Liposome construction recommendation method and device based on drug molecular structure

Abstract

The invention discloses a drug molecular structure-based liposome construction recommendation method and device, and the method comprises the steps: receiving the molecular information of a target drug inputted by a user, and carrying out the data preprocessing of the molecular information; and then, recommendation is performed based on the standard molecular feature data through a pre-trained liposome recommendation model. And finally, optimizing the construction parameters based on the target liposome construction type to obtain target liposome construction parameters. According to the method, the constructed liposome structure type is recommended based on the drug molecule information through the pre-trained neural network model, and the construction parameters of the liposome are optimized to obtain the optimal liposome construction parameters. The molecular structure characteristics of the drug are directly associated with a liposome construction scheme, and the efficiency and accuracy of liposome construction are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and more specifically, to a method and device for recommending liposome construction based on drug molecular structure. Background Art

[0002] As a new type of drug carrier, liposomes are widely used in the field of drug delivery and controlled release due to their good biocompatibility, drug protection and targeting. The key to liposome construction is to determine the appropriate construction type and key parameters according to the structure and physicochemical properties of different drug molecules to achieve the best drug delivery effect.

[0003] The construction of liposomes involves complex processes and diverse parameter adjustments. At present, liposome construction mainly relies on expert experience to design liposome composition, ratio and process parameters during the construction process. However, when liposome construction is based on expert experience or literature search, the extraction and analysis of drug molecular features lack standardization, making it difficult to directly associate drug molecular features with appropriate liposome construction schemes, resulting in low construction efficiency. Summary of the invention

[0004] In view of this, the present invention provides a recommended method and device for liposome construction based on drug molecular structure and physicochemical properties, which is used to solve the problem of low construction efficiency in existing liposome construction methods, in which drug molecular characteristics are difficult to directly associate with suitable liposome construction schemes.

[0005] To achieve the above objectives, the proposed solution is as follows: A recommended method for constructing liposomes based on the molecular structure of a drug, the method comprising: Receiving molecular information of a target drug input by a user; Preprocess the molecular information to obtain standard molecular characteristic data; The pre-trained liposome recommendation model makes recommendations based on standard molecular feature data to obtain the target liposome construction type; The construction parameters are optimized based on the target liposome construction type to obtain the target liposome construction parameters.

[0006] Preferably, the process of performing data preprocessing on the molecular information includes: Perform format conversion to convert molecular information into a standard format; Remove interfering information from molecular information; Molecular energy optimization of target drug molecules; Extract the physical and chemical characteristics of the target drug molecules and obtain standard molecular characteristic data.

[0007] Preferably, the standard molecular characteristic data include at least one of molecular weight, LogP, polar surface area, number of hydrogen bond donors, number of hydrogen bond acceptors, charge distribution and free energy.

[0008] Preferably, the method further comprises: Stores molecular information and target liposome construction parameters.

[0009] Preferably, the target liposome construction type is a unilamellar liposome, a multilamellar liposome, a targeted liposome, a thermosensitive liposome, a PEG liposome or a double-loaded liposome.

[0010] Preferably, the construction parameters include: particle size, lipid / cholesterol ratio and charge potential.

[0011] Preferably, the method further comprises: Displays the target liposome construction type and target liposome construction parameters.

[0012] A recommended device for constructing liposomes based on the molecular structure of a drug, the device comprising: A drug molecule input module is used to receive the molecular information of the target drug input by the user; The drug molecule preprocessing module is used to preprocess the molecular information to obtain standard molecular characteristic data; A liposome construction recommendation module is used to make recommendations based on standard molecular feature data using a pre-trained liposome recommendation model to obtain a target liposome construction type; The parameter optimization module is used to optimize the construction parameters based on the target liposome construction type to obtain the target liposome construction parameters.

[0013] Preferably, the device further comprises: The data storage and management module is used to store molecular information and target liposome construction parameters.

[0014] Preferably, the device further comprises: The result display module is used to display the target liposome construction type and target liposome construction parameters.

[0015] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: The liposome construction recommendation method based on drug molecular structure provided by the present invention, after receiving the molecular information of the target drug input by the user, firstly, performs data preprocessing on the molecular information. Then, a liposome recommendation model that has been pre-trained is used to make recommendations based on standard molecular feature data. Finally, the construction parameters are optimized based on the target liposome construction type to obtain the target liposome construction parameters. The present invention recommends the type of liposome structure to be constructed based on drug molecular information through a pre-trained neural network model, and optimizes the liposome construction parameters to obtain the optimal liposome construction parameters. The drug molecular structure characteristics and physicochemical properties are directly associated with the liposome construction scheme, thereby improving the efficiency and accuracy of liposome construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0017] Figure 1 A flowchart of a recommended method for constructing liposomes based on drug molecular structure provided in an embodiment of the present invention; Figure 2 A schematic diagram of a recommended device structure for constructing liposomes based on the molecular structure of a drug is provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] First, combine Figure 1 The recommended method for constructing liposomes based on the drug molecular structure provided in the embodiments of the present invention is introduced as follows: Figure 1 As shown, the recommended methods include: Step S01, receiving molecular information of a target drug input by a user.

[0020] Specifically, files containing molecular information of target drugs in different formats input by users can be received, such as SMILES, PDB, MOL formats, or molecular structures drawn by chemdraw.

[0021] Step S02, performing data preprocessing on the molecular information.

[0022] Specifically, first, format conversion is performed to convert the molecular information into a standard format. The molecular structure format conversion can be performed through Open Babel to convert the format of the input file containing molecular information (SMILES, PDB, MOL format) into a unified standard format.

[0023] Then, the interference information in the molecular information can be removed by PyMOL, where the interference information includes water molecules, organic solvents, unnecessary small molecules, etc., thereby ensuring the purity of the drug molecular structure.

[0024] Then, we can optimize the molecular energy of the target drug molecule through ORCA, and obtain a relatively stable molecular structure of the target drug, providing basic data for subsequent calculations and experiments. The whole process can be visualized through PyMOL.

[0025] Finally, the physical and chemical characteristics of the target drug molecules can be extracted through RDKit to obtain standard molecular characteristic data, such as molecular weight, LogP, polar surface area, number of hydrogen bond donors, number of hydrogen bond acceptors, charge distribution and free energy.

[0026] Molecular Weight (MW): Basic information about the molecular size of the target drug.

[0027] LogP: The lipid solubility of the target drug molecule, which mainly affects the behavior of the target drug molecule in liposomes.

[0028] Polar Surface Area (PSA): used to predict the membrane penetration ability of target drug molecules.

[0029] Number of hydrogen bond donors (HBD): The interaction characteristics between the target drug molecule and the surrounding environment.

[0030] Number of hydrogen bond acceptors (HBA): the ability of target drug molecules to interact with solvents and lipids.

[0031] Charge Distribution: The surface charge of target drug molecules affects their behavior in aqueous solution.

[0032] Free Energy: An important parameter that describes the stability of target drug molecules.

[0033] Step S03: making recommendations based on standard molecular feature data using a pre-trained liposome recommendation model.

[0034] Specifically, a liposome recommendation model that has been trained in advance is used to recommend liposome construction types based on standard molecular feature data to obtain a target liposome construction type. The liposome recommendation model can be a feedforward neural network model built using TensorFlow. The liposome recommendation model is learned and predicted based on the structural feature data of drug molecules, using the structural features of training drug molecules as training samples and the liposome construction types as training labels. Among them, liposome construction types include unilamellar liposomes, multilamellar liposomes, targeted liposomes, thermosensitive liposomes, PEG liposomes, and double-loaded liposomes, and the target liposome construction type is one of them.

[0035] Step S04, optimizing the construction parameters based on the target liposome construction type.

[0036] Specifically, the construction parameters can be optimized based on spatial coordinate information, target liposome construction type, and standard molecular feature data through a pre-trained parameter optimization model to obtain target liposome construction parameters. Scikit-learn can be used to optimize the construction parameters based on spatial coordinate information, target liposome construction type, and standard molecular feature data to ensure the stability and high encapsulation efficiency of liposome construction. Among them, the construction parameters include particle size, lipid / cholesterol ratio, and charge potential.

[0037] Particle Size: The physical size of liposomes (unit: nm).

[0038] Lipid-to-Cholesterol Ratio: Common range is 1:1 to 5:1.

[0039] Zeta Potential: The surface charge value of liposomes, usually -50 to 100 mV.

[0040] A liposome construction recommendation method based on drug molecular structure provided by an embodiment of the present invention, after receiving the molecular information of the target drug input by the user, first, the molecular information is preprocessed. Then, a liposome recommendation model that has been pre-trained is used to make recommendations based on standard molecular feature data. Finally, the construction parameters are optimized based on the target liposome construction type to obtain the target liposome construction parameters. The present invention recommends the type of liposome structure to be constructed based on drug molecular information through a pre-trained neural network model, and optimizes the liposome construction parameters to obtain the optimal liposome construction parameters. The drug molecular structure characteristics are directly associated with the liposome construction scheme, thereby improving the efficiency and accuracy of liposome construction.

[0041] In order to make more accurate liposome construction recommendations, based on the liposome construction recommendation method based on the drug molecular structure of the aforementioned embodiment, the liposome construction recommendation method of the embodiment of the present invention may further include the following steps: Stores molecular information and target liposome construction parameters.

[0042] Specifically, the molecular information and target liposome construction parameters can be stored through a pre-built database. The drug molecular information and liposome construction recommendation data can be stored through a MongoDB database. The molecular information uploaded by the user, the extracted standard molecular feature data, the recommended target liposome construction type and the target liposome construction parameters can be stored for subsequent model training and recommendation improvement.

[0043] The database can contain molecular information of related drugs from known targeted drug sources collected from ChemDraw. The information collected in the database mainly consists of two parts. One is the basic information related to the collected drug molecules, such as the name of the drug molecule, spatial coordinates, related IDs, etc., which are used as information annotations and as information carriers. The other is the extracted standard molecular feature data, etc.

[0044] Furthermore, user feedback data on target liposome construction types and target liposome construction parameters can also be received and stored. The feedback data can be used to improve the liposome recommendation model and parameter optimization model, thereby improving the accuracy of liposome construction scheme recommendation.

[0045] In addition, the following steps may be included: Displays the target liposome construction type and target liposome construction parameters.

[0046] Specifically, the recommended results of the liposome construction scheme are visualized. The visualization can be realized based on the React Native framework. The displayed content may include: target liposome construction type, target liposome construction parameters, and recommended construction parameter ranges.

[0047] Next, the embodiment of the present invention introduces a liposome recommendation model, which can be constructed by TensorFlow, wherein TensorFlow is a classic binary classification model used for classification problems. In the present invention, the liposome construction type is recommended based on the input feature data.

[0048] The input feature data can be the spatial coordinate information of the drug molecule and the standard molecular feature data. The number of output layer nodes corresponds to the number of liposome construction types. Using Softmax as the activation function, the value of each output node is converted into a probability value, and the sum of these probability values ​​is 1. The output value of each node represents the prediction confidence of the construction type. Categorical Crossentropy can be used as the loss function to train the liposome recommendation model. The liposome recommendation model formula is as follows: in, z i It is i The raw values ​​(logits) of the output nodes, z j No. j The original value of the output node, n is the number of output nodes, P ( y i ) is the output node i The predicted probability of the corresponding liposome construction type.

[0049] Liposome recommendation model training data processing process: Deduplication: Check whether there are duplicate molecular entries in the dataset and remove the duplicate data.

[0050] Missing value processing: If the physical and chemical properties of some molecules are missing, the mean, median or interpolation method can be used to fill in the missing values; if important fields (such as molecular weight, LogP) are missing, the incomplete data can be directly eliminated.

[0051] Feature vector processing: Convert the physical and chemical properties of the molecule into feature vectors. Among them, molecular weight, LogP, PSA, HBD, HBA, charge distribution and free energy can be directly retained as numerical values; charge distribution and free energy need to be normalized. If the molecular information contains chemical structure, tools (such as RDKit) can be used to extract molecular fingerprints as supplementary features.

[0052] Label liposome types: Attach a liposome construction type to each drug molecule as a training label.

[0053] Data standardization: Standardize all feature vectors, such as Z-Score standardization.

[0054] The embodiment of the present invention introduces a parameter optimization model. The parameter optimization model of the embodiment of the present invention can be constructed based on Scikit-learn. The Scikit-learn model is mainly used to optimize the construction parameters during the liposome construction process. In the present invention, the liposome construction parameters are optimized according to the input spatial coordinate information, standard molecular feature data and target liposome construction type.

[0055] The parameter optimization model can adopt the support vector regression model. The core of the support vector regression model is to find the best hyperplane to ensure the minimum deviation between the predicted value and the true value. The polynomial kernel is suitable for situations where there is a polynomial relationship between the input features and the target. The parameter optimization model is set with a penalty coefficient, an error range without penalty, and kernel function parameters. Among them, the penalty coefficient is used to control the tolerance of the parameter optimization model to errors; the error range without penalty is the error range that the target value can fall into; the kernel function parameters are used to control the complexity of data mapping.

[0056] In addition, the parameter optimization model can be evaluated by the mean square error and the coefficient of determination. The mean square error (MSE) can be used to measure the error between the predicted value and the true value; the coefficient of determination (R² Score) can be used to measure the explanatory power of the parameter optimization model for the target variable, ranging from [0,1]. The closer to 1, the better the fitting effect of the parameter optimization model.

[0057] Taking the liposome recommendation model with 3 output nodes, corresponding to single-layer liposomes, multi-layer liposomes and targeted liposomes, as an example, the training data processing process of the parameter optimization model is introduced as follows: The target liposome construction type output by the liposome recommendation model is converted into a category label. The corresponding relationship between each liposome construction type and the category label is: unilamellar liposome: 0; multilamellar liposome: 1; targeted liposome: 2.

[0058] The construction parameters of each liposome type are obtained from experiments or literature as training labels, wherein parameters such as particle size and charge potential are normalized.

[0059] The following describes a recommended device for constructing liposomes based on drug molecular structures provided in an embodiment of the present invention. The recommended device for constructing liposomes based on drug molecular structures described below and the recommended method for constructing liposomes based on drug molecular structures described above can be referenced to each other.

[0060] First combine Figure 2, the recommended device for constructing liposomes based on the drug molecular structure is introduced, such as Figure 2 As shown, the recommended device for constructing liposomes based on the drug molecular structure may include: The drug molecule input module 100 is used to receive the molecular information of the target drug input by the user and transmit it to the drug molecule preprocessing module 200 .

[0061] The drug molecule preprocessing module 200 is used to perform data preprocessing on the molecular information to obtain standard molecular characteristic data; A liposome construction recommendation module 300 is used to make recommendations based on standard molecular feature data using a pre-trained liposome recommendation model to obtain a target liposome construction type; The parameter optimization module 400 is used to optimize the construction parameters based on the target liposome construction type to obtain the target liposome construction parameters.

[0062] Further, based on the recommended device for constructing liposomes in the aforementioned embodiment, the device may further include: The data storage and management module is used to store molecular information and target liposome construction parameters.

[0063] Further, based on the recommended device for constructing liposomes in the aforementioned embodiment, the device may further include: The result display module is used to display the target liposome construction type and target liposome construction parameters.

[0064] An embodiment of the present invention further provides a storage medium, which can store a program suitable for execution by a processor, wherein the program is used to implement each processing flow in the aforementioned recommended scheme for constructing liposomes based on drug molecular structure.

[0065] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0066] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0067] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A recommended method for constructing liposomes based on drug molecular structure, characterized in that: Methods include: Receiving molecular information of a target drug input by a user; Preprocess the molecular information to obtain standard molecular characteristic data; The pre-trained liposome recommendation model makes recommendations based on standard molecular feature data to obtain the target liposome construction type; The construction parameters are optimized based on the target liposome construction type to obtain the target liposome construction parameters.

2. The recommended method for constructing liposomes based on drug molecular structure according to claim 1, characterized in that: The process of performing data preprocessing on the molecular information includes: Perform format conversion to convert molecular information into a standard format; Remove interfering information from molecular information; Molecular energy optimization of target drug molecules; Extract the physical and chemical characteristics of the target drug molecules and obtain standard molecular characteristic data.

3. The recommended method for constructing liposomes based on drug molecular structure according to claim 1, characterized in that: The standard molecular characteristic data include at least one of molecular weight, LogP, polar surface area, number of hydrogen bond donors, number of hydrogen bond acceptors, charge distribution and free energy.

4. The recommended method for constructing liposomes based on drug molecular structure according to claim 1, characterized in that: The method also includes: Stores molecular information and target liposome construction parameters.

5. The recommended method for constructing liposomes based on drug molecular structure according to claim 1, characterized in that: The target liposome construction type is a single-layer liposome, a multi-layer liposome, a targeted liposome, a temperature-sensitive liposome, a PEG liposome or a double-loaded liposome.

6. The recommended method for constructing liposomes based on drug molecular structure according to claim 1, characterized in that: The construction parameters include: particle size, lipid / cholesterol ratio and charge potential.

7. The recommended method for constructing liposomes based on drug molecular structure according to any one of claims 1 to 6, characterized in that: The method also includes: Displays the target liposome construction type and target liposome construction parameters.

8. A liposome construction recommendation device based on drug molecular structure, characterized in that: The device includes: A drug molecule input module is used to receive the molecular information of the target drug input by the user; The drug molecule preprocessing module is used to preprocess the molecular information to obtain standard molecular characteristic data; A liposome construction recommendation module is used to make recommendations based on standard molecular feature data using a pre-trained liposome recommendation model to obtain a target liposome construction type; The parameter optimization module is used to optimize the construction parameters based on the target liposome construction type to obtain the target liposome construction parameters.

9. The device for recommending liposome construction based on drug molecular structure according to claim 8, characterized in that: The device also includes: The data storage and management module is used to store molecular information and target liposome construction parameters.

10. The device for recommending liposome construction based on drug molecular structure according to any one of claims 8 to 9, characterized in that: The device also includes: The result display module is used to display the target liposome construction type and target liposome construction parameters.

Images