adjusted dimension reduction parameters until the total loss value meets a first preset loss condition.

[0012] With reference to the first aspect, in a possible implementation manner, the training process of the preset omics model includes: performing correlation analysis on the plurality of omics data sets to obtain inter-group correlation features, and performing clustering processing on the inter-group correlation features to obtain predicted labels of the molecules corresponding to the features; determining a loss value of the preset omics model according to the predicted labels and the labeled labels; adjusting model parameters of the preset omics model according to the loss value until the loss value meets a second preset loss condition.

[0013] With reference to the first aspect, in a possible implementation manner, the determination of the influence value of each feature in the inter-group correlation features on the preset omics model includes: adjusting each feature data in the inter-group correlation features to a preset value in sequence to obtain updated inter-group correlation features; training the preset omics model according to the updated inter-group correlation features to obtain an updated omics model; and determining the influence value of each feature data according to a performance index difference value between the preset omics model and the updated omics model.

[0014] The second aspect, the present application also provides a molecular detection device. The device comprises:

[0015] The acquisition module is configured to acquire omics data sets of a plurality of omics molecules.

[0016] The analysis module is configured to, for each omics data set, perform correlation analysis on omics data in the omics data set to obtain intra-group correlation features of the omics data set; and perform correlation analysis on the plurality of intra-group correlation features to obtain inter-group correlation features of the plurality of intra-group correlation features.

[0017] The determination module is configured to determine an influence value of each feature in the inter-group correlation features on a preset omics model, and determine a molecule corresponding to a feature whose influence value meets a preset condition as a target molecule; wherein the preset omics model is trained based on the inter-group correlation features of the plurality of omics data sets.

[0018] The third aspect, the present application also provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the method of the first aspect when executing the computer program.

[0019] In a fourth aspect, the present application provides a computer readable storage medium. The computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the method according to the first aspect.

[0020] In a fifth aspect, the present application provides a computer program product. The computer program product comprises a computer program which, when executed by a processor, implements the method according to the first aspect.

[0021] The embodiments of the present application provide a molecular detection method, device, computer equipment, storage medium and program product. A plurality of omics data sets of a to-be-detected object are collected, and correlation analysis is performed on omics data in each omics data set to obtain a plurality of intra-group correlation features. Then, correlation analysis is performed on the plurality of intra-group correlation features to obtain inter-group correlation features. The embodiments of the present application realize learning of interaction relationships between molecules in a substance and between molecules of different substances through two correlation analyses, effectively improve the richness and accuracy of feature data of inter-group correlation features, and improve the accuracy of a preset omics model trained based on the group correlation features. Further, the target molecule determined based on the influence of the preset omics model is more accurate, that is, the accuracy of molecular detection of the to-be-detected object is effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0022] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments, read in conjunction with the accompanying drawings:

[0023] Figures 1-6 A flowchart of a molecular detection method in an embodiment;

[0024] Figure 7 Another flowchart of a molecular detection method in an embodiment;

[0025] Figure 8 A comparison chart of detection results of various molecular detection methods;

[0026] Figure 9 A block diagram of a structure of a molecular detection device in an embodiment;

[0027] Figure 10 An internal structure diagram of a computer equipment in an embodiment. DETAILED DESCRIPTION

[0028] The present application will be further described in conjunction with the accompanying drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related application, and not to limit the application. In addition, it should be noted that, for the convenience of description, only the parts related to the application are shown in the drawings.

[0029] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present application will now be described in detail with reference to the accompanying drawings and embodiments. Furthermore, the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The terms "first" and "second," etc., in the specification and claims of the embodiments of this application are used to distinguish different objects, not to describe a specific order of objects.

[0030] With the development of personalized medicine technology, a technology based on model detection of user bioinformation has emerged. This model can detect important omics molecules in the user's omics data and provide users with services such as disease prediction and health assessment.

[0031] Currently, deep learning and other methods are commonly used for feature extraction and model training of multi-omics data. However, there are complex interactions between omics molecules of the same substance and between omics molecules of different substances, resulting in low accuracy of models obtained based on deep learning.

[0032] In one embodiment, a molecular detection method is provided. This embodiment illustrates the application of this method to a terminal. It is understood that the method can also be applied to a server, or to a system including a terminal and a server, and can be implemented through the interaction between the terminal and the server.

[0033] like Figure 1 As shown, the method includes the following steps:

[0034] Step 101: Obtain a collection of omics data for various omics molecules.

[0035] In this embodiment of the application, bodily fluid samples such as blood and urine of the subject to be tested can be obtained first, and routine blood tests and multi-omics sequencing can be performed on the obtained bodily fluid samples to obtain multi-omics data of the subject to be tested, that is, a set of omics data corresponding to the omics molecules of multiple substances.

[0036] Among them, omics molecules can include genomics, transcriptomics, proteomics, phenome, metabolomics, etc., and each omics molecule corresponds to an omics dataset.

[0037] In a possible implementation manner, after the multi-omics data is acquired, the multi-omics data can be preprocessed. For example, difference analysis can be performed on each omics data, that is, the P value of each omics data is calculated by variance analysis, or chi-square distribution, or t test, or Z test, and the omics data with a P value less than or equal to a preset threshold is determined, which indicates that the substance (gene, protein, etc.) corresponding to the omics data has a significant difference relative to a normal substance state, and a pathological molecule may exist, and further molecular detection is required, so the omics data is retained, and the omics data with a P value greater than the preset threshold is discarded; the preset threshold can be 0.05. Further, the missing values and abnormal values in the retained omics data can be processed by zero filling.

[0038] In step 102, for each omics data set, correlation analysis is performed on the omics data in the omics data set to obtain intra-group correlation features of the omics data set; and correlation analysis is performed on the plurality of intra-group correlation features to obtain inter-group correlation features of the plurality of intra-group correlation features.

[0039] In the embodiment of the present application, considering the interaction relationship between omics molecules of the same substance and the interaction relationship between omics molecules of different substances, correlation analysis can be performed on each omics data set first, the correlation relationship between the omics data in each omics data set is learned, and thus the intra-group correlation features of each omics data set are obtained. Each feature data in the intra-group correlation features can represent the interaction relationship between the omics molecules of the corresponding substance.

[0040] Then, correlation analysis is performed on the plurality of intra-group correlation features, the correlation relationship between the feature data in the plurality of different intra-group correlation features is learned, and thus the inter-group correlation features of the plurality of intra-group correlation features are obtained. Each feature data in the inter-group correlation features can represent the interaction relationship between the omics molecules of the corresponding substance and the interaction relationship between the omics molecules of different substances.

[0041] In a possible implementation manner, self-attention mechanism can be used to perform correlation analysis on each omics data set first. That is, for each omics data in the omics data set, the influence of the remaining omics data on the omics data is determined, and thus a similarity matrix is calculated, as shown in the following formulas (1)-(3):

[0042] (1)

[0043] (2)

[0044] (3)

[0045] wherein,

[0046] wherein, denotes the query matrix corresponding to the i-th set of omics data; denotes the linear transformation matrix of the query value; denotes the key matrix corresponding to the i-th set of omics data; denotes the linear transformation matrix of the key value; denotes the value matrix corresponding to the i-th set of omics data; denotes the linear transformation matrix of the value; i denotes the i-th kind of omics molecule, i.e., the i-th set of omics data; denotes the dimensional characteristics of the omics data; n denotes the number of samples, i.e., the number of omics data; v denotes the number of omics molecule types; d denotes the dimension.

[0047] Further, the intra-group correlation characteristics are calculated according to the plurality of similarity matrices, as shown in the following formula (4):

[0048] (4)

[0049] wherein, denotes the intra-group correlation characteristics of the i-th set of omics data.

[0050] Then, the self-attention mechanism is also used to analyze the correlation of the plurality of intra-group correlation characteristics. For each characteristic data in the plurality of intra-group correlation characteristics, the influence of the remaining characteristic data on the characteristic data is determined, or the influence of the remaining characteristic data on the characteristic data is determined, except for the characteristic data of the intra-group correlation characteristics in which the characteristic data is located. Thus, the similarity matrix is calculated, as shown in the following formulas (5)-(7):

[0051] (5)

[0052] (6)

[0053] (7)

[0054] wherein, denotes the query matrix corresponding to the plurality of intra-group correlation characteristics; denotes the linear transformation matrix of the query value; denotes the key matrix corresponding to the plurality of intra-group correlation characteristics; denotes the linear transformation matrix of the key value; denotes the value matrix corresponding to the plurality of intra-group correlation characteristics; denotes the linear transformation matrix of the value; is the integrated representation of the plurality of intra-group correlation characteristics.

[0055] Further, the inter-group correlation characteristics of the plurality of intra-group correlation characteristics are calculated according to the plurality of similarity matrices, as shown in the following formula (8):

[0056] (8)

[0057] Step 103, determining the influence value of each feature in the inter-group correlation feature on the preset omics model, and determining the molecule corresponding to the feature with the influence value meeting the preset condition as the target molecule.

[0058] The preset omics model is trained based on the inter-group correlation features of a plurality of omics data sets. The preset condition can be that the influence value is greater than a preset influence threshold, or the top M influence values in descending order.

[0059] In the embodiments of the present application, the preset omics model can be trained using inter-group correlation features as sample data. The preset omics model is trained based on the multi-omics data of the to-be-detected object, so it can accurately classify the omics molecules contained in any multi-omics data of the to-be-detected object.

[0060] Similarly, the influence of each feature data in the inter-group correlation features used to train the preset omics model on the preset omics model can represent the importance of the feature data, and further can represent the abnormality of the omics molecule corresponding to the feature data. Therefore, the influence of each feature data in the inter-group correlation features can be analyzed to determine the influence value of each feature data on the preset omics model, and the omics molecule corresponding to the feature data with a larger influence value is determined as the target molecule. For example, the omics molecule corresponding to the influence value greater than the preset influence threshold is determined as the target molecule, or the omics molecule corresponding to the top M influence values in descending order is determined as the target molecule.

[0061] The method provided in the embodiments of the present application first collects a plurality of omics data sets of a to-be-detected object, and performs correlation analysis on the omics data in each omics data set to obtain a plurality of intra-group correlation features, and then performs correlation analysis on the plurality of intra-group correlation features to obtain inter-group correlation features. The embodiments of the present application learn the interaction relationship between the molecules in a substance and the molecules between different substances through two correlation analyses in turn, effectively improve the richness and accuracy of the feature data of the inter-group correlation features, and improve the accuracy of the preset omics model trained based on the inter-group correlation features. Further, the target molecule determined based on the influence of the preset omics model is more accurate, that is, the accuracy of the molecule detection of the to-be-detected object is effectively improved.

[0062] The foregoing embodiments introduce that the plurality of intra-group correlation features can be fused into a feature data set, and then the fused feature data set is analyzed for correlation. For example, the foregoing “performing correlation analysis on the plurality of intra-group correlation features to obtain inter-group correlation features of the plurality of intra-group correlation features” includes, for example, Figure 2The steps shown are:

[0063] Step 201, fusion processing is performed on the multiple intra-group correlation features to obtain a fusion feature.

[0064] Step 202, correlation analysis is performed on the feature data in the fusion feature to obtain inter-group correlation features.

[0065] In the embodiments of the present application, in order to facilitate correlation analysis on the multiple intra-group correlation features, the multiple intra-group correlation features can be first fusion processed to fuse multiple groups of data into one group of data. For example, the multiple intra-group correlation features can be spliced to obtain a fusion feature, and the splicing manner is shown in the following formula (9):

[0066] (9)

[0067] Then, correlation analysis is performed on the fusion feature , and the component correlation features can be obtained.

[0068] The method provided in the embodiments of the present application can first fuse the multiple intra-group correlation features into one feature data set, and perform correlation analysis on the fused feature data set, thereby improving the convenience of correlation analysis on the multiple intra-group correlation features.

[0069] In one embodiment, in order to improve the correlation analysis efficiency and shorten the acquisition time of the fusion feature, the multiple intra-group correlation features can be dimensionally reduced before fusion processing, and the embodiment includes the steps shown as follows: Figure 3

[0070] Step 301, dimensionally reducing the intra-group correlation features to obtain the dimensionally reduced features corresponding to the intra-group correlation features.

[0071] Step 302, fusion processing is performed on the dimensionally reduced features to obtain a fusion feature.

[0072] In the embodiments of the present application, a fully connected layer can be configured to dimensionally reduce the intra-group correlation features to reduce the data amount of the intra-group correlation features. Further, in order to ensure that the data amount of the intra-group correlation features after dimensionally reduction is small enough or within a preset data amount range, the multiple intra-group correlation features can be processed by multiple fully connected layers, for example, the multiple intra-group correlation features are processed by two fully connected layers to obtain the dimensionally reduced features corresponding to the intra-group correlation features. Then, the dimensionally reduced features are fusion processed, for example, spliced, to obtain a fusion feature.

[0073] ​The method provided in the embodiments of the present application first performs dimension reduction processing on the intra-group correlation features before performing fusion processing on the intra-group correlation features, effectively reducing the data amount of the intra-group correlation features, i.e., reducing the data amount of the fused features, and effectively improving the correlation analysis efficiency of the fused features.

[0074] In one embodiment, to ensure the effectiveness of the feature data in the dimension-reduced features, difference analysis can be performed on the dimension-reduced features and the data before dimension reduction. If the difference is large, the intra-group correlation features are reprocessed for dimension reduction until the difference meets the expectation. The embodiment includes the steps as shown in Figure 4

[0075] Step 401: determining a total loss value of the plurality of omics data sets according to the dimension-reduced features and the omics data set corresponding to the dimension-reduced features.

[0076] Step 402: adjusting the dimension reduction parameter according to the total loss value, and performing dimension reduction processing on the intra-group correlation features according to the adjusted dimension reduction parameter until the total loss value meets a first preset loss condition.

[0077] The first preset loss condition can be that the total loss value is less than or equal to a first preset loss threshold.

[0078] In the embodiments of the present application, difference analysis can be performed on the dimension-reduced features and the data before dimension reduction (for example, the intra-group correlation features corresponding to the dimension-reduced features or the omics data set), and the total loss value of the intra-group correlation features of the plurality of groups is determined, as shown in the following formula (10):

[0079] (10)

[0080] wherein, represents the data before dimension reduction; represents the dimension-reduced features; F represents a norm; represents the total loss value, which can represent the difference between the dimension-reduced features and the data before dimension reduction.

[0081] If the total loss value is greater than the first preset loss threshold, the parameters in the fully connected layer, i.e., the dimension reduction parameter, are adjusted, then the intra-group correlation features are reprocessed for dimension reduction according to the fully connected layer after parameter adjustment, and difference analysis is performed on the dimension-reduced features obtained after the re-dimension reduction processing and the data before dimension reduction until the difference meets the expectation, i.e., the total loss value is less than or equal to the first preset loss threshold.

[0082] ​The method provided in this application embodiment can perform a difference analysis between the dimensionality-reduced features and the data before dimensionality reduction after dimensionality reduction processing of the intra-group correlation features. When the difference is large, the dimensionality reduction parameters are adjusted, and the intra-group correlation features are dimensionality-reduced again until the difference reaches the expected level. While reducing the amount of intra-group correlation feature data, the effectiveness of each feature data in the dimensionality-reduced features is guaranteed, and the effectiveness and accuracy of the fused features obtained based on the dimensionality-reduced feature fusion are improved.

[0083] In one embodiment, a training process for a pre-defined omics model is provided, which includes, as follows: Figure 5 The steps shown are as follows:

[0084] Step 501: Perform correlation analysis on multiple omics datasets to obtain inter-group correlation features, and perform clustering on the inter-group correlation features to obtain the predicted labels of molecules corresponding to each feature.

[0085] Step 502: Determine the loss value of the preset omics model based on the predicted labels and the labeled labels.

[0086] Step 503: Adjust the model parameters of the preset omics model according to the loss value until the loss value meets the second preset loss condition.

[0087] The second preset loss condition can be that the loss value is less than or equal to the second preset loss threshold.

[0088] In this embodiment, association analysis is performed on multiple omics datasets (i.e., multi-omics data) of the collected target object to obtain inter-group association features. The association analysis process is the same as in the above embodiment, and will not be repeated here.

[0089] After obtaining the inter-group association features, each feature data in these features can be used as training samples, and each feature data can be labeled to obtain labeled tags. Clustering the inter-group association features yields predicted labels for each feature data. For example, the K-means algorithm can be used to cluster the inter-group association features. Based on the training samples, labeled tags, and predicted labels, a predefined omics model can be trained.

[0090] Then, a difference analysis is performed on the predicted labels and the labeled labels to obtain the loss value of the preset omics model, as shown in equations (11)-(12):

[0091] (11)

[0092] (12)

[0093] in, Let p represent the number of samples in the training set, and p-th sample. represents a true label of the pth sample, represents a predicted label of the pth sample, represents a cross-entropy loss, represents a mean square error loss, a is a hyperparameter.

[0094] If the loss value L is greater than the second preset loss threshold, the inter-group associated features are re-clustered to obtain updated predicted labels, and an updated preset omics model is trained according to the training samples, the labeled labels and the updated predicted labels (this step completes the adjustment of the model parameters of the preset omics model). The loss value corresponding to the updated preset omics model is calculated until the loss value is less than or equal to the second preset loss threshold, and the training of the preset omics model is completed.

[0095] The foregoing embodiments introduce a scheme for determining the influence of each feature data in the inter-group associated features on the preset omics model. In another embodiment, the influence of each feature data on the preset omics model can be determined by changes in the performance of the preset omics model. For example, the "determining of the influence value of each feature in the inter-group associated features on the preset omics model" involves the steps as shown in Figure 6

[0096] Step 601: sequentially adjust each feature data in the inter-group associated features to a preset value to obtain updated inter-group associated features.

[0097] Step 602: training the preset omics model according to the updated inter-group associated features to obtain an updated omics model.

[0098] Step 603: determining the influence value of each feature data according to the performance index difference value between the preset omics model and the updated omics model.

[0099] The preset value can be 0.

[0100] In the embodiments of the present application, each feature data of the inter-group associated features is set to 0, and the remaining feature data remains unchanged, thereby obtaining the updated inter-group associated features corresponding to each feature data. Further, the updated preset omics model is retrained based on the updated inter-group associated features, thereby obtaining the updated preset omics model corresponding to each feature data.

[0101] Then, the performance parameters of each updated preset omics model are determined, and the performance index difference value is calculated based on the performance parameters of the preset omics model, thereby obtaining the performance index difference value corresponding to each feature data. The performance index difference value can represent the influence of the feature data on the preset omics model, and therefore the performance index difference value can be used as the influence value of the feature data.

[0102] ​The method provided by the embodiments of the present application can determine the influence value of each feature data of the inter-group correlation feature on the omics data model through performance changes, and then determine important target molecules in the omics molecules corresponding to the multi-omics data based on the influence value, thereby improving the accuracy of molecular detection of the to-be-detected object.

[0103] In summary, the molecular detection method (i.e., Omicsformer model) provided by the embodiments of the present application can be as shown in Figure 7 Further, the classification effect of the preset omics model is tested, as shown in Figure 8 The experimental results of the method described in the embodiments of the present application, (7) k - nearest neighbor (KNN) classifier, support vector machine SVM classifier, fully connected neural network classifier, classification based on Lasso regularization, multi-omics graph convolutional network (Multi-Omics Graph Convolutional Networks, MOGONET), multi-omics variational autoencoder (Multi-Omics Variational Autoencoders, MOVE) and DeepIMV are compared respectively.

[0104] Among them, the KNN method is to connect the multi-omics data and use the KNN algorithm for classification. The SVM method is to classify the multi-omics data by using the SVM algorithm. The fully connected neural network classifier method uses a fully connected deep network architecture and uses cross-entropy loss for training. The classification method based on Lasso regularization uses linear regression and Lasso regularization for classification, and assigns the highest probability prediction value for each sample as the final class prediction label. The MOGONET method uses graph convolutional network to perform specific and cross-omics learning in multi-omics data classification. The MOVE method uses variational autoencoder to explore the correlation between omics layers and evaluate the effect of drug treatment. The DeepIMV method realizes an information bottleneck framework, focusing on the edge and joint representation of omics to obtain the best performance.

[0105] In order to evaluate the performance of the above-mentioned models, three key performance indicators, accuracy (i.e., ACC), F1 score (F1) and purity, are compared. Each method is executed multiple times (for example, 10 times), and the mean and variance of the results show that the new model based on Omicsformer is superior to other state-of-the-art methods in multi-omics supervised task classification. The present application has significant improvement in the above three performance indicators, with an average increase of 8.3%, 6.0% and 8.3% respectively, indicating that the model based on Omicsformer provided by the embodiments of the present application has the advantage of capturing the complex interrelationships within and between groups in omics research.

[0106] Wherein, ACC represents the proportion of samples correctly classified to the total number of samples, the higher the value, the better the model performance; F1 score (F1) is a weighted average of precision and recall, precision represents the proportion of actual positive samples in the predicted positive samples, recall represents the proportion of actual positive samples in the predicted positive samples, F1 score considers both precision and recall, the value is generally between 0 and 1, the larger the value, the better the model performance; the closer the purity to 1, the better the clustering result.

[0107] It should be noted that although the operations of the method of the present application are described in a specific order in the accompanying drawings, this does not require or imply that the operations must be performed in this specific order, or that all of the shown operations must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart can change the order of execution. Additionally or alternatively, certain steps can be omitted, combined into one step, and / or divided into multiple steps.

[0108] Further reference is made to Figure 9 which shows an exemplary block diagram of an apparatus for molecular detection according to an embodiment of the present application.

[0109] In one embodiment, the molecular detection apparatus comprises: an acquisition module 701, an analysis module 702 and a determination module 703, wherein:

[0110] The acquisition module 701 is configured to acquire a plurality of sets of omics data of omics molecules.

[0111] The analysis module 702 is configured to, for each set of omics data, perform correlation analysis on the omics data in the set of omics data to obtain intra-group correlation features of the set of omics data; and perform correlation analysis on a plurality of intra-group correlation features to obtain inter-group correlation features of the plurality of intra-group correlation features.

[0112] The determination module 703 is configured to determine an influence value of each feature in the inter-group correlation features on a preset omics model, and determine a molecule corresponding to a feature whose influence value meets a preset condition as a target molecule; wherein the preset omics model is trained based on the inter-group correlation features of the plurality of sets of omics data.

[0113] In one embodiment, the analysis module 702 is specifically configured to perform fusion processing on the plurality of intra-group correlation features to obtain fusion features; and perform correlation analysis on feature data in the fusion features to obtain the inter-group correlation features.

[0114] In one embodiment, the analysis module 702 is further configured to perform dimensionality reduction processing on each intra-group correlation feature to obtain a dimensionality reduction feature corresponding to the intra-group correlation feature; and perform fusion processing on each dimensionality reduction feature to obtain the fusion features.

[0115] In an embodiment, the molecular detection apparatus further comprises a feature optimization module 704 configured to determine a total loss value of the plurality of omics data sets according to the reduced dimension features and the omics data sets corresponding to the reduced dimension features; adjust the reduced dimension parameters according to the total loss value, and perform reduced dimension processing on the intra-group correlation features according to the adjusted reduced dimension parameters until the total loss value meets a first preset loss condition.

[0116] In an embodiment, the training process of the preset omics model comprises: performing correlation analysis on the plurality of omics data sets to obtain inter-group correlation features, and performing clustering processing on the inter-group correlation features to obtain predicted labels of the molecules corresponding to each feature; determining a loss value of the preset omics model according to the predicted labels and the annotated labels; adjusting model parameters of the preset omics model according to the loss value until the loss value meets a second preset loss condition.

[0117] In an embodiment, the determining module 703 is specifically configured to adjust each feature data in the inter-group correlation features to a preset value in sequence to obtain updated inter-group correlation features; train the preset omics model according to the updated inter-group correlation features to obtain an updated omics model; and determine an influence value of each feature data according to a performance index difference value between the preset omics model and the updated omics model.

[0118] It should be understood that the units or modules described in the molecular detection apparatus correspond to the respective steps in the method described above. Therefore, the operations and features described above for the method are also applicable to the molecular detection apparatus and the units contained therein, and will not be described here again. The molecular detection apparatus can be pre- implemented in a browser or other secure application of an electronic device, or can be loaded into the browser or secure application of the electronic device through downloading or the like. The corresponding units in the molecular detection apparatus can cooperate with the units in the electronic device to realize the schemes of the embodiments of the present application. Figure 1 The respective steps in the method described above correspond to the units or modules described in the molecular detection apparatus. Therefore, the operations and features described above for the method are also applicable to the molecular detection apparatus and the units contained therein, and will not be described here again. The molecular detection apparatus can be pre- implemented in a browser or other secure application of an electronic device, or can be loaded into the browser or secure application of the electronic device through downloading or the like. The corresponding units in the molecular detection apparatus can cooperate with the units in the electronic device to realize the schemes of the embodiments of the present application.

[0119] The following refers to the accompanying drawings, which illustrate Figure 10 which shows a structural schematic diagram of a computer system 800 suitable for realizing the terminal device or server of the embodiments of the present application.

[0120] As shown in FIG. 8, the computer system 800 can include a processor 801, a memory 802, a storage 803, an input device 804, an output device 805 and a communication interface 806. Figure 10As shown, the computer system 800 includes a central processing unit (CPU) 801 which can perform various appropriate actions and processes according to programs stored in a read only memory (ROM) 802 or loaded from a storage section 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the system 800 are also stored. The CPU 801, the ROM 802, and the RAM 803 are connected to each other through a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0121] Connected to the I / O interface 805 are an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the I / O interface 805 as necessary. A removable recording medium 811 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 810 as necessary, so that a computer program read therefrom is installed into the storage section 808 as necessary.

[0122] In particular, the processes described above in reference to the Figure 1 application can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program tangibly embodied on a machine readable medium, the computer program comprising program code for executing the methods described above. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 809 and / or installed from the removable recording medium 811. Figure 1

[0123] ​It should be noted that the computer-readable medium shown in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0124] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0125] The units or modules described in the embodiments of the present application can be implemented in the form of software or in the form of hardware. The described units or modules can also be arranged in a processor. In some cases, the names of the units or modules do not constitute a limitation on the units or modules themselves.

[0126] As another aspect, the present application also provides a computer readable storage medium, which can be included in the computer device described in the above embodiments, or can exist separately and not be assembled into the computer device. The computer readable storage medium stores one or more programs, when the programs are used by one or more processors to execute the methods described in the present application. For example, the computer readable storage medium can store the programs for executing Figure 1 each step of the method.

[0127] The embodiments of the present application provide a computer program product, which includes instructions that, when executed, cause the method described in the embodiments of the present application to be performed. For example, the computer program product can store the programs for executing Figure 1 each step of the method.

[0128] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0129] The above description is merely preferred embodiments of the present application and a description of the principles of the technology used. Those skilled in the art should understand that the scope of the application involved in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and also covers other technical solutions formed by any combination of the above technical features or equivalent features without departing from the inventive concept. For example, the above features can be replaced with technical features disclosed in the present application (but not limited to) having similar functions to form technical solutions.

Claims

1. A molecular detection method, characterized in that, The method comprises obtaining a plurality of omics data sets of omics molecules of substances; each omics molecule of a substance corresponds to an omics data set; performing correlation analysis on omics data in the omics data set corresponding to each substance to obtain intra-group correlation features of the omics data set; performing correlation analysis on a plurality of intra-group correlation features to obtain inter-group correlation features of the plurality of intra-group correlation features; sequentially adjusting each feature data in the inter-group correlation features to a preset value to obtain updated inter-group correlation features; training a preset omics model according to the updated inter-group correlation features to obtain an updated omics model; determining an influence value of each feature data according to a performance index difference value between the preset omics model and the updated omics model, and determining a target molecule corresponding to a feature data meeting a preset condition as a target molecule; wherein the preset omics model is trained based on the inter-group correlation features of the plurality of omics data sets; and the preset value is 0.

2. The method of claim 1, wherein, The method further comprises: determining a total loss value of the plurality of omics data sets according to the dimension-reduced features and the omics data sets corresponding to the dimension-reduced features; adjusting a dimension-reduction parameter according to the total loss value, and performing dimension-reduction processing on each intra-group correlation feature according to the adjusted dimension-reduction parameter until the total loss value meets a first preset loss condition.

3. The method of claim 2, wherein, The training process of the preset omics model comprises: performing correlation analysis on the plurality of omics data sets to obtain the inter-group correlation features, and performing clustering processing on the inter-group correlation features to obtain a predicted label of a molecule corresponding to each feature; determining a loss value of the preset omics model according to the predicted label and a labeled label; 4. The method of claim 3, wherein, adjusting a model parameter of the preset omics model according to the loss value until the loss value meets a second preset loss condition. The device comprises: an obtaining module configured to obtain a plurality of omics data sets of omics molecules of substances; each omics molecule of a substance corresponds to an omics data set; 5. The method of claim 1, wherein, an analysis module configured to perform correlation analysis on omics data in the omics data set corresponding to each substance to obtain intra-group correlation features of the omics data set; and perform correlation analysis on a plurality of intra-group correlation features to obtain inter-group correlation features of the plurality of intra-group correlation features. ​ ​ ​ 6. A molecular detection device, characterized by, ​ ​ ​ The determining module is configured to sequentially adjust each feature data in the inter-group correlation features to a preset value to obtain updated inter-group correlation features; train a preset omics model according to the updated inter-group correlation features to obtain an updated omics model; determine an influence value of each feature data according to a performance index difference value between the preset omics model and the updated omics model, and determine a molecule corresponding to a feature meeting a preset condition as a target molecule; wherein the preset omics model is trained based on the inter-group correlation features of the plurality of omics data sets; and the preset value is 0. 7.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-6 when the computer program is executed by the processor. The processor, when executing the computer program, implements the steps of the method of any one of claims 1 to 5.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 5. The computer program, when executed by the processor, implements the steps of the method of any one of claims 1 to 5.

Images