WO2024239598A1 - 蛋白质鉴定方法、电子设备及存储介质 - Google Patents

蛋白质鉴定方法、电子设备及存储介质 Download PDF

Info

Publication number
WO2024239598A1
WO2024239598A1 PCT/CN2023/137165 CN2023137165W WO2024239598A1 WO 2024239598 A1 WO2024239598 A1 WO 2024239598A1 CN 2023137165 W CN2023137165 W CN 2023137165W WO 2024239598 A1 WO2024239598 A1 WO 2024239598A1
Authority
WO
WIPO (PCT)
Prior art keywords
protein
category
prediction
training
protein sequence
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2023/137165
Other languages
English (en)
French (fr)
Inventor
余函
罗小舟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen Institute of Advanced Technology of CAS
Original Assignee
Shenzhen Institute of Advanced Technology of CAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shenzhen Institute of Advanced Technology of CAS filed Critical Shenzhen Institute of Advanced Technology of CAS
Publication of WO2024239598A1 publication Critical patent/WO2024239598A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B30/00—ICT specially adapted for sequence analysis involving nucleotides or amino acids
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
    • G16B40/00—ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
    • G16B40/20—Supervised data analysis
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Definitions

  • the present application relates to the biological field, and in particular to a protein identification method, electronic device and storage medium.
  • the present application provides a protein identification method, which includes: obtaining a protein sequence, using multiple pre-trained self-supervised models to extract information from the protein sequence to obtain multiple embedding vectors of the protein sequence, generating a target probability vector for the protein sequence based on the multiple embedding vectors and multiple pre-trained classification models, and predicting the target probability vector based on the pre-trained prediction model to obtain a prediction result of the protein category of the protein sequence.
  • the present application uses multiple self-supervised models to extract information from protein sequences. Since the multiple protein models are self-supervised models of multiple different types of network architectures, it can ensure that the multiple embedded vectors extracted are different levels of information in the protein sequence. In addition, the present application relies on multiple pre-trained self-supervised models to extract information from protein sequences, and does not rely on artificially designed features to extract information from protein sequences. Therefore, it can solve the technical problem of limited expression ability of artificially extracted features, and can also improve the speed of information extraction.
  • the present application inputs each embedded vector into each classification model, the output result of each embedded vector on each classification model can be obtained, and the output results of the multiple embedded vectors on the multiple classification models are spliced to obtain the target probability vector.
  • the target probability vector incorporates the information extracted by the multiple self-supervised models and the multiple classification models on the protein sequence, so the target probability vector can fully reflect the information contained in the protein sequence, and the target probability vector is predicted, which can improve the efficiency of protein sequence identification.
  • the method further comprises: obtaining a data set, dividing the data set to obtain a first training set and a second training set, wherein the first training set comprises a plurality of first protein sequences and a first category label corresponding to each first protein sequence, and the second training set comprises a plurality of second protein sequences and a second category label corresponding to each second protein sequence, and the first category
  • the label and the second category label both include preset protein categories, obtain multiple initial classifiers corresponding to the multiple classification models and a prediction learner corresponding to the prediction model, train each initial classifier according to any one of the multiple self-supervised models, the multiple first protein sequences and the first category label corresponding to each first protein sequence, and obtain the classification model corresponding to each initial classifier, combine any self-supervised model and any classification model to obtain multiple pairs of combined models, extract information from each second protein sequence according to each pair of combined models, and obtain a training probability vector corresponding to each second protein sequence, use the prediction learner to predict each training probability vector, and obtain a training prediction result
  • generating the prediction model based on the training prediction results, the second category label and the prediction learner includes: calculating a training loss value based on multiple training prediction results and the second category label corresponding to each training prediction result, adjusting parameters of the prediction learner based on the training loss value until the training loss value meets a preset condition, and obtaining the prediction model.
  • calculating the training loss value based on the multiple training prediction results and the second category label corresponding to each training prediction result includes: determining whether each training prediction result is correct based on the second category label corresponding to each training prediction result, and calculating the training loss value based on the correct training prediction results and the number of labels of the second category label.
  • generating the target probability vector of the protein sequence based on the multiple embedding vectors and the pre-trained multiple classification models includes: inputting each embedding vector into the multiple classification models to obtain multiple initial probability vectors for each embedding vector, and concatenating all the initial probability vectors of the multiple embedding vectors to obtain the target probability vector.
  • the protein category includes preset protein categories and non-preset protein categories
  • the predicting the target probability vector according to the pre-trained prediction model to obtain a prediction result of the protein category of the protein sequence includes: encoding the preset protein category and the non-preset protein category respectively to obtain a category vector, classifying the target probability vector based on the category vector, determining the predicted probability that the protein sequence belongs to the preset protein category, and determining the prediction result according to a comparison result of the predicted probability and a preset threshold.
  • determining the prediction result based on the comparison result of the prediction probability and the preset threshold includes: if the prediction probability is greater than or equal to the preset threshold, determining the prediction result is that the protein sequence belongs to the preset protein category, or, if the prediction probability is less than the preset threshold, determining the prediction result is that the protein sequence does not belong to the preset protein category.
  • the predetermined protein classes include thermophilic proteins and psychrophilic proteins.
  • the present application provides a protein identification device that runs on an electronic device, the device comprising: an acquisition unit for acquiring a protein sequence, an extraction unit for extracting information from the protein sequence using a plurality of pre-trained self-supervised models to obtain a plurality of embedding vectors of the protein sequence, a generation unit for generating a target probability vector of the protein sequence based on the plurality of embedding vectors and a plurality of pre-trained classification models, and a prediction unit for predicting the target probability vector based on a pre-trained prediction model to obtain a prediction result of the protein category of the protein sequence.
  • the present application provides an electronic device, comprising: a memory storing at least one instruction; and a processor executing the at least one instruction to implement the protein identification method.
  • the present application provides a computer-readable storage medium, wherein at least one instruction is stored in the computer-readable storage medium, and the at least one instruction is executed by a processor in an electronic device to implement the protein identification method.
  • FIG. 1 is a structural diagram of an electronic device provided in an embodiment of the present application.
  • FIG. 2 is a flow chart of a protein identification method provided in an embodiment of the present application.
  • FIG3 is a flowchart of a method for generating prediction results provided in an embodiment of the present application.
  • FIG4 is a flowchart of a method for generating multiple self-supervisory models, multiple classification models, and a prediction model provided in an embodiment of the present application.
  • FIG. 5 is a functional module diagram of a protein identification device provided in one embodiment of the present application.
  • words such as “exemplary” or “for example” are used to indicate examples, illustrations or descriptions. Any embodiment or design described as “exemplary” or “for example” in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as “exemplary” or “for example” is intended to present related concepts in a specific way.
  • thermostable proteins are a class of proteins that can maintain their structure and function under high temperature conditions.
  • Thermophilic proteins are usually derived from thermophilic bacteria, especially archaea. Due to the feasibility of thermophilic proteins in enhancing protein stability, enhancing enzyme activity and being suitable for high-temperature industrial processes, they have been widely used in various fields such as biotechnology, food processing and pharmaceuticals.
  • thermostable cellulases from thermophilic bacteria have been used to efficiently convert lignocellulosic biomass into biofuels.
  • identification and screening of thermophilic proteins is a laborious and tedious process. In order to accelerate the development of related fields and better understand the mechanism of thermophilic proteins, it is crucial to develop high-throughput screening methods to achieve rapid identification of thermophilic proteins.
  • thermophilic proteins Many related computational methods have been developed to identify thermophilic proteins. However, these methods often rely on complex artificially designed features to extract key information of protein sequences, and then identify protein sequences based on the extracted key information. However, artificially designed features often have limited representation capabilities, which makes it difficult for the extracted key information to fully reflect the information contained in the protein sequence, resulting in The identification efficiency of protein sequences is low.
  • the present application provides a protein identification method, an electronic device and a storage medium.
  • the present application is described in detail below in conjunction with the accompanying drawings and specific embodiments.
  • the protein identification method provided in the embodiments of the present application can be applied to one or more electronic devices.
  • the electronic device 1 includes, but is not limited to, a memory 12, a processor 13, and a computer program stored in the memory 12 and executable on the processor 13, such as a protein identification program.
  • the electronic device 1 is a device that can automatically calculate parameter values and/or process information according to pre-set or stored instructions. Its hardware includes, but is not limited to: microprocessors, application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), digital signal processors (DSP), embedded devices, etc.
  • ASIC application specific integrated circuits
  • FPGA field-programmable gate arrays
  • DSP digital signal processors
  • the electronic device 1 may be any electronic product that can perform human-computer interaction with a user, such as a personal computer, a server, a tablet computer, a smart phone, etc.
  • the electronic device 1 may also include a network device and/or a user device.
  • the network device includes, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud consisting of a large number of hosts or network servers based on cloud computing.
  • Figure 1 is only an example of the electronic device 1 and does not constitute a limitation on the electronic device 1. It may include more or fewer components than shown in the figure, or combine certain components, or different components.
  • the electronic device 1 may also include input and output devices, network access devices, buses, etc.
  • the network where the electronic device 1 is located includes but is not limited to: the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.
  • the Internet includes but is not limited to: the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN), etc.
  • VPN virtual private network
  • FIG2 it is a flow chart of a protein identification method provided by an embodiment of the present application. According to different requirements, the order of each step in the flow chart can be adjusted according to actual requirements, and some steps can be omitted.
  • the execution subject of the method is an electronic device, such as the electronic device 1 shown in FIG1 .
  • the protein sequence refers to a protein sequence whose category needs to be identified.
  • the electronic device can obtain the protein sequence from a data set.
  • the data set can be part of or all of the data in the Uniprot database, the BRENDA database, and a common database containing protein sequences.
  • the protein sequence may be a sequence in a data set whose category needs to be identified.
  • the protein sequence can be obtained by other means, and the present application does not limit the protein sequence.
  • the multiple self-supervised models include, but are not limited to: SeqVec model, ProtCNN model, ProtTrans model and CPCProt model.
  • the multiple self-supervised models are pre-trained models to convergence.
  • the multiple self-supervisory models may also be other models, and the present application does not limit the multiple self-supervisory models.
  • the electronic device uses a plurality of pre-trained self-supervised models to extract information from the protein sequence to obtain a plurality of embedding vectors of the protein sequence, including: the electronic device inputs each embedding vector into each self-supervised model to obtain an embedding vector corresponding to the output of each embedding vector on each self-supervised model.
  • the information extraction process of each protein model for the protein sequence may refer to the process of the model performing representation learning on the protein sequence in the related art to obtain the protein embedded representation.
  • the dimension and information amount of each embedding vector are related to the self-supervised model corresponding to the embedding vector.
  • the multiple protein models are the SeqVec model, the ProtCNN model, the ProtTrans model and the CPCProt model
  • the multiple embedding vectors are 4 vectors
  • the dimension of the embedding vector extracted by the SeqVec model is 1024 dimensions
  • the dimension of the embedding vector extracted by the ProtCNN model is 1100 dimensions
  • the dimension of the embedding vector extracted by the ProtTrans model is 1024 dimensions
  • the dimension of the embedding vector extracted by the CPCProt model is 1536 dimensions.
  • multiple self-supervised models are used to extract features from the protein sequence, instead of relying on manually extracted features to extract information from the protein sequence, thereby solving the technical problem of limited expressive power of manually extracted features and improving the speed of information extraction. Since the architecture of each self-supervised model is different, embedding vectors of different levels and dimensions in the protein sequence can be extracted.
  • the number of the multiple self-supervisory models may be greater.
  • the multiple classification models include, but are not limited to: Random Forest model (RF), Adaptive Boosting model (AdaBoost, AB), guided clustering model or bagging model (Bagging, BA), gradient boosting decision tree model XGBoost (eXtremeGradient Boosting, XGB) and (LightGBM, LGB), etc.
  • RF Random Forest model
  • AdaBoost Adaptive Boosting model
  • AB guided clustering model or bagging model
  • BA gradient boosting decision tree model
  • XGBoost eXtremeGradient Boosting, XGB
  • LightGBM LightGBM
  • the electronic device generates a target probability vector for the protein sequence based on the multiple embedding vectors and the pre-trained multiple classification models, including: the electronic device inputs each embedding vector into the multiple classification models to obtain multiple initial probability vectors for each embedding vector, and then the electronic device concatenates all the initial probability vectors of the multiple embedding vectors to obtain the target probability vector.
  • each classification model generating each embedding vector according to each input embedding vector can refer to the classification method or classification formula of each classification model in the relevant technology.
  • the multiple embedding vectors are 4 embedding vectors output by the SeqVec model, the ProtCNN model, the ProtTrans model, and the CPCProt model, and the multiple classification models are the AdaBoost model, the Bagging model, the XGBoost model, and the LightGBM model
  • the 4 embedding vectors are respectively input into the 5 classification models, 20 initial probability vectors are obtained. If the dimension of each initial probability vector is 1 dimension, the 20 1-dimensional initial probability vectors are concatenated to obtain a target probability vector with a dimension of 20 dimensions.
  • the multiple classification models may be other models, which are not limited in the present application.
  • the above examples of the multiple classification models are only examples and do not constitute a limitation on the multiple classification models.
  • the multiple classification models can be cascaded, and each embedding vector can be classified by a model integrated by the multiple classification models to obtain the target probability vector, which can reduce the overfitting problem of a single model.
  • the target probability vector since the target probability vector is obtained by splicing multiple initial probability vectors, the target probability vector incorporates the information in the multiple initial probability vectors, so that the target probability vector can fully reflect the information contained in the protein sequence.
  • the protein category includes a preset protein category and a non-preset protein category.
  • the preset protein category includes, but is not limited to, thermophilic protein and psychrophilic protein.
  • the non-preset protein category is a protein category other than the preset protein category. For example, if the preset protein category is thermophilic protein, the non-preset protein category may be non-thermophilic protein, or, if the preset protein category is psychrophilic protein, the non-preset protein category may be non-psychrophilic protein.
  • the prediction result includes that the protein sequence belongs to the preset protein category and that the protein sequence does not belong to the preset protein category.
  • the prediction model may be a machine learning model.
  • the prediction model may be a support vector machine model (SVM).
  • the target probability vector can fully reflect the information contained in the protein sequence, predicting the target probability vector can improve the accuracy of protein sequence identification.
  • the present application uses multiple self-supervised models to extract information from protein sequences. Since the multiple protein models are self-supervised models of multiple different types of network architectures, it can ensure that the multiple embedded vectors extracted are different levels of information in the protein sequence. In addition, the present application relies on multiple pre-trained self-supervised models to extract information from protein sequences, and does not rely on artificially designed features to extract information from protein sequences. Therefore, it can solve the technical problem of limited expression ability of artificially extracted features, and can also improve the speed of information extraction.
  • the present application inputs each embedded vector into each classification model, the output result of each embedded vector on each classification model can be obtained, and the output results of the multiple embedded vectors on the multiple classification models are spliced to obtain the target probability vector.
  • the target probability vector incorporates the information extracted by the multiple self-supervised models and the multiple classification models on the protein sequence, so the target probability vector can fully reflect the information contained in the protein sequence, and the target probability vector is predicted, which can improve the efficiency of protein sequence identification.
  • the electronic device inputs the target probability vector into the prediction model to obtain a prediction result of the protein category of the protein sequence.
  • FIG. 3 is a flowchart of a method for generating prediction results provided by an embodiment of the present application, comprising the following steps:
  • the electronic device encodes the preset protein category and the non-preset protein category respectively to obtain a category vector.
  • the preset protein category includes, but is not limited to: thermophilic protein and psychrophilic protein.
  • the non-preset protein category is a category other than the preset protein category.
  • the preset protein category is thermophilic protein
  • the non-preset protein category may be non-thermophilic protein
  • the preset protein category is psychrophilic protein
  • the non-preset protein category may be non-psychrophilic protein.
  • the electronic device may perform one-hot encoding on the preset protein category and the non-preset protein category to obtain the category vector. For example, the electronic device may encode the preset protein category as 1 and encode the non-preset protein category as 0 to obtain a category vector consisting of 1 and 0.
  • the electronic device may encode the preset protein category and the non-preset protein category in other ways, and the present application does not limit the encoding method.
  • S142 The electronic device classifies the target probability vector based on the category vector to determine a predicted probability that the protein sequence belongs to the preset protein category.
  • the electronic device classifies the target probability vector based on the category vector, and the method for determining the predicted probability that the protein sequence belongs to the preset protein category can refer to the classification formula corresponding to the sigmoid function or the softmax function in the relevant technology.
  • the electronic device determines the prediction result according to a comparison result between the prediction probability and a preset threshold.
  • the electronic device determines the prediction result based on the comparison result of the prediction probability and the preset threshold, including: if the prediction probability is greater than or equal to the preset threshold, the electronic device determines the prediction result as the protein sequence belongs to the preset protein category, or, if the prediction probability is less than the preset threshold, the electronic device determines the prediction result as the protein sequence does not belong to the preset protein category, or the protein sequence belongs to the non-preset protein category.
  • the electronic device determines that the prediction result is that the protein sequence belongs to the preset protein category
  • the electronic device outputs first preset data to indicate that the protein sequence belongs to the preset protein category, or, if the electronic device determines that the prediction result is that the protein sequence does not belong to the preset protein category (the electronic device determines that the prediction result is that the protein sequence belongs to the non-preset protein category), the electronic device outputs second preset data to indicate that the protein sequence does not belong to the preset protein category.
  • the first preset data and the second preset data may be in the form of a numerical value, a letter, or a combination of a numerical value and a letter.
  • the first preset data and the second preset data may be set arbitrarily, and this application does not limit this. For example, when the first preset data and the second preset data are both numerical values, the first preset value may be 1, and the second preset value may be 0.
  • FIG4 it is a flowchart of a method for generating multiple self-supervisory models, multiple classification models, and a prediction model provided by an embodiment of the present application, comprising the following steps:
  • the electronic device acquires a data set, and divides the data set to obtain a first training set and a second training set.
  • the data set includes a plurality of protein sequences, each protein sequence in the data set has a corresponding category label, and the data set may include, but is not limited to: partial data or all data in the Uniprot database, the BRENDA database, and common databases containing protein sequences.
  • the first training set includes a plurality of first protein sequences and a first category label corresponding to each first protein sequence
  • the second training set includes a plurality of second protein sequences and a second category label corresponding to each second protein sequence
  • the first category label and the second category label both include preset protein categories.
  • each first protein sequence has a corresponding first category label
  • each second protein sequence has a corresponding second category label
  • the first category label is used to indicate the category of each first protein sequence
  • the second category label is used to indicate the category of each second protein sequence
  • the electronic device can set the way to divide the data set by itself, and this application does not limit this.
  • the electronic device can use 80% of the data in the data set as the first training set, and the category label of the 80% of the data as the first category label, and then the electronic device uses 20% of the data in the data set as the second training set, and the category label of the 20% of the data as the second category label.
  • the above example is only an example of a way to divide the data set, and does not constitute a limitation on the way to divide the data set.
  • the electronic device obtains a plurality of initial classifiers corresponding to the plurality of classification models and a prediction learner corresponding to the prediction model.
  • the multiple initial classifiers may include a classifier corresponding to the random forest model Random Forest, a classifier corresponding to the adaptive boosting model AdaBoost, a classifier corresponding to the bagging model Bagging, a classifier corresponding to the gradient boosting decision tree model XGBoost, and a classifier corresponding to the gradient boosting decision tree model LightGBM, and the prediction learner may be a learner of a support vector machine model SVM.
  • the electronic device trains each initial classifier according to any one of the multiple self-supervised models, the multiple first protein sequences, and the first category label corresponding to each first protein sequence to obtain a classification model corresponding to each initial classifier.
  • the electronic device inputs each first protein sequence into each self-supervised model to obtain an initial embedding vector corresponding to the output of each first protein sequence on each self-supervised model, wherein the multiple initial embedding vectors corresponding to each first protein sequence correspond to the first category label of the first protein sequence, and then the electronic device inputs the multiple initial embedding vectors corresponding to each first protein sequence into the multiple initial classifiers respectively to obtain a classification vector corresponding to the output of each initial embedding vector of each first protein sequence on each initial classifier, splices the multiple classification vectors corresponding to each first protein sequence to obtain a splicing probability vector corresponding to each first protein sequence, and the electronic device predicts the splicing probability vector corresponding to each first protein sequence to obtain a predicted category of each first protein sequence, and
  • the first category label calculates an initial loss value, and adjusts the parameters of the multiple initial classifiers according to the initial loss value until the initial loss value meets the configuration condition, thereby obtaining a classification model corresponding to the initial classifier
  • the multiple initial embedding vectors are generated in a manner substantially the same as the multiple embedding vectors described above, and the splicing probability vector is generated in a manner substantially the same as the target probability vector described above, so this application will not repeat the description.
  • the electronic device predicts each splicing probability vector with reference to the prediction process of the target probability vector described below, the calculation method of the initial loss value can refer to the calculation method of the training loss value described below, and the configuration conditions are substantially the same as the preset conditions described below.
  • the electronic device combines any self-supervision model and any classification model to obtain multiple pairs of combined models.
  • the multiple self-supervisory models are represented by A, B, C and D respectively
  • the multiple classification models are represented by a, b, c, d and e respectively.
  • the electronic device combines any self-supervisory model and any classification model to obtain multiple pairs of combined models as shown in Table 1.
  • the electronic device extracts information from each second protein sequence according to each pair of combined models to obtain a training probability vector corresponding to each second protein sequence.
  • the generation method of each training probability vector is substantially the same as the generation method of the target probability vector, and thus the description will not be repeated in this application.
  • the electronic device uses the prediction learner to predict each training probability vector to obtain a training prediction result for each second protein sequence.
  • the training prediction results and the prediction results of the protein sequence are generated in substantially the same manner, and thus will not be described again in this application.
  • the electronic device generates the prediction model according to the training prediction result, the second category label and the prediction learner.
  • the electronic device generates the prediction model based on the training prediction results, the second category label and the prediction learner, including: the electronic device calculates the training loss value based on the multiple training prediction results and the second category label corresponding to each training prediction result, adjusts the parameters of the prediction learner according to the training loss value until the training loss value meets the preset conditions, and obtains the prediction model.
  • the electronic device can be used to analyze the multiple self-supervisory models, the multiple initial classifiers and The parameters such as the weight and bias of the prediction learner are adjusted.
  • the preset condition can be set by oneself, and this application does not limit this.
  • the preset condition corresponds to the training prediction result.
  • the preset condition can be that the training loss value drops to a preset interval, the training loss value drops to a minimum, or the training loss value no longer changes, wherein the preset interval can be set by oneself, and this application does not limit this.
  • the electronic device calculates the training loss value based on the multiple training prediction results and the second category label corresponding to each training prediction result, including: the electronic device determines whether each training prediction result is correct based on the second category label corresponding to each training prediction result, and then the electronic device calculates the training loss value based on the correct training prediction results and the number of labels of the second category label.
  • the electronic device can compare each training prediction result with the corresponding second category label. If the protein indicated by each training prediction result is the same as the corresponding second category label, the electronic device determines that each training prediction result is correct; or, if each training prediction result is different from the corresponding second category label, the electronic device determines that each training prediction result is wrong.
  • the electronic device determines that the training prediction result of any second protein sequence is correct.
  • the training prediction result of any second protein sequence is that the any second protein sequence belongs to the preset protein category
  • the electronic device determines that the training prediction result of any second protein sequence is wrong.
  • the electronic device uses an integration algorithm (such as a bagging algorithm or a boosting algorithm) to perform integration training on the multiple initial classifiers to obtain a model formed by cascading the multiple classification models.
  • an integration algorithm such as a bagging algorithm or a boosting algorithm
  • the electronic device calculates the correct number of correct training prediction results, and determines the ratio between the correct number and the number of labels as the training loss value.
  • the electronic device stops adjusting to obtain the multiple self-supervised models, the multiple classification models, and the prediction model.
  • the training loss value may also be calculated in other ways.
  • the electronic device may determine the ratio between the number of errors in the wrong training prediction results and the number of labels as the training loss value, and when the training loss value no longer decreases or decreases to a preset interval, the electronic device stops adjusting to obtain the multiple self-supervisory models, the multiple classification models, and the prediction model.
  • FIG5 it is a functional module diagram of a protein identification device provided in an embodiment of the present application.
  • the protein identification device 11 includes an acquisition unit 110, an extraction unit 111, a generation unit 112, and a prediction unit 113.
  • the module/unit referred to in the present application refers to a series of computer-readable instruction segments that can be acquired by the processor 13 in FIG1 and can perform fixed functions, which are stored in the memory 12 in FIG1 . In this embodiment, the functions of each module/unit will be described in detail in subsequent embodiments.
  • the acquisition unit 110 is used to acquire a protein sequence.
  • the extraction unit 111 is used to extract information from the protein sequence using multiple pre-trained self-supervised models to obtain multiple embedding vectors of the protein sequence.
  • the generating unit 112 is used to generate a target probability vector of the protein sequence according to the multiple embedding vectors and the multiple pre-trained classification models.
  • the generating unit 112 is further used to input each embedding vector into the multiple classification models to obtain multiple initial probability vectors for each embedding vector, and concatenate all the initial probability vectors of the multiple embedding vectors to obtain the target probability vector.
  • the prediction unit 113 is used to predict the target probability vector according to the pre-trained prediction model to obtain a prediction result of the protein category of the protein sequence.
  • the preset protein category includes thermophilic protein and psychrophilic protein.
  • the protein category includes a preset protein category and a non-preset protein category
  • the prediction unit 113 is further used to predict the target probability vector according to a pre-trained prediction model to obtain a prediction result of the protein category of the protein sequence, including: encoding the preset protein category and the non-preset protein category respectively to obtain a category vector, classifying the target probability vector based on the category vector, determining the predicted probability that the protein sequence belongs to the preset protein category, and determining the prediction result according to a comparison result of the predicted probability and a preset threshold.
  • the prediction unit 113 is also used to determine the prediction result based on the comparison result of the prediction probability and the preset threshold, including: if the prediction probability is greater than or equal to the preset threshold, determining that the prediction result is that the protein sequence belongs to the preset protein category, or, if the prediction probability is less than the preset threshold, determining that the prediction result is that the protein sequence does not belong to the preset protein category.
  • the generating unit 112 is further used to obtain a data set, divide the data set to obtain a first training set and a second training set, wherein the first training set includes multiple first protein sequences and a first category label corresponding to each first protein sequence, and the second training set includes multiple second protein sequences and a second category label corresponding to each second protein sequence, and the first category label and the second category label both include preset protein categories, obtain multiple initial classifiers corresponding to the multiple classification models and a prediction learner corresponding to the prediction model, train each initial classifier according to any one of the multiple self-supervised models, the multiple first protein sequences and the first category label corresponding to each first protein sequence to obtain a classification model corresponding to each initial classifier, combine any self-supervised model and any classification model to obtain multiple pairs of combined models, extract information from each second protein sequence according to each pair of combined models to obtain a training probability vector corresponding to each second protein sequence, use the prediction learner to predict each training probability vector to obtain a training prediction result for each second protein sequence, and generate
  • the generating unit 112 is further configured to generate the prediction model according to the training prediction results, the second category label and the prediction learner, including: calculating a training loss value according to a plurality of the training prediction results and the second category label corresponding to each training prediction result, adjusting parameters of the prediction learner according to the training loss value until the training loss value meets the prediction Set conditions and obtain the prediction model.
  • the generating unit 112 is further used to calculate the training loss value based on the multiple training prediction results and the second category label corresponding to each training prediction result, including: determining whether each training prediction result is correct based on the second category label corresponding to each training prediction result, and calculating the training loss value based on the correct training prediction results and the number of labels of the second category label.
  • the present application uses multiple self-supervised models to extract information from protein sequences. Since the multiple protein models are self-supervised models of multiple different types of network architectures, it can ensure that the multiple embedded vectors extracted are different levels of information in the protein sequence. In addition, the present application relies on multiple pre-trained self-supervised models to extract information from protein sequences, and does not rely on artificially designed features to extract information from protein sequences. Therefore, it can solve the technical problem of limited expression ability of artificially extracted features, and can also improve the speed of information extraction.
  • the present application inputs each embedded vector into each classification model, the output result of each embedded vector on each classification model can be obtained, and the output results of the multiple embedded vectors on the multiple classification models are spliced to obtain the target probability vector.
  • the target probability vector incorporates the information extracted by the multiple self-supervised models and the multiple classification models on the protein sequence, so the target probability vector can fully reflect the information contained in the protein sequence, and the target probability vector is predicted, which can improve the efficiency of protein sequence identification.
  • the processor 13 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • a general-purpose processor may be a microprocessor or any conventional processor, etc.
  • the processor 13 is the computing core and control center of the electronic device 1, and uses various interfaces and lines to connect various parts of the entire electronic device 1, and obtain the operating system of the electronic device 1 and various installed applications, program codes, etc.
  • the processor 13 obtains the operating system and various installed applications of the electronic device 1.
  • the processor 13 obtains the application to implement the steps in the above-mentioned various protein identification method embodiments, such as the steps shown in FIG. 2, FIG. 3 and FIG. 4.
  • the computer program may be divided into one or more modules/units, such as an acceleration unit, and one or more modules/units are stored in the memory 12 and acquired by the processor 13 to complete the present application.
  • One or more modules/units may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the acquisition process of the computer program in the electronic device 1.
  • the memory 12 can be used to store computer programs and/or modules.
  • the processor 13 implements various functions of the electronic device 1 by running or acquiring computer programs and/or modules stored in the memory 12, and calling data stored in the memory 12.
  • the memory 12 can mainly include a program storage area and a data storage area.
  • the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the server, etc.
  • the memory 12 can include non-volatile memory, such as a hard disk, a memory, a plug-in A hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device.
  • non-volatile memory such as a hard disk, a memory, a plug-in A hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device.
  • the memory 12 may be an external memory and/or an internal memory of the electronic device 1. Furthermore, the memory 12 may be a physical memory, such as a memory stick, a TF card (Trans-flash Card), and the like.
  • TF card Trans-flash Card
  • the module/unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.
  • the present application implements all or part of the processes in the above-mentioned embodiment method, and can also instruct the relevant hardware to complete it through a computer program.
  • the computer program can be stored in a computer-readable storage medium, and when the computer program is obtained by the processor, it can implement the steps of each of the above-mentioned method embodiments.
  • the computer program includes computer program code, which can be in source code form, object code form, accessible file or some intermediate form, etc.
  • Computer readable media can include: any entity or device that can carry computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM).
  • the memory 12 in the electronic device 1 stores multiple instructions to implement a protein identification method
  • the processor 13 can obtain multiple instructions to achieve: obtaining a protein sequence; using multiple pre-trained self-supervised models to extract information from the protein sequence to obtain multiple embedding vectors of the protein sequence; generating a target probability vector of the protein sequence based on the multiple embedding vectors and multiple pre-trained classification models; predicting the target probability vector based on the pre-trained prediction model to obtain a prediction result of the protein category of the protein sequence.
  • modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
  • each functional module in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional modules.

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • General Health & Medical Sciences (AREA)
  • Data Mining & Analysis (AREA)
  • Biophysics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Biotechnology (AREA)
  • Evolutionary Biology (AREA)
  • Theoretical Computer Science (AREA)
  • Spectroscopy & Molecular Physics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Artificial Intelligence (AREA)
  • Bioethics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Databases & Information Systems (AREA)
  • Epidemiology (AREA)
  • Evolutionary Computation (AREA)
  • Public Health (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

提供一种蛋白质鉴定方法、电子设备及存储介质。所述方法包括：获取蛋白质序列；使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量；根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量；根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果。利用上述方法，能够提高蛋白质序列的鉴定效率。

Description

蛋白质鉴定方法、电子设备及存储介质 技术领域
本申请涉及生物领域，尤其涉及一种蛋白质鉴定方法、电子设备及存储介质。
背景技术
目前的蛋白质鉴定方案中，依赖于人工设计的复杂特征提取蛋白质序列的关键信息，然后根据提取的关键信息实现蛋白质序列的鉴定。但是人工设计的特征往往表示能力有限，导致提取到的关键信息难以充分地反映蕴含在蛋白质序列内部的信息，造成对蛋白质序列的鉴定效率较低。
发明内容
鉴于以上内容，有必要提供一种蛋白质鉴定方法、电子设备及存储介质，解决了由于人工设计的特征在表达能力上的局限性导致蛋白质序列的鉴定效率较低的技术问题。
一方面，本申请提供一种蛋白质鉴定方法，所述方法包括：获取蛋白质序列，使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量，根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量，根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果。
由上述技术方案可知，本申请采用多个自监督模型对蛋白质序列进行信息提取，由于所述多个蛋白质模型为多个不同类型网络架构的自监督模型，因此，能够确保提取到的多个嵌入式向量为所述蛋白质序列中不同级别的信息。此外，本申请依靠预训练的多个自监督模型从蛋白质序列中提取信息，并不依赖于人工设计的特征从蛋白质序列中提取信息，因此能够解决的人工提取的特征的表达能力有限的技术问题，还能够提高信息提取的速度。由于本申请是将每个嵌入向量输入至每个分类模型中，因此可以得到每个嵌入向量在每个分类模型上的输出结果，将所述多个嵌入向量在所述多个分类模型上的输出结果进行拼接，得到所述目标概率向量，所述目标概率向量中融合了所述多个自监督模型以及所述多个分类模型对所述蛋白质序列进行提取的信息，因此所述目标概率向量能够充分地反映蕴含在所述蛋白质序列内部的信息，对所述目标概率向量进行预测，能够提高对蛋白质序列的鉴定效率。
在一些实施例中，所述方法还包括：获取数据集，对所述数据集进行划分，得到第一训练集以及第二训练集，其中，所述第一训练集包括多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签，所述第二训练集包括多个第二蛋白质序列以及每个第二蛋白质序列对应的第二类别标签，所述第一类别 标签以及所述第二类别标签均包括预设蛋白质类别，获取所述多个分类模型对应的多个初始分类器以及所述预测模型对应的预测学习器，根据所述多个自监督模型中的任意一个模型、所述多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签对每个初始分类器进行训练，得到每个初始分类器对应的分类模型，将任一自监督模型以及任一分类模型进行组合，得到多对组合模型，根据每对组合模型对每个第二蛋白质序列进行信息提取，得到每个第二蛋白质序列对应的训练概率向量，使用所述预测学习器对每个训练概率向量进行预测，得到每个第二蛋白质序列的训练预测结果，根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型。
在一些实施例中，所述根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型包括：根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值，根据所述训练损失值对所述预测学习器进行参数调整，直至所述训练损失值满足预设条件，得到所述预测模型。
在一些实施例中，所述根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值包括：根据每个训练预测结果对应的第二类别标签，确定每个训练预测结果是否正确，根据正确的训练预测结果以及所述第二类别标签的标签数量，计算所述训练损失值。
在一些实施例中，所述根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量包括：将每个嵌入向量输入至所述多个分类模型中，得到每个嵌入向量的多个初始概率向量，将所述多个嵌入向量的所有初始概率向量进行拼接，得到所述目标概率向量。
在一些实施例中，所述蛋白质类别包括预设蛋白质类别和非预设蛋白质类别，所述根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果，包括：对所述预设蛋白质类别和所述非预设蛋白质类别分别进行编码，得到类别向量，基于所述类别向量对所述目标概率向量进行分类，确定所述蛋白质序列属于所述预设蛋白质类别的预测概率，根据所述预测概率与预设阈值的比较结果确定所述预测结果。
在一些实施例中，所述根据所述预测概率与预设阈值的比较结果确定所述预测结果包括：若所述预测概率大于或者等于预设阈值，确定所述预测结果为所述蛋白质序列属于所述预设蛋白质类别，或者，若所述预测概率小于所述预设阈值，确定所述预测结果为所述蛋白质序列不属于所述预设蛋白质类别。
在一些实施例中，所述预设蛋白质类别包括嗜热蛋白和嗜冷蛋白。
另一方面，本申请提供一种蛋白质鉴定装置，运行于电子设备，所述装置包括：获取单元，用于获取蛋白质序列，提取单元，用于使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量，生成单元，用于根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量，预测单元，用于根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果。
另一方面，本申请提供一种电子设备，所述电子设备包括：存储器，存储至少一个指令；及处理器，执行所述至少一个指令以实现所述的蛋白质鉴定方法。
另一方面，本申请提供一种计算机可读存储介质，所述计算机可读存储介质中存储有至少一个指令，所述至少一个指令被电子设备中的处理器执行以实现所述的蛋白质鉴定方法。
附图说明
图1是本申请一实施例提供的电子设备的结构图。
图2是本申请一实施例提供的蛋白质鉴定方法的流程图。
图3是本申请一实施例提供的预测结果的生成方法的流程图。
图4是本申请一实施例提供的多个自监督模型、多个分类模型以及预测模型的生成方法的流程图。
图5是本申请一实施例提供的蛋白质鉴定装置的功能模块图。
具体实施方式
需要说明的是，本申请中“至少一个”是指一个或者多个，“多个”是指两个或多于两个。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B可以表示：单独存在A，同时存在A和B，单独存在B的情况，其中A，B可以是单数或者复数。本申请的说明书和权利要求书及附图中的术语“第一”、“第二”、“第三”、“第四”等(如果存在)是用于区别类似的对象，而不是用于描述特定的顺序或先后次序。
在本申请实施例中，“示例性的”或者“例如”等词用于表示作例子、例证或说明。本申请实施例中被描述为“示例性的”或者“例如”的任何实施例或设计方案不应被解释为比其它实施例或设计方案更优选或更具优势。确切而言，使用“示例性的”或者“例如”等词旨在以具体方式呈现相关概念。
嗜热蛋白，通常也被称为耐高温的蛋白，是一类在高温条件下仍然能够保持其结构和功能的蛋白质。嗜热蛋白质通常来自于嗜热菌，特别是古菌。由于嗜热蛋白在增强蛋白稳定性、增强酶活以及适用于高温工业过程的可行性，它们已经在生物技术、食品加工和制药等各个领域实现了广泛的应用。例如，来自嗜热菌的热稳定纤维素酶已被用于将木质纤维素生物质高效转化为生物燃料。然而，鉴定和筛选嗜热蛋白是一项费力且繁琐的过程。为了加快相关领域的发展和更好的理解嗜热蛋白的机制，开发高通量的筛选方法实现快速的嗜热蛋白的鉴定就至关重要。
很多相关计算方法被开发出来用于鉴定嗜热蛋白，然而，相关计算方法往往依赖于人工设计的复杂特征以提取蛋白质序列的关键信息，然后根据提取的关键信息实现对蛋白质序列的鉴定。但是人工设计的特征往往表示能力有限，导致提取到的关键信息难以充分地反映蕴含在蛋白质序列内部的信息，造成对 蛋白质序列的鉴定效率较低。
为了解决上述问题，本申请提供一种蛋白质鉴定方法、电子设备及存储介质。为了使本申请的目的、技术方案和优点更加清楚，下面结合附图和具体实施例对本申请进行详细描述。
本申请实施例提供的蛋白质鉴定方法可应用于一个或者多个电子设备中。
如图1所示，是本申请一实施例提供的电子设备的结构图。在图1中，电子设备1包括，但不限于，存储器12、处理器13，以及存储在存储器12中并可在处理器13上运行的计算机程序，例如蛋白质鉴定程序。
所述电子设备1是一种能够按照事先设定或存储的指令，自动进行参数值计算和/或信息处理的设备，其硬件包括，但不限于：微处理器、专用集成电路(Application Specific Integrated Circuit，ASIC)、可编程门阵列(Field-Programmable Gate Array，FPGA)、数字信号处理器(Digital Signal Processor，DSP)、嵌入式设备等。
所述电子设备1可以是任何一种可与用户进行人机交互的电子产品，例如，个人计算机、服务器、平板电脑、智能手机等。
所述电子设备1还可以包括网络设备和/或用户设备。其中，所述网络设备包括，但不限于单个网络服务器、多个网络服务器组成的服务器组或基于云计算(Cloud Computing)的由大量主机或网络服务器构成的云。图1仅仅是电子设备1的示例，并不构成对电子设备1的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如电子设备1还可以包括输入输出设备、网络接入设备、总线等。
所述电子设备1所处的网络包括但不限于：互联网、广域网、城域网、局域网、虚拟专用网络(Virtual Private Network，VPN)等。
如图2所示，是本申请一实施例提供的蛋白质鉴定方法的流程图。根据不同的需求，该流程图中各个步骤的顺序可以根据实际要求进行调整，某些步骤可以省略。所述方法的执行主体为电子设备，例如图1所示的电子设备1。
S11，获取蛋白质序列。
在一实施例中，所述蛋白质序列是指需要鉴定类别的蛋白质序列。所述电子设备可以从数据集中获取所述蛋白质序列。例如，所述数据集可以为Uniprot数据库、BRENDA数据库以及常见的包含蛋白质序列的数据库中的部分数据或者所有数据。
在一实施例中，所述蛋白质序列可以为需要鉴定类别的数据集中的序列。
在另一实施例中，所述蛋白质序列可以通过其它方式获取，本申请对所述蛋白质序列不作限制。
S12，使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量。
在一实施例中，所述多个自监督模型包括，但不限于：SeqVec模型、ProtCNN模型、ProtTrans模型和CPCProt模型。其中，所述多个自监督模型为预训练至收敛的模型。
在另一实施例中，所述多个自监督模型还可以为其它模型，本申请对所述多个自监督模型不作限制。
在一些实施例中，所述电子设备使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量包括：所述电子设备将每个嵌入向量输入至每个自监督模型中，得到每个嵌入向量在每个自监督模型上对应输出的嵌入向量。
其中，每个蛋白质模型对所述蛋白质序列的信息提取过程可参考相关技术中该模型对蛋白质序列进行表示学习，得到蛋白质嵌入式表示的过程。
在本实施例中，每个嵌入向量的维度以及信息量等信息与所述嵌入向量对应的自监督模型相关。例如，若所述多个蛋白质模型为所述SeqVec模型、所述ProtCNN模型、所述ProtTrans模型和所述CPCProt模型，所述多个嵌入向量为4个向量，所述SeqVec模型提取到的嵌入向量的维度为1024维、所述ProtCNN模型提取到的嵌入向量的维度为1100维、所述ProtTrans模型提取到的嵌入向量的维度为1024维以及所述CPCProt模型提取到的嵌入向量的维度为1536维。
在本实施例中，使用多个自监督模型对所述蛋白质序列进行特征提取，而并不依赖人工提取的特征从蛋白质序列中提取信息，因此能够解决的人工提取的特征的表达能力有限的技术问题，还能够提高信息的提取速度。由于每个自监督模型的架构不相同，因此能够提取到所述蛋白质序列中不同级别以及不同维度的嵌入向量。
在本申请的其它实施例中，为了使得所述蛋白质序列具有更多的嵌入表示，所述多个自监督模型的数量可以更多。
S13，根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量。
在一些实施例中，所述多个分类模型包括，但不限于：随机森林模型(Random Forest，RF)、自适应增强模型(AdaBoost，AB)、引导聚集模型或者装袋模型(Bagging，BA)、梯度提升决策树模型XGBoost(eXtremeGradient Boosting，XGB)和(LightGBM，LGB)等。
在一实施例中，所述电子设备根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量包括：所述电子设备将每个嵌入向量输入至所述多个分类模型中，得到每个嵌入向量的多个初始概率向量，然后所述电子设备将所述多个嵌入向量的所有初始概率向量进行拼接，得到所述目标概率向量。
其中，由于所述多个分类模型为不同的模型，因此每个分类模型根据输入的每个嵌入向量生成每个嵌入向量的过程可参考相关技术中每个分类模型的分类方法或者分类公式。
例如，承接上述实施例，若所述多个嵌入向量为所述SeqVec模型、所述ProtCNN模型、所述ProtTrans模型和所述CPCProt模型输出的4个嵌入向量，所述多个分类模型为所述AdaBoost模型、所述Bagging模型、所述XGBoost模型和所述LightGBM模型，将所述4个嵌入向量分别输入至5个分类模型中， 得到20个初始概率向量，若每个初始概率向量的维度均为1维，将20个1维的初始概率向量进行拼接，得到维度为20维的目标概率向量。
在本申请的另一实施例中，所述多个分类模型可以为其它模型，本申请对所述不作限制。上述对所述多个分类模型的举例仅为示例，而并不构成对所述多个分类模型的限制。
在本实施例中，可以对所述多个分类模型进行级联，通过由所述多个分类模型集成的模型对每个嵌入向量进行分类，得到所述目标概率向量，能够减少避免单一模型的过拟合问题。此外，由于所述目标概率向量是由多个所述初始概率向量拼接得到，所述目标概率向量融合了所述多个初始概率向量中的信息，使得所述目标概率向量能够充分地反映蕴含在所述蛋白质序列内部的信息。
S14，根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果。
在一实施例中，所述蛋白质类别包括预设蛋白质类别和非预设蛋白质类别。其中，所述预设蛋白质类别包括，但不限于：嗜热蛋白和嗜冷蛋白。所述非预设蛋白质类别为除了所述预设蛋白质类别之外的其它蛋白质类别。例如，若所述预设蛋白质类别为嗜热蛋白，则所述非预设蛋白质类别可以为非嗜热蛋白，或者，若所述预设蛋白质类别为嗜冷蛋白，则所述非预设蛋白质类别可以为非嗜冷蛋白。
在一些实施例中，所述预测结果包括所述蛋白质序列属于所述预设蛋白质类别和所述蛋白质序列不属于所述预设蛋白质类别。
在一些实施例中，所述预测模型可以为机器学习模型。例如，所述预测模型可以为支持向量机模型(Support Vector Machines,SVM)。
在本实施例中，由于所述目标概率向量能够充分地反映蕴含在所述蛋白质序列内部的信息，因此对所述目标概率向量进行预测，能够提高对蛋白质序列的鉴定准确性。
由上述技术方案可知，本申请采用多个自监督模型对蛋白质序列进行信息提取，由于所述多个蛋白质模型为多个不同类型网络架构的自监督模型，因此，能够确保提取到的多个嵌入式向量为所述蛋白质序列中不同级别的信息。此外，本申请依靠预训练的多个自监督模型从蛋白质序列中提取信息，并不依赖于人工设计的特征从蛋白质序列中提取信息，因此能够解决的人工提取的特征的表达能力有限的技术问题，还能够提高信息提取的速度。由于本申请是将每个嵌入向量输入至每个分类模型中，因此可以得到每个嵌入向量在每个分类模型上的输出结果，将所述多个嵌入向量在所述多个分类模型上的输出结果进行拼接，得到所述目标概率向量，所述目标概率向量中融合了所述多个自监督模型以及所述多个分类模型对所述蛋白质序列进行提取的信息，因此所述目标概率向量能够充分地反映蕴含在所述蛋白质序列内部的信息，对所述目标概率向量进行预测，能够提高对蛋白质序列的鉴定效率。
在本申请的一些实施例中，所述电子设备将所述目标概率向量输入至所述预测模型中，得到对所述蛋白质序列的蛋白质类别的预测结果。如图3所示， 是本申请一实施例提供的预测结果的生成方法的流程图，包括以下步骤：
S141，所述电子设备对所述预设蛋白质类别和所述非预设蛋白质类别分别进行编码，得到类别向量。
在一些实施例中，所述预设蛋白质类别包括，但不限于：嗜热蛋白和嗜冷蛋白。所述非预设蛋白质类别为除了所述预设蛋白质类别之外的其它类别。例如，若所述预设蛋白质类别为嗜热蛋白，则所述非预设蛋白质类别可以为非嗜热蛋白，或者，若所述预设蛋白质类别为嗜冷蛋白，则所述非预设蛋白质类别可以为非嗜冷蛋白。
在一些实施例中，所述电子设备可以对所述预设蛋白质类别和所述非预设蛋白质类别分别进行独热编码(one-hot)，得到所述类别向量。例如，所述电子设备可以将所述预设蛋白质类别编码为1，并将所述非预设蛋白质类别编码为0，得到由1和0组成的类别向量。
在本申请的其它实施例中，所述电子设备可以通过其它方式对所述预设蛋白质类别和所述非预设蛋白质类别进行编码，本申请对编码的方式不作限制。
S142，所述电子设备基于所述类别向量对所述目标概率向量进行分类，确定所述蛋白质序列属于所述预设蛋白质类别的预测概率。
在一些实施例中，所述电子设备基于所述类别向量对所述目标概率向量进行分类，确定所述蛋白质序列属于所述预设蛋白质类别的预测概率的方法可参考相关技术中sigmoid函数或者softmax函数对应的分类公式。
S143，所述电子设备根据所述预测概率与预设阈值的比较结果确定所述预测结果。
在一些实施例中，所述电子设备根据所述预测概率与预设阈值的比较结果确定所述预测结果，包括：若所述预测概率大于或者等于预设阈值，所述电子设备确定所述预测结果为所述蛋白质序列属于所述预设蛋白质类别，或者，若所述预测概率小于所述预设阈值，所述电子设备确定所述预测结果为所述蛋白质序列不属于所述预设蛋白质类别，也可以是所述蛋白质序列属于所述非预设蛋白质类别。
在本申请的一实施例中，若所述电子设备确定所述预测结果为所述蛋白质序列属于所述预设蛋白质类别，所述电子设备输出第一预设数据以表示所述蛋白质序列属于所述预设蛋白质类别，或者，若所述电子设备确定所述预测结果为所述蛋白质序列不属于所述预设蛋白质类别(所述电子设备确定所述预测结果为所述蛋白质序列属于所述非预设蛋白质类别)，所述电子设备输出第二预设数据以表示所述蛋白质序列不属于所述预设蛋白质类别。
其中，所述第一预设数据与所述第二预设数据可以为数值、字母或者数值和字母的组合等形式。所述第一预设数据以及所述第二预设数据可以自行设置，本申请对此不作限制。例如，在所述第一预设数据与所述第二预设数据均为数值时，所述第一预设数值可以为1，所述第二预设数值可以为0。
在本申请的一些实施例中，在使用所述多个自监督模型、所述多个分类模型以及所述预测模型之前，需要生成所述多个自监督模型、所述多个分类模型 以及所述预测模型。如图4所示，是本申请一实施例提供的多个自监督模型、多个分类模型以及预测模型的生成方法的流程图，包括以下步骤：
S21，所述电子设备获取数据集，对所述数据集进行划分，得到第一训练集以及第二训练集。
在一些实施例中，所述数据集包括多个蛋白质序列，所述数据集中的每个蛋白质序列具有对应的类别标签，所述数据集可以包括，但不限于：Uniprot数据库、BRENDA数据库以及常见的包含蛋白质序列的数据库中的部分数据或者所有数据。其中，所述第一训练集包括多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签，所述第二训练集包括多个第二蛋白质序列以及每个第二蛋白质序列对应的第二类别标签，所述第一类别标签以及所述第二类别标签均包括预设蛋白质类别。
在一些实施例中，每个第一蛋白质序列具有对应的第一类别标签，每个第二蛋白质具有对应的第二类别标签，所述第一类别标签用于指示每个第一蛋白质序列的类别，所述第二类别标签用于指示每个第二蛋白质序列的类别。
在本实施例中，所述电子设备对所述数据集进行划分的方式可以自行设置，本申请对此不作限制。例如，所述电子设备可以将所述数据集中80％的数据作为所述第一训练集，将所述80％的数据的类别标签作为所述第一类别标签，然后所述电子设备将所述数据集中20％的数据作为所述第二训练集，将所述20％的数据的类别标签作为所述第二类别标签。上述举例仅为对所述数据集的一种划分方式的举例，而不构成对所述数据集的划分方式的限制。
S22，所述电子设备获取所述多个分类模型对应的多个初始分类器以及所述预测模型对应的预测学习器。
在一些实施例中，所述多个初始分类器可以包括所述随机森林模型Random Forest对应的分类器、所述自适应增强模型AdaBoost对应的分类器、所述装袋模型Bagging对应的分类器、所述梯度提升决策树模型XGBoost对应的分类器和所述梯度提升决策树模型LightGBM对应的分类器，所述预测学习器可以为支持向量机模型SVM的学习器。
S23，所述电子设备根据所述多个自监督模型中的任意一个模型、所述多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签对每个初始分类器进行训练，得到每个初始分类器对应的分类模型。
在一些实施例中，所述电子设备将每个第一蛋白质序列输入至每个自监督模型中，得到每个第一蛋白质序列在每个自监督模型上对应输出的初始嵌入向量，其中，每个第一蛋白质序列对应的多个初始嵌入向量与该第一蛋白质序列的第一类别标签相对应，然后所述电子设备将每个第一蛋白质序列对应的多个初始嵌入向量分别输入至所述多个初始分类器中，得到每个第一蛋白质序列的每个初始嵌入向量在每个初始分类器上对应输出的分类向量，将每个第一蛋白质序列对应的多个分类向量进行拼接，得到每个第一蛋白质序列对应的拼接概率向量，所述电子设备对每个第一蛋白质序列对应的拼接概率向量进行预测，得到每个第一蛋白质序列的预测类别，根据所述多个第一蛋白质的预测类别与 第一类别标签计算初始损失值，并根据所述初始损失值调整所述多个初始分类器的参数，直至所述初始损失值满足配置条件，得到所述初始分类器对应的分类模型。
在本实施例中，所述多个初始嵌入向量与上文中的多个嵌入向量的生成方式基本相同，所述拼接概率向量与上文中的目标概率向量的生成方式基本相同，故本申请不再重复描述。此外，所述电子设备对每个拼接概率向量进行预测可参考下文中对目标概率向量的预测过程，所述初始损失值的计算方式可参考下文中的训练损失值的计算方式，所述配置条件与下文中的预设条件基本相同。
S24，所述电子设备将任一自监督模型以及任一分类模型进行组合，得到多对组合模型。
例如，若所述多个自监督模型分别以A、B、C和D进行表示，所述多个分类模型分别以a、b、c、d和e进行表示。所述电子设备将任一自监督模型以及任一分类模型进行组合，得到的多对组合模型如表1所示。
表1多对组合模型示例
S25，所述电子设备根据每对组合模型对每个第二蛋白质序列进行信息提取，得到每个第二蛋白质序列对应的训练概率向量。
在一些实施例中，每个训练概率向量的生成方式与所述目标概率向量的生成方式基本相同，故本申请不再重复描述。
S26，所述电子设备使用所述预测学习器对每个训练概率向量进行预测，得到每个第二蛋白质序列的训练预测结果。
在一些实施例中，所述训练预测结果与所述蛋白质序列的预测结果的生成方式基本相同，故本申请在此不再重复描述。
S27，所述电子设备根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型。
在一些实施例中，所述电子设备根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型包括：所述电子设备根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值，根据所述训练损失值对所述预测学习器进行参数调整，直至所述训练损失值满足预设条件，得到所述预测模型。
其中，所述电子设备可以对所述多个自监督模型、所述多个初始分类器和 所述预测学习器的权重和偏置等参数进行调整。所述预设条件可以自行设置，本申请对此不作限制。例如，所述预设条件与所述训练预测结果相对应。例如，所述预设条件可以为所述训练损失值下降至一预设区间、所述训练损失值下降至最低或者所述训练损失值不再变化，其中，所述预设区间可以自行设置，本申请对此不作限制。
在一些实施例中，所述电子设备根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值包括：所述电子设备根据每个训练预测结果对应的第二类别标签，确定每个训练预测结果是否正确，然后所述电子设备根据正确的训练预测结果以及所述第二类别标签的标签数量，计算所述训练损失值。
其中，所述电子设备可以将每个训练预测结果与所述对应的第二类别标签进行比对，若每个训练预测结果指示的蛋白质与所述对应的第二类别标签相同，则所述电子设备确定每个训练预测结果正确，或者，若每个训练预测结果与所述对应的第二类别标签不相同，则所述电子设备确定每个训练预测结果错误。
例如，在任一第二蛋白质序列的训练预测结果为所述任一第二蛋白质序列属于所述预设蛋白质类别时，若所述任一第二蛋白质序列的第二类别标签为所述预设蛋白质类别，则所述电子设备确定任一第二蛋白质序列的训练预测结果正确。或者，在任一第二蛋白质序列的训练预测结果为所述任一第二蛋白质序列属于所述预设蛋白质类别时，若所述任一第二蛋白质序列的第二类别标签不是所述预设蛋白质类别，则所述电子设备确定任一第二蛋白质序列的训练预测结果错误。
在一实施例中，所述电子设备使用集成算法(例如bagging算法或者boosting算法)对所述多个初始分类器进行集成训练，得到由所述多个分类模型级联而成的模型。
在本实施例中，所述电子设备计算正确的训练预测结果的正确数量，并将所述正确数量与所述标签数量之间的比值确定为所述训练损失值。在所述训练损失值不再变化时，所述电子设备停止调整，得到所述多个自监督模型、所述多个分类模型以及所述预测模型。
在本申请的另外一些实施例中，所述训练损失值还可以通过其它方式进行计算。例如，所述电子设备可以将错误的训练预测结果的错误数量与所述标签数量之间的比值确定为所述训练损失值，在所述训练损失值不再下降或者下降至一预设区间，所述电子设备停止调整，得到所述多个自监督模型、所述多个分类模型以及所述预测模型。
如图5所示，是本申请一实施例提供的蛋白质鉴定装置的功能模块图。所述蛋白质鉴定装置11包括获取单元110、提取单元111、生成单元112以及预测单元113。本申请所称的模块/单元是指一种能够被图1中的处理器13所获取，并且能够完成固定功能的一系列计算机可读指令段，其存储在图1中的存储器12中。在本实施例中，关于各模块/单元的功能将在后续的实施例中详述。
在一些实施例中，所述获取单元110用于获取蛋白质序列。
所述提取单元111，用于使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量。
所述生成单元112，用于根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量。
在一些实施例中，所述生成单元112，还用于将每个嵌入向量输入至所述多个分类模型中，得到每个嵌入向量的多个初始概率向量，将所述多个嵌入向量的所有初始概率向量进行拼接，得到所述目标概率向量。
所述预测单元113，用于根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果。所述预设蛋白质类别包括嗜热蛋白和嗜冷蛋白。
在一些实施例中，所述蛋白质类别包括预设蛋白质类别和非预设蛋白质类别，所述预测单元113，还用于根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果，包括：对所述预设蛋白质类别和所述非预设蛋白质类别分别进行编码，得到类别向量，基于所述类别向量对所述目标概率向量进行分类，确定所述蛋白质序列属于所述预设蛋白质类别的预测概率，根据所述预测概率与预设阈值的比较结果确定所述预测结果。
在一些实施例中，所述预测单元113，还用于根据所述预测概率与预设阈值的比较结果确定所述预测结果包括：若所述预测概率大于或者等于预设阈值，确定所述预测结果为所述蛋白质序列属于所述预设蛋白质类别，或者，若所述预测概率小于所述预设阈值，确定所述预测结果为所述蛋白质序列不属于所述预设蛋白质类别。
在一些实施例中，所述生成单元112，还用于获取数据集，对所述数据集进行划分，得到第一训练集以及第二训练集，其中，所述第一训练集包括多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签，所述第二训练集包括多个第二蛋白质序列以及每个第二蛋白质序列对应的第二类别标签，所述第一类别标签以及所述第二类别标签均包括预设蛋白质类别，获取所述多个分类模型对应的多个初始分类器以及所述预测模型对应的预测学习器，根据所述多个自监督模型中的任意一个模型、所述多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签对每个初始分类器进行训练，得到每个初始分类器对应的分类模型，将任一自监督模型以及任一分类模型进行组合，得到多对组合模型，根据每对组合模型对每个第二蛋白质序列进行信息提取，得到每个第二蛋白质序列对应的训练概率向量，使用所述预测学习器对每个训练概率向量进行预测，得到每个第二蛋白质序列的训练预测结果，根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型。
在一些实施例中，所述生成单元112，还用于根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型包括：根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值，根据所述训练损失值对所述预测学习器进行参数调整，直至所述训练损失值满足预 设条件，得到所述预测模型。
在一些实施例中，所述生成单元112，还用于根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值包括：根据每个训练预测结果对应的第二类别标签，确定每个训练预测结果是否正确，根据正确的训练预测结果以及所述第二类别标签的标签数量，计算所述训练损失值。
由上述技术方案可知，本申请采用多个自监督模型对蛋白质序列进行信息提取，由于所述多个蛋白质模型为多个不同类型网络架构的自监督模型，因此，能够确保提取到的多个嵌入式向量为所述蛋白质序列中不同级别的信息。此外，本申请依靠预训练的多个自监督模型从蛋白质序列中提取信息，并不依赖于人工设计的特征从蛋白质序列中提取信息，因此能够解决的人工提取的特征的表达能力有限的技术问题，还能够提高信息提取的速度。由于本申请是将每个嵌入向量输入至每个分类模型中，因此可以得到每个嵌入向量在每个分类模型上的输出结果，将所述多个嵌入向量在所述多个分类模型上的输出结果进行拼接，得到所述目标概率向量，所述目标概率向量中融合了所述多个自监督模型以及所述多个分类模型对所述蛋白质序列进行提取的信息，因此所述目标概率向量能够充分地反映蕴含在所述蛋白质序列内部的信息，对所述目标概率向量进行预测，能够提高对蛋白质序列的鉴定效率。
在一实施例中，接续前文对图1中的电子设备的描述，处理器13可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现场可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，处理器13是电子设备1的运算核心和控制中心，利用各种接口和线路连接整个电子设备1的各个部分，及获取电子设备1的操作系统以及安装的各类应用程序、程序代码等。
处理器13获取电子设备1的操作系统以及安装的各类应用程序。处理器13获取应用程序以实现上述各个蛋白质鉴定方法实施例中的步骤，例如图2、图3及图4所示的步骤。
示例性的，计算机程序可以被分割成一个或多个模块/单元，例如加速单元，一个或者多个模块/单元被存储在存储器12中，并由处理器13获取，以完成本申请。一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述计算机程序在电子设备1中的获取过程。
存储器12可用于存储计算机程序和/或模块，处理器13通过运行或获取存储在存储器12内的计算机程序和/或模块，以及调用存储在存储器12内的数据，实现电子设备1的各种功能。存储器12可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等；存储数据区可存储根据服务器的使用所创建的数据等。此外，存储器12可以包括非易失性存储器，例如硬盘、内存、插接 式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。
存储器12可以是电子设备1的外部存储器和/或内部存储器。进一步地，存储器12可以是具有实物形式的存储器，如内存条、TF卡(Trans-flash Card)等等。
电子设备1集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请实现上述实施例方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器获取时，可实现上述各个方法实施例的步骤。
其中，计算机程序包括计算机程序代码，计算机程序代码可以为源代码形式、对象代码形式、可获取文件或某些中间形式等。计算机可读介质可以包括：能够携带计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(Read-Only Memory，ROM)。
结合图2，电子设备1中的存储器12存储多个指令以实现一种蛋白质鉴定方法，处理器13可获取多个指令从而实现：获取蛋白质序列；使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量；根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量；根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果。
具体地，处理器13对上述指令的具体实现方法可参考图2对应实施例中相关步骤的描述，在此不赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，模块的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
作为分离部件说明的模块可以是或者也可以不是物理上分开的，作为模块显示的部件可以是或者也可以不是物理单元，即可以处于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选取其中的部分或者全部模块来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能模块的形式实现。
因此，无论从哪一点来看，均应将实施例看作是示范性的，而且是非限制性的，本申请的范围由所附权利要求而不是上述说明限定，因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本申请内。不应将权利要求中的任何附关联图标记视为限制所涉及的权利要求。
此外，显然“包括”一词不排除其他单元或步骤，单数不排除复数。本申请中陈述的多个单元或装置也可以由一个单元或装置通过软件或者硬件来实现。第一、第二等词语用来表示名称，而并不表示任何特定的顺序。
最后应说明的是，以上实施例仅用以说明本申请的技术方案而非限制，尽管参照较佳实施例对本申请进行了详细说明，本领域的普通技术人员应当理解，可以对本申请的技术方案进行修改或等同替换，而不脱离本申请技术方案的精神和范围。

Claims (10)

  1. 一种蛋白质鉴定方法，其特征在于，所述方法包括：
    获取蛋白质序列；
    使用预训练的多个自监督模型对所述蛋白质序列进行信息提取，得到所述蛋白质序列的多个嵌入向量；
    根据所述多个嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量；
    根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果。
  2. 如权利要求1所述的蛋白质鉴定方法，其特征在于，所述方法还包括：
    获取数据集，对所述数据集进行划分，得到第一训练集以及第二训练集，其中，所述第一训练集包括多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签，所述第二训练集包括多个第二蛋白质序列以及每个第二蛋白质序列对应的第二类别标签，所述第一类别标签以及所述第二类别标签均包括预设蛋白质类别；
    获取所述多个分类模型对应的多个初始分类器以及所述预测模型对应的预测学习器；
    根据所述多个自监督模型中的任意一个模型、所述多个第一蛋白质序列以及每个第一蛋白质序列对应的第一类别标签对每个初始分类器进行训练，得到每个初始分类器对应的分类模型；
    将任一自监督模型以及任一分类模型进行组合，得到多对组合模型；
    根据每对组合模型对每个第二蛋白质序列进行信息提取，得到每个第二蛋白质序列对应的训练概率向量；
    使用所述预测学习器对每个训练概率向量进行预测，得到每个第二蛋白质序列的训练预测结果；
    根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型。
  3. 如权利要求2所述的蛋白质鉴定方法，其特征在于，所述根据所述训练预测结果、所述第二类别标签和所述预测学习器，生成所述预测模型包括：
    根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值；
    根据所述训练损失值对所述预测学习器进行参数调整，直至所述训练损失值满足预设条件，得到所述预测模型。
  4. 如权利要求2所述的蛋白质鉴定方法，其特征在于，所述根据多个所述训练预测结果以及每个训练预测结果对应的第二类别标签计算训练损失值包括：
    根据每个训练预测结果对应的第二类别标签，确定每个训练预测结果是否正确；
    根据正确的训练预测结果以及所述第二类别标签的标签数量，计算所述训练损失值。
  5. 如权利要求1所述的蛋白质鉴定方法，其特征在于，所述根据所述多个 嵌入向量以及预训练的多个分类模型，生成所述蛋白质序列的目标概率向量包括：
    将每个嵌入向量输入至所述多个分类模型中，得到每个嵌入向量的多个初始概率向量；
    将所述多个嵌入向量的所有初始概率向量进行拼接，得到所述目标概率向量。
  6. 如权利要求1所述的蛋白质鉴定方法，其特征在于，所述蛋白质类别包括预设蛋白质类别和非预设蛋白质类别，所述根据预训练的预测模型对所述目标概率向量进行预测，得到对所述蛋白质序列的蛋白质类别的预测结果，包括：
    对所述预设蛋白质类别和所述非预设蛋白质类别分别进行编码，得到类别向量；
    基于所述类别向量对所述目标概率向量进行分类，确定所述蛋白质序列属于所述预设蛋白质类别的预测概率；
    根据所述预测概率与预设阈值的比较结果确定所述预测结果。
  7. 如权利要求6所述的蛋白质鉴定方法，其特征在于，所述根据所述预测概率与预设阈值的比较结果确定所述预测结果包括：
    若所述预测概率大于或者等于预设阈值，确定所述预测结果为所述蛋白质序列属于所述预设蛋白质类别；或者
    若所述预测概率小于所述预设阈值，确定所述预测结果为所述蛋白质序列不属于所述预设蛋白质类别。
  8. 如权利要求6所述的蛋白质鉴定方法，其特征在于，所述预设蛋白质类别包括嗜热蛋白和嗜冷蛋白。
  9. 一种电子设备，其特征在于，所述电子设备包括：
    存储器，存储至少一个指令；及
    处理器，执行所述至少一个指令以实现如权利要求1至8中任意一项所述的蛋白质鉴定方法。
  10. 一种计算机可读存储介质，其特征在于：所述计算机可读存储介质中存储有至少一个指令，所述至少一个指令被电子设备中的处理器执行以实现如权利要求1至8中任意一项所述的蛋白质鉴定方法。
PCT/CN2023/137165 2023-05-22 2023-12-07 蛋白质鉴定方法、电子设备及存储介质 Ceased WO2024239598A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202310588503.4A CN116646009A (zh) 2023-05-22 2023-05-22 蛋白质鉴定方法、电子设备及存储介质
CN202310588503.4 2023-05-22

Publications (1)

Publication Number Publication Date
WO2024239598A1 true WO2024239598A1 (zh) 2024-11-28

Family

ID=87639281

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/137165 Ceased WO2024239598A1 (zh) 2023-05-22 2023-12-07 蛋白质鉴定方法、电子设备及存储介质

Country Status (2)

Country Link
CN (1) CN116646009A (zh)
WO (1) WO2024239598A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119479836A (zh) * 2025-01-15 2025-02-18 之江实验室 一种基于预训练大语言模型的蛋白质功能注释方法和装置

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116646009A (zh) * 2023-05-22 2023-08-25 深圳先进技术研究院 蛋白质鉴定方法、电子设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112420123A (zh) * 2020-11-30 2021-02-26 上海商汤智能科技有限公司 自监督学习模型的训练方法和装置、设备以及存储介质
CN114333982A (zh) * 2021-11-26 2022-04-12 北京百度网讯科技有限公司 蛋白质表示模型预训练、蛋白质相互作用预测方法和装置
US20220165356A1 (en) * 2020-11-23 2022-05-26 NE47 Bio, Inc. Protein database search using learned representations
WO2022185179A1 (en) * 2021-03-02 2022-09-09 Glaxosmithkline Biologicals Sa Natural language processing to predict properties of proteins
CN116013428A (zh) * 2023-02-10 2023-04-25 中南大学 基于自监督学习的药物靶标通用预测方法、设备及介质
CN116646009A (zh) * 2023-05-22 2023-08-25 深圳先进技术研究院 蛋白质鉴定方法、电子设备及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220165356A1 (en) * 2020-11-23 2022-05-26 NE47 Bio, Inc. Protein database search using learned representations
CN112420123A (zh) * 2020-11-30 2021-02-26 上海商汤智能科技有限公司 自监督学习模型的训练方法和装置、设备以及存储介质
WO2022185179A1 (en) * 2021-03-02 2022-09-09 Glaxosmithkline Biologicals Sa Natural language processing to predict properties of proteins
CN114333982A (zh) * 2021-11-26 2022-04-12 北京百度网讯科技有限公司 蛋白质表示模型预训练、蛋白质相互作用预测方法和装置
CN116013428A (zh) * 2023-02-10 2023-04-25 中南大学 基于自监督学习的药物靶标通用预测方法、设备及介质
CN116646009A (zh) * 2023-05-22 2023-08-25 深圳先进技术研究院 蛋白质鉴定方法、电子设备及存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WU, JIANG: "Protein Structural Classes Prediction by Using Laplace Support Vector Machine and Based on Semi-supervised Method", MICROCOMPUTER APPLICATIONS, vol. 36, no. 8, 1 January 2020 (2020-01-01), XP093241809 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN119479836A (zh) * 2025-01-15 2025-02-18 之江实验室 一种基于预训练大语言模型的蛋白质功能注释方法和装置

Also Published As

Publication number Publication date
CN116646009A (zh) 2023-08-25

Similar Documents

Publication Publication Date Title
CN111462761A (zh) 声纹数据生成方法、装置、计算机装置及存储介质
US10262272B2 (en) Active machine learning
WO2024239598A1 (zh) 蛋白质鉴定方法、电子设备及存储介质
CN112883990A (zh) 数据分类方法及装置、计算机存储介质、电子设备
CN113591881B (zh) 基于模型融合的意图识别方法、装置、电子设备及介质
CN117390213B (zh) 基于oscar的图文检索模型的训练方法和实现图文检索的方法
CN116204726A (zh) 一种基于多模态模型的数据处理方法、装置及设备
Singh et al. Sequence based prediction of enhancer regions from DNA random walk
CN113609948B (zh) 一种视频时序动作的检测方法、装置及设备
Lee et al. Protein family classification with neural networks
CN117992765B (zh) 基于动态新兴标记的偏标签学习方法、装置、设备及介质
CN112801006A (zh) 表情表征模型的训练方法、面部的表情表征方法及装置
WO2022142019A1 (zh) 基于智能机器人的问题分发方法、装置、电子设备及存储介质
Zhu et al. Malware homology determination using visualized images and feature fusion
CN111860508A (zh) 图像样本选择方法及相关设备
CN115222443A (zh) 客户群体划分方法、装置、设备及存储介质
CN111582315B (zh) 样本数据处理方法、装置及电子设备
CN116150367B (zh) 一种基于方面的情感分析方法及系统
CN107688744B (zh) 基于图像特征匹配的恶意文件分类方法及装置
Ranjini et al. An ensemble of heterogeneous incremental classifiers for assisted reproductive technology outcome prediction
WO2021212753A1 (zh) 计算机性能数据确定方法、装置、计算机设备及存储介质
Yu et al. iDNA-ITLM: An interpretable and transferable learning model for identifying DNA methylation
CN118194044A (zh) 一种训练数据的获取方法、装置及电子设备
CN117609864A (zh) 一种基于双通道图对比学习的虚假信息检测方法及系统
CN117932016A (zh) 对话生成方法及相关装置、设备和存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23938265

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE