Enzyme kinetics parameter prediction model training method, enzyme kinetics parameter prediction method, and related device
By using a quaternary data set and a two-layer architecture model containing environmental factors, combining enzyme sequence and substrate structure, the problems of low accuracy of the prediction of enzyme kinetic parameters and poor model universality in the existing technology are solved, and more accurate and General prediction of enzyme kinetic parameters.
Patent Information
- Application Number
- PCT/CN2023/097103
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-30
- Publication Date
- 2024-12-05
AI Technical Summary
The existing methods for predicting enzyme kinetic parameters fail to effectively consider environmental factors, which makes it difficult to achieve expectations for prediction accuracy, and the model is difficult to generalize to other species, making it impossible to achieve generalized predictions.
A method for training an enzyme dynamics parameter prediction model is proposed. By obtaining a quaternary data set containing environmental factors, a machine learning model with a two-layer architecture is adopted, combining enzyme sequence, substrate structure and environmental factor values for training, and optimizing the model to improve the Predict accuracy and suitable for any enzyme sequence and substrate structure.
The accuracy of the prediction of enzyme kinetic parameter and the universality of the model are improved, and the influence of environmental factors is systematically considered to achieve more accurate prediction of enzyme kinetic parameter, which is suitable for enzyme sequences of different species.
Smart Images

Figure CN2023097103_05122024_PF_FP_ABST
Abstract
Description
Enzyme kinetic parameter prediction model training and prediction methods and related equipment Technical Field
[0001] The present application relates to the field of biomedicine technology, and in particular to an enzyme kinetic parameter prediction model training and prediction method and related equipment. Background Art
[0002] In the biomedical field, enzymes are catalysts of cellular metabolism, and accurate prediction of enzyme kinetic parameters is of great research significance. Existing methods for predicting enzyme kinetic parameters directly rely on the protein structure of the enzyme and the structure of the substrate. These methods fail to consider the role of environmental factors in predicting enzyme kinetic parameters, resulting in poor prediction accuracy.
[0003] Summary of the Invention
[0004] In view of the above, it is necessary to propose an enzyme kinetic parameter prediction model training and prediction method and related equipment, which can save the model training cost while improving the accuracy of the model in predicting enzyme kinetic parameters.
[0005] An embodiment of the present application provides a method for training an enzyme kinetic parameter prediction model, the method comprising: obtaining sample data, the sample data comprising a quaternary data set consisting of quaternary data, the quaternary data comprising an environmental factor value-an enzyme sequence-a substrate structure-a first true enzyme kinetic parameter; dividing the sample data into a training set and a test set, and using the training set to train an enzyme kinetic parameter prediction model; determining the prediction accuracy of the enzyme kinetic parameter prediction model based on the test set, and optimizing the prediction model according to the prediction accuracy.
[0006] In one embodiment, the environmental factor value includes pH value and / or temperature value.
[0007] In one embodiment, the sample data further includes a ternary data set consisting of ternary data, wherein the ternary data includes enzyme sequence-substrate structure-second real enzyme kinetic parameter.
[0008] In one embodiment, dividing the sample data into a training set and a test set includes: using the ternary data set as a first training set; dividing the quaternary data set into a second training set and a test set according to a preset first ratio, and dividing the second training set into a third training set and a fourth training set according to a preset second ratio.
[0009] In one embodiment, the enzyme kinetic parameter prediction model includes a two-layer architecture, the two-layer architecture includes a basic layer and a meta-layer, the basic layer includes a first model and a second model, the meta-layer includes a machine learning model, and the input of the machine learning model includes the output of the first model and the output of the second model.
[0010] In one embodiment, the use of the training set to train the enzyme kinetic parameter prediction model includes: using the first training set in the training set to train the first model, the first model outputting the first prediction parameter of the enzyme based on the enzyme sequence-substrate structure in the first training set; using the third training set in the training set to train the second model, the second model outputting the second prediction parameter of the enzyme based on the environmental factor value-enzyme sequence-substrate structure in the third training set; training the machine learning model based on the fourth training set in the training set, including: inputting the enzyme sequence-substrate structure in the fourth training set into the trained first model, the first model outputting the third prediction parameter of the enzyme; inputting the environmental factor value-enzyme sequence-substrate structure in the fourth training set into the trained second model, the second model outputting the fourth prediction parameter of the enzyme, and using the third prediction parameter and the fourth prediction parameter to train the machine learning model.
[0011] In one embodiment, the use of the test set to determine the prediction accuracy of the enzyme kinetic parameter prediction model includes: inputting the environmental factor value-enzyme sequence-substrate structure in the test set into the enzyme kinetic parameter prediction model to obtain the fourth prediction parameter of the enzyme output by the enzyme kinetic parameter prediction model; and determining the prediction accuracy based on the first true enzyme kinetic parameter in the test set and the fourth prediction parameter.
[0012] An embodiment of the present application provides an enzyme kinetic parameter prediction method, which includes: obtaining the environmental factor value-enzyme sequence-substrate structure corresponding to the enzyme to be tested; inputting the environmental factor value-enzyme sequence-substrate structure corresponding to the enzyme to be tested into an enzyme kinetic parameter prediction model, and using the enzyme kinetic parameter prediction model to output the enzyme kinetic parameters of the enzyme to be tested, the enzyme kinetic parameters including the enzyme turnover number, and the enzyme kinetic parameter prediction model is obtained using the enzyme kinetic parameter prediction model training method.
[0013] An embodiment of the present application provides an enzyme kinetic parameter prediction model training device, which includes: an acquisition module for acquiring sample data, the sample data including a four-element data set composed of four-element data, the four-element data including environmental factor value-enzyme sequence-substrate structure-first real enzyme kinetic parameter; a training module for dividing the sample data into a training set and a test set, and using the training set to train the enzyme kinetic parameter prediction model; an optimization module for determining the prediction accuracy of the enzyme kinetic parameter prediction model based on the test set, and optimizing the prediction model according to the prediction accuracy.
[0014] An embodiment of the present application provides an electronic device, which includes a processor and a memory, and the processor is used to implement the enzyme kinetic parameter prediction model training and enzyme kinetic parameter prediction method when executing the computer program stored in the memory.
[0015] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the enzyme kinetic parameter prediction model training and enzyme kinetic parameter prediction method are implemented.
[0016] In summary, the enzyme kinetic parameter prediction model training and prediction method and related equipment described in this application can take into account the impact of environmental factors on enzyme kinetic parameters when training the model, thereby improving the prediction accuracy of enzyme kinetic parameters, and the trained model is applicable to any enzyme sequence and substrate structure, improving the versatility and practicality of the model. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.
[0018] FIG1 is a structural diagram of an electronic device provided in one embodiment of the present application.
[0019] FIG2 is a flow chart of an enzyme kinetic parameter prediction model training method provided in an embodiment of the present application.
[0020] FIG3 is an example diagram of a data distribution histogram in a quaternary data set provided by an embodiment of the present application.
[0021] FIG4 is a logic example diagram of model training provided in an embodiment of the present application.
[0022] FIG5 is a flow chart of an enzyme kinetic parameter prediction method provided in one embodiment of the present application.
[0023] FIG6 is a structural diagram of an enzyme kinetic parameter prediction model training device provided in one embodiment of the present application.
[0024] The following specific implementation methods will further illustrate the present application in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION
[0025] In order to more clearly understand the above-mentioned objectives, features and advantages of the present application, the present application is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features therein can be combined with each other in the absence of conflict.
[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application pertains. The terms used herein in the specification of this application are for the purpose of describing embodiments in one embodiment only and are not intended to limit this application.
[0027] In one embodiment, in the field of biomedicine, enzymes are catalysts of cellular metabolism, and accurate prediction of enzyme kinetic parameters is of great research significance. Existing enzyme kinetic parameter prediction methods directly predict enzyme kinetic parameters based on the protein structure of the enzyme and the structure of the substrate, without considering the role of environmental factors in the prediction process. Consequently, the prediction accuracy is less than expected.
[0028] In addition, there are enzyme kinetic parameter prediction methods that predict enzyme kinetic parameters of specific species based on multi-level data such as enzyme biochemistry, enzyme protein structure and network environment. The models used rely on detailed annotations of specific species and are difficult to effectively generalize to other species, and cannot be used for universal enzyme kinetic parameter prediction.
[0029] In order to solve the above problems, the embodiments of the present application provide an enzyme kinetic parameter prediction model training method, which can take into account the impact of environmental factors on enzyme kinetic parameters when training the model, thereby improving the prediction accuracy of enzyme kinetic parameters, and the trained model is applicable to any enzyme sequence and substrate structure, improving the versatility and practicality of the model.
[0030] For example, Figure 1 shows a block diagram of an electronic device provided in one embodiment of the present application. The enzyme kinetic parameter prediction model training method provided in this embodiment of the present application is performed by an electronic device, which can be a computer, server, laptop computer, mobile phone, or other device. The electronic device 1 includes a memory 11, at least one processor 12, at least one communication bus 13, and a transceiver 14.
[0031] The structure of the electronic device shown in FIG1 does not constitute a limitation of the embodiments of the present application, and can be either a bus structure or a star structure. The electronic device 1 can also include more or less other hardware or software than shown in the figure, or a different component arrangement.
[0032] In some embodiments, the electronic device 1 is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, a microprocessor, an application-specific integrated circuit, a programmable gate array, a digital processor, and an embedded device. The electronic device 1 may also include other external devices, such as input and output devices such as a keyboard, a mouse, a remote control, a display, a touchpad, or a voice control device.
[0033] It should be noted that the electronic device 1 is only an example. Other existing or future electronic products that are suitable for this application should also be included in the scope of protection of this application and included here by reference.
[0034] FIG2 is a flow chart of an enzyme kinetic parameter prediction model training method provided in one embodiment of the present application. The enzyme kinetic parameter prediction model training method is applied to an electronic device, such as the electronic device 1 in FIG1 , and specifically includes the following steps. The order of the steps in the flow chart may be changed, and some steps may be omitted, depending on different requirements.
[0035] S21, obtain sample data.
[0036] In one embodiment, the sample data includes a four-element data set consisting of four-element data, wherein the four-element data includes environmental factor value-enzyme sequence-substrate structure-first real enzyme kinetic parameter, and the environmental factor value includes pH value and / or temperature value.
[0037] In one embodiment, in order to obtain a prediction model that can take into account the impact of environmental factors on enzyme kinetic parameters, the present application embodiment constructs a four-element data set including environmental factors. Wherein, the environmental factors include but are not limited to common environmental factors such as acid-base pH value, temperature value, co-substrate, sodium chloride concentration. For example, when the environmental factor value is pH value, the four-element data in the four-element data set is pH value-enzyme sequence-substrate structure-first true enzyme kinetic parameter.
[0038] In addition, the enzyme sequence represents the protein (amino acid) sequence of the enzyme, the quaternary data represents the first enzyme kinetics real parameter of the enzyme protein sequence in the corresponding environmental factor value and the specific substrate structure, and the first enzyme kinetics real parameter represents the real enzyme kinetic parameter obtained according to the experiment. For example, when the quaternary data is pH value-enzyme sequence-substrate structure-first real enzyme turnover number k catWhen the enzyme protein sequence is in the corresponding pH value and the k cat In addition, the first true enzyme kinetic parameter may also include other parameters, such as the Michaelis constant K m wait.
[0039] In one embodiment, the quaternary data set can be obtained by constructing the quaternary data set based on an open source database, for example, the quaternary data set can be constructed based on enzyme sequences, substrate structures and corresponding pH values or temperature values from databases such as UniProt and PubChem.
[0040] In one embodiment, as shown in FIG3 , an example diagram of a data distribution histogram in a quaternary data set provided in an embodiment of the present application is shown. The horizontal axis of the pH value distribution histogram on the left side of FIG3 represents the pH value, and the vertical axis represents the number of corresponding pH value quaternary data groups; the horizontal axis of the temperature value distribution histogram on the right side of FIG3 represents the temperature value (in degrees Celsius), and the vertical axis represents the number of corresponding temperature value quaternary data groups.
[0041] For example, as shown in Figure 3, the number of samples of the quaternary data group corresponding to the pH value in the quaternary data set constructed in the embodiment of the present application is 636, where the pH value ranges from 3 to 10.5, which can almost cover all possible values of the pH value in the actual enzyme kinetics experimental scenario; the number of samples of the quaternary data group corresponding to the temperature value in the quaternary data set is 572, where the temperature value ranges from 4 to 85 degrees Celsius, which can almost cover all possible values of the temperature value in the actual enzyme kinetics experimental scenario.
[0042] In one embodiment, the sample data further includes a ternary data set consisting of ternary data, wherein the ternary data includes enzyme sequence-substrate structure-second real enzyme kinetic parameter.
[0043] In one embodiment, due to the small number of samples in the quaternary dataset, directly using the quaternary dataset to train the prediction model may not achieve the expected results. Furthermore, since related technologies typically predict enzyme kinetic parameters based solely on enzyme sequence and substrate structure, in order to study the impact of environmental factors on enzyme kinetic parameters, it is necessary to obtain a ternary dataset in addition to the quaternary dataset. This allows the model to be trained using both the ternary and quaternary datasets to improve the model's prediction accuracy.
[0044] In one embodiment, the ternary data set can be constructed based on the DLKcat data set, wherein the enzyme sequence in the ternary data represents the protein sequence of the enzyme, and the ternary data represents the second real enzyme kinetic parameter of the enzyme protein sequence in a specific substrate structure, and the second real enzyme kinetic parameter in the ternary data is a real enzyme kinetic parameter obtained from an experiment. For example, when the ternary data is enzyme sequence-substrate structure-second real enzyme turnover number k cat When , it indicates the k of the enzyme protein sequence in a specific substrate structure cat In addition, the second true enzyme kinetic parameter may also include other parameters, such as the Michaelis constant K m In one embodiment, the enzyme kinetic parameter k cat Give an example.
[0045] S22, dividing the sample data into a training set and a test set, and using the training set to train an enzyme kinetic parameter prediction model.
[0046] In one embodiment, the enzyme kinetic parameter prediction model comprises a two-layer architecture comprising a base layer and a meta-layer, wherein the base layer comprises a first model and a second model, and the meta-layer comprises a machine learning model (e.g., a linear regression model), wherein the input of the machine learning model comprises the output of the first model and the output of the second model. The specific model structure and training method will be described in conjunction with subsequent examples.
[0047] In one embodiment, the dividing of the sample data into a training set and a test set includes: using the ternary data set as the first training set; dividing the quaternary data set into a second training set and a test set according to a preset first ratio; and dividing the second training set into a third training set and a fourth training set according to a preset second ratio. For example, when the first ratio is 8:2, the second training set accounts for 80% of the entire quaternary data set, and the test set accounts for 20% of the entire quaternary data set; when the second ratio is 8:2, the third training set accounts for 80% of the second training set and 64% of the entire quaternary data set, and the fourth training set accounts for 20% of the second training set and 16% of the entire quaternary data set. The specific reasons for the division will be explained in conjunction with subsequent embodiments.
[0048] In one embodiment, the use of the training set to train the enzyme kinetic parameter prediction model includes the following (1)-(4):
[0049] (1) The first model is trained (or inputted) using the first training set in the training set, and the first model outputs a first prediction parameter of the enzyme according to the enzyme sequence-substrate structure in the first training set.
[0050] In one embodiment, the first model is inputted with an enzyme sequence and substrate structure, and is used to predict enzyme kinetic parameters based on the enzyme sequence and substrate structure to obtain first predicted parameters. Therefore, the first model is trained using a first training set corresponding to the ternary data set. Due to the large number of samples in the ternary data set, the first model, which does not consider environmental factors, has a higher prediction accuracy.
[0051] In addition, since the first model has the same function as the enzyme kinetic parameter prediction model in the related art that does not consider environmental factors, there is generally no need to train and optimize the first model, and the model architecture and parameters in the related art can be directly used.
[0052] For example, the first model includes but is not limited to the DLKcat model, which is a deep learning model provided by the research team of Chalmers University of Technology. It can be used to output the corresponding k according to the input enzyme sequence-substrate structure. cat Specifically, the SMILES (Simplified Molecular Input Line Entry System) corresponding to the enzyme sequence and substrate structure is input into the first model. The first model uses the convolutional neural network to encode the information of the enzyme sequence to generate A, and uses the graph neural network to encode the information of the enzyme sequence to generate B. Then, the first model uses the concatenated vector composed of A and B as the input of the neural network mechanism and outputs the predicted k cat The predicted value K1.
[0053] In other embodiments, the first training set may be divided into a test set M and a training set N to train the first model.
[0054] (2) The second model is trained using a third training set in the training set, and the second model outputs a second prediction parameter of the enzyme based on the environmental factor value-enzyme sequence-substrate structure in the third training set.
[0055] In one embodiment, the second model is trained using a quaternary dataset, which is an input of an environmental factor value, an enzyme sequence, and a substrate structure, and is used to predict enzyme kinetic parameters based on the environmental factor value, the enzyme sequence, and the substrate structure to obtain a second predicted parameter.
[0056] In one embodiment, similar to the first model, the second model may also include but is not limited to the DLKcat model. However, since the existing DLKcat model does not take environmental factors into consideration, it is necessary to divide the quaternary data set into a training set (third training set) and a test set to train the second model, and obtain updated model parameters to improve the prediction accuracy of the model. Specifically, the input of the second model is compared with the input of the first model, and the second model uses the series vector composed of A, B and C as the input of the neural network mechanism, and outputs the predicted k cat The predicted value K2.
[0057] In one embodiment, the reason why the third training set divided from the second training set is used to train the second model instead of directly using the second training set will be explained in conjunction with subsequent embodiments.
[0058] (3) training the machine learning model based on the fourth training set in the training set, comprising: inputting the enzyme sequence-substrate structure in the fourth training set into the trained first model, and the first model outputting the third prediction parameter of the enzyme; inputting the environmental factor value-enzyme sequence-substrate structure in the fourth training set into the trained second model, and the second model outputting the fourth prediction parameter of the enzyme, and using the third prediction parameter and the fourth prediction parameter to train the machine learning model.
[0059] In one embodiment, the first model of the base layer is used to obtain the prediction value without considering environmental factors, and the second model of the base layer is used to obtain the prediction value considering environmental factors. In order to study the impact of environmental factors, a meta-layer is set after the base layer.
[0060] The input of the meta-layer is the output of the first model and the output of the second model, which is used to correct the output of the first model according to the output of the second model, thereby realizing the prediction of enzyme kinetic parameters after adding environmental factors on the basis of relevant technologies and improving the model prediction accuracy.
[0061] In one embodiment, the enzyme sequence-substrate structure in the fourth training set is used as the input of the trained first model to obtain the third prediction parameter of the enzyme output by the first model; the environmental factor value-enzyme sequence-substrate structure in the fourth training set is used as the input of the trained second model to obtain the fourth prediction parameter of the enzyme output by the second model; and the third prediction parameter and the fourth prediction parameter are then used as the input of the meta-layer to train the meta-layer.
[0062] In one embodiment, the second training set cannot be used directly to train the second model and the meta-layer model. The reason is that if the second training set is used directly to train the second model, when training the meta-layer, the data containing environmental factors in the second training set must still be input into the second model to obtain second prediction parameters. If the second prediction parameters are then used to train the meta-layer model, the generalization ability of the meta-layer model will be insufficient, overfitting will occur, and the prediction accuracy of the meta-layer model cannot be trained to meet expectations.
[0063] Therefore, when training the meta-layer, different training samples should be used from those used when training the second model. In the embodiment of the present application, after training the second model using the third training set, the fourth training set is used as the input of the second model to obtain the third prediction parameters for training the meta-layer model.
[0064] In one embodiment, the meta-layer may include but is not limited to common machine learning models such as linear regression and random forest. For example, when the meta-layer is a linear regression model, the concatenated vector of the first prediction parameter and the third prediction parameter is used as the input of the linear regression model, and the predicted k is output. cat The predicted value K.
[0065] In one embodiment, as shown in Figure 4, a logical example diagram of model training provided in an embodiment of the present application is provided. Here, the environmental factors are pH and temperature, Model 1 represents the first model, Model 2 represents the second model, and the meta-layer model is a linear regression model.
[0066] Step S23: determining the prediction accuracy of the enzyme kinetic parameter prediction model based on the test set, and optimizing the prediction model according to the prediction accuracy.
[0067] In one embodiment, the use of the test set to determine the prediction accuracy of the enzyme kinetic parameter prediction model includes: inputting the environmental factor value-enzyme sequence-substrate structure in the test set into the enzyme kinetic parameter prediction model to obtain the fourth prediction parameter of the enzyme output by the enzyme kinetic parameter prediction model; and determining the prediction accuracy based on the first true enzyme kinetic parameter in the test set and the fourth prediction parameter.
[0068] In order to optimize the prediction model and train the model prediction accuracy to the expected level, the least squares method can be used to approximate the curve formed by the model prediction values to the curve fitted with the true values, thereby adjusting the model parameters to obtain the corresponding updated model parameters until the prediction accuracy is trained to the expected level.
[0069] In one embodiment, the enzyme kinetic parameter prediction model training method provided in the embodiment of the present application includes at least the following beneficial effects: combining data sets from different sources and using a two-layer learning framework to improve the versatility and prediction accuracy of the model in the application of enzymes in different species; a new four-element data set including environmental factors is created, and the influence of environmental factors such as pH or temperature is systematically considered to achieve more accurate prediction of enzyme kinetic parameters.
[0070] After the enzyme kinetic parameter prediction model is obtained through training, the enzyme kinetic parameter prediction model can be used to predict enzyme kinetic parameters. The enzyme kinetic parameter prediction method provided in the embodiment of the present application is performed by an electronic device (e.g., the electronic device 1 shown in FIG1 ). Accordingly, in one embodiment, the electronic device can include an enzyme kinetic parameter prediction device running therein.
[0071] Figure 5 is a flow chart of an enzyme kinetic parameter prediction method provided in one embodiment of the present application. The enzyme kinetic parameter prediction method specifically includes the following steps. According to different requirements, the order of the steps in the flow chart can be changed, and some steps can be omitted.
[0072] S51, obtaining the environmental factor value-enzyme sequence-substrate structure corresponding to the enzyme to be tested.
[0073] In one embodiment, the environmental factor value-enzyme sequence-substrate structure can refer to the description in step S21 and can be collected through experiments.
[0074] S52, input the environmental factor value-enzyme sequence-substrate structure corresponding to the enzyme to be tested into an enzyme kinetic parameter prediction model, and use the enzyme kinetic parameter prediction model to output the enzyme kinetic parameters of the enzyme to be tested, wherein the enzyme kinetic parameters include the enzyme turnover number, and the enzyme kinetic parameter prediction model is obtained using the enzyme kinetic parameter prediction model training method.
[0075] In one embodiment, the application process of the enzyme kinetic parameter prediction model is roughly the same as the training process of the enzyme kinetic parameter prediction model. Specifically, after the environmental factor value-enzyme sequence-substrate structure of the enzyme to be tested is input into the prediction model, the first model of the basic layer of the prediction model outputs the prediction parameter Y1 according to the enzyme sequence-substrate structure of the enzyme to be tested, and the second model of the basic layer of the prediction model outputs the prediction parameter Y2 according to the environmental factor value-enzyme sequence-substrate structure of the enzyme to be tested, and the meta-layer of the prediction model outputs the final prediction parameter Y according to the prediction parameter Y1 and the prediction parameter Y2.
[0076] FIG6 is a structural diagram of an enzyme kinetic parameter prediction model training device provided in one embodiment of the present application.
[0077] In some embodiments, the enzyme kinetic parameter prediction model training device 60 may include multiple functional modules composed of computer program segments. The computer program of each program segment in the enzyme kinetic parameter prediction model training device 60 may be stored in the memory of an electronic device and executed by at least one processor to perform the enzyme kinetic parameter prediction model training function (see FIG. 2 for details).
[0078] In the present embodiment, the enzyme kinetics parameter prediction model training device 60 can be divided into multiple functional modules according to the functions performed by it. The functional modules may include: acquisition module 601, training module 602, optimization module 603. The module referred to in the present application refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in a memory. In the present embodiment, the limitation of the enzyme kinetics parameter prediction model training device 60 can be found in the above limitation of the enzyme kinetics parameter prediction model training method, which will not be described in detail here.
[0079] The acquisition module 601 is used to acquire sample data, which includes a four-element data set composed of four-element data, and the four-element data includes environmental factor value-enzyme sequence-substrate structure-first real enzyme kinetic parameter.
[0080] The training module 602 is used to divide the sample data into a training set and a test set, and use the training set to train an enzyme kinetic parameter prediction model.
[0081] The optimization module 603 is used to determine the prediction accuracy of the enzyme kinetic parameter prediction model based on the test set, and optimize the prediction model according to the prediction accuracy.
[0082] 1 , the memory 11 stores a computer program that, when executed by the at least one processor 12, implements all or part of the steps in the enzyme kinetic parameter prediction method. The memory 11 includes a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.
[0083] Furthermore, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of the blockchain node, etc.
[0084] In one embodiment of the present application, a computer program is stored on the computer-readable storage medium, and when the computer program is executed by the processor 12 , the process shown in FIG. 2 is implemented.
[0085] In some embodiments, the at least one processor 12 is the control core (Control Unit) of the electronic device 1, which uses various interfaces and lines to connect the various components of the entire electronic device 1, and executes the various functions and processes data of the electronic device 1 by running or executing the programs or modules stored in the memory 11, and calling the data stored in the memory 11. For example, when the at least one processor 12 executes the computer program stored in the memory, it implements all or part of the steps of the enzyme kinetic parameter prediction method described in the embodiment of the present application; or implements all or part of the functions of the enzyme kinetic parameter prediction device. The at least one processor 12 can be composed of an integrated circuit, for example, it can be composed of a single packaged integrated circuit, or it can be composed of multiple integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors and a combination of various control chips.
[0086] In some embodiments, the at least one communication bus 13 is configured to implement connection and communication between the memory 11 and the at least one processor 12 .
[0087] Although not shown, the electronic device 1 may also include a power source (such as a battery) to power various components. Preferably, the power source may be logically connected to the at least one processor 12 via a power management device, thereby enabling the power management device to manage charging, discharging, and power consumption. The power source may also include one or more DC or AC power sources, a recharging device, a power failure detection circuit, a power converter or inverter, a power status indicator, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, cameras, etc., which will not be described in detail here.
[0088] The above-mentioned integrated unit implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, electronic device, or network device, etc.) or a processor to execute part of the method described in each embodiment of the present application.
[0089] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the module division is only a logical function division, and other division methods may be used in actual implementation.
[0090] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical units, and may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected to achieve the purpose of this embodiment based on actual needs.
[0091] In addition, the functional modules in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0092] It is obvious to those skilled in the art that the present application is not limited to the details of the above-mentioned exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the spirit or basic characteristics of the present application. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and scope of the equivalent elements of the claims are included in the present application. Any figure mark in the claims should not be regarded as limiting the claims involved. In addition, it is obvious that the word "including" does not exclude other units or, and the singular does not exclude the plural. Multiple units or devices stated in the specification may also be implemented by one unit or device through software or hardware. Words such as first, second, etc. are used to indicate names and do not indicate any particular order.
[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.
Claims
1. A method for training an enzyme kinetic parameter prediction model, characterized in that: The method comprises: Acquire sample data, wherein the sample data includes a quaternary data set consisting of quaternary data, wherein the quaternary data includes an environmental factor value-an enzyme sequence-a substrate structure-a first real enzyme kinetic parameter; Dividing the sample data into a training set and a test set, and using the training set to train an enzyme kinetic parameter prediction model; The prediction accuracy of the enzyme kinetic parameter prediction model is determined based on the test set, and the prediction model is optimized according to the prediction accuracy.
2. The enzyme kinetic parameter prediction model training method according to claim 1, characterized in that: The environmental factor value includes a pH value and / or a temperature value.
3. The enzyme kinetic parameter prediction model training method according to claim 1, characterized in that: The sample data also includes a ternary data set consisting of ternary data, wherein the ternary data includes enzyme sequence-substrate structure-second real enzyme kinetic parameters.
4. The enzyme kinetic parameter prediction model training method according to claim 3, characterized in that: The dividing the sample data into a training set and a test set comprises: Using the ternary data set as a first training set; The quaternion data set is divided into a second training set and a test set according to a preset first ratio, and the second training set is divided into a third training set and a fourth training set according to a preset second ratio.
5. The enzyme kinetic parameter prediction model training method according to claim 1, characterized in that: The enzyme kinetic parameter prediction model includes a two-layer architecture, which includes a basic layer and a meta-layer. The basic layer includes a first model and a second model, and the meta-layer includes a machine learning model. The input of the machine learning model includes the output of the first model and the output of the second model.
6. The enzyme kinetic parameter prediction model training method according to claim 5, characterized in that: The method of using the training set to train the enzyme kinetic parameter prediction model comprises: Using a first training set in the training set to train the first model, the first model outputting a first prediction parameter of the enzyme according to the enzyme sequence-substrate structure in the first training set; The second model is trained using a third training set in the training set, wherein the second model outputs a second prediction parameter of the enzyme according to the environmental factor value-enzyme sequence-substrate structure in the third training set; Training the machine learning model based on the fourth training set in the training set includes: inputting the enzyme sequence-substrate structure in the fourth training set into the trained first model, and the first model outputs the third prediction parameter of the enzyme; inputting the environmental factor value-enzyme sequence-substrate structure in the fourth training set into the trained second model, and the second model outputs the fourth prediction parameter of the enzyme, and using the third prediction parameter and the fourth prediction parameter to train the machine learning model.
7. The enzyme kinetic parameter prediction model training method according to claim 6, characterized in that: Determining the prediction accuracy of the enzyme kinetic parameter prediction model using the test set includes: Inputting the environmental factor value-enzyme sequence-substrate structure in the test set into the enzyme kinetic parameter prediction model to obtain a fourth prediction parameter of the enzyme output by the enzyme kinetic parameter prediction model; The predicted enzyme kinetics parameter is determined based on the first true enzyme kinetic parameter in the test set and the fourth prediction parameter. Measure accuracy.
8. A method for predicting enzyme kinetic parameters, characterized in that: The method comprises: Obtain the environmental factor value-enzyme sequence-substrate structure corresponding to the enzyme to be tested; The environmental factor value-enzyme sequence-substrate structure corresponding to the enzyme to be tested is input into an enzyme kinetic parameter prediction model, and the enzyme kinetic parameter prediction model is used to output the enzyme kinetic parameters of the enzyme to be tested, wherein the enzyme kinetic parameters include the enzyme turnover number, wherein the enzyme kinetic parameter prediction model is obtained using the enzyme kinetic parameter prediction model training method as described in any one of claims 1 to 7.
9. An electronic device, characterized in that: The electronic device includes a processor and a memory, and the processor is used to implement the enzyme kinetic parameter prediction model training method as described in any one of claims 1 to 7, or implement the enzyme kinetic parameter prediction method as described in claim 8 when executing the computer program stored in the memory.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the enzyme kinetic parameter prediction model training method as described in any one of claims 1 to 7, or implements the enzyme kinetic parameter prediction method as described in claim 8.
Citation Information
Patent Citations
Prediction method for ubiquitination degradation of target protein
CN109785902A
Methods for enzyme engineering
WO2022254192A1
Cited By
Enzyme kinetic parameter prediction model training and prediction method and related equipment
CN116580776A
