Method and apparatus for predicting interactions between peptides and lipid membranes

Abstract

The present invention relates to a method and apparatus for predicting the interaction between a peptide and a lipid membrane. According to one embodiment of the present invention, a method may be provided comprising: a step of inputting an arbitrary first peptide and a first lipid membrane into a previously trained artificial intelligence model; and a step of obtaining first interaction information between the first peptide and the first lipid membrane as the output of the previously trained artificial intelligence model; wherein the first interaction information includes contact map information.

Description

Technology Field

[0001] The present invention relates to a method and apparatus for predicting the interaction between a peptide and a lipid membrane. Specifically, the present invention relates to a method and apparatus for predicting the interaction between a peptide and a lipid membrane by utilizing an artificial intelligence model and molecular dynamics simulation. Background Technology

[0002] The process by which peptides penetrate membranes through interaction with lipid membranes is one of the major processes occurring within the human body. The membrane-penetrating properties of peptides are utilized in various fields, such as drug delivery, anticancer therapy, and vaccine development. Therefore, it is necessary to properly design peptides to enable them to penetrate membranes.

[0003] Conventionally, predictions regarding whether a peptide penetrates a membrane were limited to information about the peptide itself. However, it was difficult to accurately predict the interaction between the peptide and the membrane based on information about the peptide itself. The problem to be solved

[0004] The problem that the present invention aims to solve is to provide a method and apparatus for predicting the interaction between a peptide and a lipid membrane corresponding to membrane permeability, etc., by directly utilizing information on the interaction between a peptide and a lipid membrane as training data for an artificial intelligence model.

[0005] The problems that the present invention aims to solve are not limited to those mentioned above, and other problems and advantages of the present invention not mentioned can be understood from the following description and will be more clearly understood by the embodiments of the present invention. Furthermore, it will be understood that the problems and advantages that the present invention aims to solve can be realized by the means and combinations thereof set forth in the claims. means of solving the problem

[0006] As a technical means for achieving the technical problem described above, a method for predicting the interaction between a peptide and a lipid membrane according to one embodiment of the present invention comprises: a step of inputting an arbitrary first peptide and a first lipid membrane into a previously trained artificial intelligence model; and a step of obtaining first interaction information between the first peptide and the first lipid membrane as the output of the previously trained artificial intelligence model; wherein the first interaction information may include contact map information.

[0007] Additionally, the step of acquiring the first interaction information may include: generating contact map information including distance information between the first peptide and the first lipid membrane that changes over time; and acquiring the first interaction information including the contact map information.

[0008] In addition, the first interaction information may further include at least one of the gradient of the peptide, structural change of the peptide (RMSD, RMSF), electrostatic energy between the peptide and the lipid membrane, change in hydrogen bonding between the peptide and the lipid membrane, and change in hydrogen bonding between the peptide and water molecules.

[0009] Additionally, the method further comprises the steps of: obtaining MD simulation data by applying weighted ensemble MD (Molecular Dynamics) simulation to a second peptide and a second lipid membrane; extracting second interaction information between the second peptide and the second lipid membrane from the MD simulation data; and training the artificial intelligence model by inputting the second peptide and the second lipid membrane as input data for the artificial intelligence model and obtaining the second interaction information as output data for the artificial intelligence model; wherein the second interaction information may include contact map information.

[0010] In addition, the above MD simulation data may be preprocessed to reduce data dimensions or highlight key information.

[0011] An interaction prediction device for a peptide and a lipid membrane according to one embodiment of the present invention comprises: a memory in which at least one program is stored; and a processor that operates by executing the at least one program, wherein the processor inputs an arbitrary first peptide and a first lipid membrane to a previously trained artificial intelligence model, obtains first interaction information between the first peptide and the first lipid membrane as the output of the previously trained artificial intelligence model, and the first interaction information may include contact map information.

[0012] A computer-readable recording medium according to one embodiment of the present invention may include a computer-readable recording medium having a program for executing the above-described method on a computer.

[0013] Other aspects, features, and advantages other than those described above will become clear from the following drawings, claims, and detailed description of the invention. Effects of the invention

[0014] According to one embodiment of the present invention, efficient membrane permeability prediction is possible by directly utilizing information on the interaction between a peptide and a lipid membrane as training data for an artificial intelligence model.

[0015] In addition, based on the above membrane permeability prediction, the time and cost required in the peptide design process can be significantly reduced, and more precise peptide design is possible.

[0016] In addition, the contact map produced in the present invention can be used as auxiliary data for structural verification or additional analysis in the Alphafold program, and can be directly used for structure generation and design in the Rosetta program.

[0017] The effects of the embodiments of the present invention are not limited to the effects mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art from the description in this specification. Brief explanation of the drawing

[0018] FIG. 1 is an example of a method for predicting the interaction between a peptide and a lipid membrane according to one embodiment. FIG. 2 is a flowchart illustrating the steps in which an artificial intelligence model is utilized in a method for predicting the interaction between a peptide and a lipid membrane according to one embodiment. FIG. 3 is a flowchart illustrating the step of training an artificial intelligence model in a method for predicting the interaction between a peptide and a lipid membrane according to one embodiment. FIG. 4 is an exemplary drawing for illustrating the number of contacts that change over time according to one embodiment. FIG. 5a is an exemplary diagram illustrating the slope of a peptide corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment. FIG. 5b is an exemplary diagram illustrating the RMSD of a peptide corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment. FIG. 5c is an exemplary diagram illustrating the RMSF of a peptide corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment. FIG. 5d is an exemplary diagram illustrating the electrostatic energy of a peptide and a lipid membrane corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment. FIG. 5e is an exemplary diagram illustrating changes in hydrogen bonding between a peptide and a lipid membrane and between a peptide and a water molecule, corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment. FIG. 6 is a block diagram of a device for predicting the interaction between a peptide and a lipid membrane according to one embodiment. Specific details for implementing the invention

[0019] The advantages and features of the present invention, and the methods for achieving them, will become clear by referring to the embodiments described in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments presented below, but can be implemented in various different forms and should be understood to include all modifications, equivalents, and substitutions that fall within the spirit and scope of the present invention. The embodiments presented below are provided to make the present invention complete and to fully inform those skilled in the art of the scope of the invention. In describing the present invention, detailed descriptions of related known technologies are omitted if it is determined that such detailed descriptions may obscure the essence of the present invention.

[0020] The terms used herein are used merely to describe specific embodiments and are not intended to limit the invention. Unless otherwise defined, all terms used herein have the same meaning as generally understood by those skilled in the art to which the invention pertains.

[0021] In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. Furthermore, terms such as "comprising" or "having" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0022] Additionally, terms including ordinal numbers, such as "first" or "second" as used herein, may be used to describe various components, but they are used solely for the purpose of distinguishing one component from another. The order and / or importance of the components are not limited by terms including ordinal numbers. For example, a component named as the first component in parts of this specification may be named as the second component without departing from the scope of the technology disclosed herein.

[0023] Phrases such as "in one embodiment," "according to one embodiment," "related to one embodiment," or "according to an implementation of one embodiment" in this specification do not necessarily refer to the same embodiment. Furthermore, throughout this specification, "examples" are arbitrary classifications to facilitate the description of the invention, and each embodiment does not need to be mutually exclusive. For example, configurations mentioned for the description of one embodiment may be applied and / or implemented in other embodiments, and may be modified and applied and / or implemented to the extent that they do not depart from the scope of the invention.

[0024] Some embodiments of the present invention may be represented by functional block configurations and various processing steps. Some or all of these functional blocks may be implemented by various numbers of hardware and / or software configurations that execute specific functions. For example, the functional blocks of the present invention may be implemented by one or more microprocessors or by circuit configurations for a specific function.

[0025] For example, the functional blocks of the present invention may be implemented in various programming or scripting languages. The functional blocks may be implemented as algorithms executed on one or more processors. Additionally, the present invention may employ prior art for electronic configuration, signal processing, and / or data processing. Terms such as "mechanism," "element," "means," and "configuration" may be used broadly and are not limited to mechanical and physical configurations. Furthermore, terms such as "-part" and "-module" refer to a unit that processes at least one function or operation, which may be implemented in hardware or software, or as a combination of hardware and software.

[0026] Furthermore, the connecting lines or connecting members between the components depicted in the drawings are merely illustrative of functional connections and / or physical or circuit connections. In the actual device, connections between components may be represented by various alternative or added functional connections, physical connections, or circuit connections.

[0027] In addition, some components in the drawings may be depicted with their size or proportions slightly exaggerated. Also, components depicted in one drawing may not be depicted in another drawing.

[0028] The present invention will be described in detail below with reference to the attached drawings.

[0029] FIG. 1 is an example of a method for predicting the interaction between a peptide and a lipid membrane according to one embodiment.

[0030] Referring to FIG. 1, the method for predicting the interaction between a peptide and a lipid membrane may consist of an artificial intelligence model learning step (110) and an artificial intelligence model utilization step (120).

[0031] The artificial intelligence model training step (110) may be a step in which a device for predicting the interaction between a peptide and a lipid membrane (hereinafter referred to as the 'device') trains an artificial intelligence model used in the method for predicting the interaction between a peptide and a lipid membrane. The device may be the device (600) of FIG. 6 described later.

[0032] The artificial intelligence model training step (110) may be a step of training an artificial intelligence model based on dynamic data using a weighted ensemble MD simulation described later. The dynamic data may be data that is difficult to obtain with conventional MD simulation methods. In addition, the dynamic data may include interaction information. For example, the dynamic data may include interactions between peptides and lipid membranes, peptide permeation processes, and interactions between peptides.

[0033] The artificial intelligence model utilization step (120) may be a step in which the device uses a previously trained artificial intelligence model to predict information regarding the interaction between the peptide and the lipid membrane, such as membrane permeability.

[0034] In the artificial intelligence model utilization step (120), the device can efficiently perform peptide screening in a short period of time. In addition, the device can design peptides based on the acquired first interaction information. The preceding peptide design can be made more precise by the interaction information between the peptide and the lipid membrane acquired by the artificial intelligence model utilization step (120).

[0035] In machine learning technology and cognitive science, an artificial intelligence model may refer to a statistical learning algorithm implemented based on the structure of a biological neural network, or a structure that executes such an algorithm.

[0036] For example, an artificial intelligence model can represent a model with problem-solving capabilities by having nodes, which are artificial neurons forming a network through the combination of synapses as in biological neural networks, learn to reduce the error between the correct output corresponding to a specific input and the inferred output by repeatedly adjusting the weights of the synapses. For example, an artificial intelligence model may include arbitrary probability models and artificial intelligence models used in artificial intelligence learning methods such as machine learning and deep learning, as well as CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network).

[0037] For example, an artificial intelligence model can be implemented as a multilayer perceptron (MLP) composed of multiple layers of nodes and connections between them. An artificial intelligence model according to the present embodiment can be implemented using one of various artificial neural network model structures including an MLP. For example, an artificial intelligence model may be composed of an input layer that receives input signals or data from the outside, an output layer that outputs output signals or data corresponding to the input data, and at least one hidden layer located between the input layer and the output layer, which receives a signal from the input layer, extracts a feature, and transmits it to the output layer. The output layer receives signals or data from the hidden layer and outputs them to the outside.

[0038] FIG. 2 is a flowchart illustrating the steps in which an artificial intelligence model is utilized in a method for predicting the interaction between a peptide and a lipid membrane according to one embodiment.

[0039] Referring to FIG. 2, in step 210, the device can input the first peptide and the first lipid membrane into a pre-trained artificial intelligence model.

[0040] Step 210 is a step in which the device inputs a first peptide and a first lipid membrane into an artificial intelligence model that has been trained by the artificial intelligence model training method described below. The first peptide and the first lipid membrane may be any peptide and any lipid membrane.

[0041] In step 220, the device can obtain information on the first interaction between the first peptide and the first lipid membrane as the output of a previously trained artificial intelligence model.

[0042] Step 220 is a step in which the device obtains first interaction information of the first peptide and the first lipid membrane as the output of an artificial intelligence model previously trained by the artificial intelligence model learning method described below. The first interaction information may be interaction information occurring between the input first peptide and the first lipid membrane. Additionally, the first interaction information may include contact map information.

[0043] A contact map can be a matrix representing the accessibility between atoms or residues. A residue can refer to an individual unit element of a molecule.

[0044] Contact maps can be useful for analyzing and visualizing information about molecular structures and interactions. One embodiment of the present invention may utilize a contact map between a peptide and a lipid membrane, a contact map between a peptide and a peptide, and a contact map between a peptide and a water molecule.

[0045] A contact map of a peptide and a lipid membrane can represent the interactions between the amino acids of the peptide and the lipid molecules constituting the lipid membrane as a two-dimensional matrix. The rows of the contact map can represent the amino acids of the peptide, and the columns can represent individual lipid molecules or specific regions constituting the lipid membrane. Specific regions may include lipid heads and tails. An example of this is given by Equation 1.

[0046]

[0047] Mathematical Equation 1 is a contact map of a lipid membrane composed of a peptide consisting of 5 amino acids (A1, A2, A3, A4, A5) and 3 lipid molecules (L1, L2, L3). Each row of the matrix can represent the amino acids constituting the peptide, and each column of the matrix can represent the lipid membrane constituting the lipid membrane. A 1 expressed in the matrix indicates contact within a certain distance range, and a 0 indicates no contact within a certain distance range.

[0048] For example, referring to Equation 1, the matrix component corresponding to row 1, column 1 of the contact map may be 1. The fact that the matrix component corresponding to row 1, column 1 of the contact map is 1 may mean that amino acid A1 and lipid molecule L1 are in contact within a certain distance range. The certain distance range can be determined arbitrarily. For example, the certain distance range may be 5 Å. Å is a unit of length and is 0.1 nm.

[0049] The device can construct a contact map using atoms or residues. For example, the device can construct a contact map between a peptide and a lipid membrane by subdividing the amino acids of a peptide and the components of a lipid membrane. Additionally, the device can construct a contact map between a peptide and a peptide by subdividing the components of an amino acid. Furthermore, the device can construct a contact map between a peptide and a water molecule by subdividing the components of an amino acid and a water molecule.

[0050] Additionally, the device can construct a contact map by subdividing the components of atoms or residues. For example, the components of an amino acid may be an alpha carbon, a backbone, a side chain, and a hydrogen atom. For example, the components of a lipid membrane may be a head and a tail. For example, the components of a water molecule may include only oxygen atoms or include both oxygen and hydrogen atoms. An example of a contact map expressed by subdividing the components of a lipid membrane into a head and a tail is given by Equation 2.

[0051]

[0052] Looking at mathematical formula 2, the row consists of 5 amino acids and the column can be divided into a head (L1_head, L2_head, …) and a tail (L1_tail, L2_tail, …).

[0053] For example, the rows may be the five amino acids A1, A2, A3, A4, and A5, and the columns may be L1_head, L2_head, L3_head corresponding to the head portion and L1_tail, L2_tail, L3_tail corresponding to the tail portion. The columns from column 1 to column 6 may be L1_head, L2_head, L3_head, L1_tail, L2_tail, L3_tail, respectively. The meaning of the components constituting the contact map is as described above.

[0054] Random peptides input into an artificial intelligence model each have different characteristics, such as length. Therefore, due to differences in characteristics such as length for each random peptide input, the time at which the interaction between the random peptide input and the lipid membrane occurs may differ.

[0055] A contact map can represent information about the interaction between a peptide and a lipid membrane. Therefore, the contact map acquired by the device by the artificial intelligence model may be multiple contact maps according to the time at which the interaction occurs.

[0056] The device can visualize three-dimensional structures, including the permeation process of peptides, using a graphic program based on a contact map obtained from an artificial intelligence model. In addition, the contact map obtained by the device can be used as auxiliary data for structure verification or further analysis in the Alphafold program, and can be directly used for structure generation and design in the Rosetta program.

[0057] FIG. 3 is a flowchart illustrating the step of training an artificial intelligence model in a method for predicting the interaction between a peptide and a lipid membrane according to one embodiment.

[0058] Referring to FIG. 3, in step 310, the device can obtain MD simulation data by applying weighted ensemble MD simulation to the second peptide and the second lipid membrane.

[0059] The device can observe the spontaneous permeation process of the peptide and acquire information on the interaction with the lipid membrane relatively quickly. The second peptide and the second lipid membrane can be any peptide and lipid membrane.

[0060] Weighted ensemble MD simulation can be composed of a step of setting progress coordinates, a step of dividing progress coordinates into several intervals, a step of managing the number of replica simulations while simulating, and a step of analyzing while iteratively executing.

[0061] More specifically, the step of setting the progress coordinates may be a step of defining group A and group B and then setting the distance between the two groups as the progress coordinates. The groups may be any atoms, residues, or molecules. For example, group A may be a peptide and group B may be a lipid membrane.

[0062] The step of dividing the progression coordinates into multiple segments may be a step of dividing the distance between two groups into multiple segments. In this case, the device may set the segment distances differently depending on the relative positions of the peptide and the lipid membrane. For example, the device may set the segment distance at the center of the lipid membrane (hydrophobic core) to be smaller than the distance of other segments so that the peptide can easily cross the energy barrier. For example, the device may set the segment distance from the center of the lipid membrane to 0.1 Å.

[0063] The step of managing the number of replicated simulations during simulation may be a step of appropriately merging or splitting replicated simulations while running weighted ensemble MD simulations. For example, the device may merge replicated simulations if the number of replicated simulations is concentrated in a specific interval within a certain time interval, and split replicated simulations if it is too low. For example, the certain time interval may be 5 ps.

[0064] The iterative analysis step may be a step of identifying the movement of peptides by repeatedly running weighted ensemble MD simulations. For example, it may be a step of identifying the path the peptides traveled and the depth of penetration. For example, the step of identifying the depth of penetration may be a step of calculating the depth to which the peptides have penetrated.

[0065] In step 320, the device can extract information on the second interaction between the second peptide and the second lipid membrane.

[0066] Step 320 may be a step of extracting second interaction information between the second peptide and the second lipid membrane from MD simulation data obtained by the device through weighted ensemble MD simulation. The second interaction information may be interaction information occurring between the second peptide and the second lipid membrane input to the device. Additionally, the second interaction information may include a contact map. Furthermore, the second interaction information may further include the gradient of the peptide, structural changes of the peptide (RMSD, RMSF), electrostatic energy between the peptide and the lipid membrane, changes in hydrogen bonding between the peptide and the lipid membrane, and changes in hydrogen bonding between the peptide and water molecules.

[0067] In step 330, the device can train the artificial intelligence model by using the second peptide and the second lipid membrane as input data for the artificial intelligence model and acquiring the second interaction information as output data for the artificial intelligence model. The second peptide and the second lipid membrane may be any peptide and any lipid membrane.

[0068] To verify the performance of the above-mentioned artificial intelligence model, the device can separate the dataset into training data, validation data, and test data, and evaluate the prediction performance on the test data. The device may use accuracy, precision, recall, etc., as evaluation metrics. For example, the device can verify the performance of the artificial intelligence model by comparing how well the contact map generated when the artificial intelligence model receives a new peptide sequence matches the results of an actual weighted ensemble MD simulation.

[0069] FIG. 4 is an exemplary drawing for illustrating the number of contacts that change over time according to one embodiment.

[0070] Referring to Fig. 4, the x-axis of the graph represents time, the y-axis represents the value of the number of contacts, and the value of the graph represents the value of the number of contacts over time.

[0071] The device may define Group A and Group B, calculate the distance between the two groups, and set the contact value to 1 if it is within a certain distance range and to 0 otherwise. The certain distance range may be arbitrarily determined. Groups may be any atoms, residues, or molecules. For example, Group A may be a peptide, and Group B may be a lipid membrane. The device may set the sum of all contact values ​​as the contact count value.

[0072] In response to changes in contact values ​​over time, the contact map can change over time. Therefore, the device can acquire multiple contact maps over time. The device can identify the connection status of Group A and Group B based on the number of contacts that change over time. Additionally, the device can identify structural changes based on the connection status of Group A and Group B.

[0073] FIG. 5a is an exemplary diagram illustrating the slope of a peptide corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment.

[0074] Figure 1 (510) is an exemplary diagram showing the inclination angle of a peptide.

[0075] The slope of the peptide may be an angle indicating the degree to which the peptide is tilted relative to the lipid membrane surface. The slope angle may be the angle formed by a specific axis (511) of the peptide with the surface normal vector (512) of the lipid membrane. The specific axis of the peptide may be the central axis of the backbone.

[0076] If a specific axis (511) of the peptide is parallel to the surface normal vector (512) of the lipid membrane, the tilt angle of the peptide may be close to 0 degrees. On the other hand, if a specific axis (511) of the peptide is perpendicular to the surface normal vector (512) of the lipid membrane, the tilt angle of the peptide may be close to 90 degrees. The tilt angle of the peptide may indicate the interaction between the peptide and the membrane. In other words, a low tilt angle may indicate that the peptide is permeable to the membrane or exists in an inserted form. On the other hand, a high tilt angle may indicate that the peptide is not permeable to the membrane but is interacting with the membrane surface.

[0077] Figure 2 (520) is an exemplary diagram showing the change in the slope angle of a peptide during the peptide's permeation process. Referring to Figure 2 (520), the x-axis of the graph represents time, the y-axis represents the slope angle of the peptide, and the values ​​of the graph represent the slope angle over time.

[0078] In addition, referring to Figure 2 (520), the graph can represent the slope angle of the peptide changing over time. The graph can represent the interaction between the peptide and the membrane through the slope angle of the peptide changing over time. For example, referring to Figure 2 (520), it can be seen that the peptide is inserted into the lipid membrane because the slope angle of the peptide decreases after 20 ns. As the slope angle of the peptide changes over time, the interaction between the peptide and the lipid membrane and the interaction between the peptide and water molecules can be understood. In addition, by referring to the slope angle of the peptide changing over time, the contact map changing over time can be effectively understood.

[0079] FIG. 5b is an exemplary diagram illustrating the RMSD of a peptide corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment.

[0080] Referring to Fig. 5b, the x-axis of the graph represents time, the y-axis represents the RMSD value, and the values ​​of the graph represent the RMSD value over time.

[0081] Root Mean Square Deviation (RMSD), which represents structural changes in peptides, can be expressed as the difference in the average distance of atomic coordinates between two structures. Additionally, RMSD can be used to quantitatively evaluate structural changes in molecules. Furthermore, RMSD can be used to analyze whether peptides undergo structural changes while binding to lipid membranes.

[0082] Referring to Fig. 5b, the fact that the change in the RMSD value of the peptide increases over time may mean that the interactions between the peptide and the lipid membrane and between the peptide and water molecules are becoming more active. For example, referring to Fig. 5b, the change in the RMSD value of the red peptide 3 over time may indicate that peptide 3 is permeating the lipid membrane.

[0083] The RMSD value changing over time can indicate that the contact state between the two structures is changing. Therefore, by referring to the RMSD value changing over time, the contact map changing over time can be effectively understood.

[0084] FIG. 5c is an exemplary diagram illustrating the RMSF of a peptide corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment.

[0085] Referring to Fig. 5c, the x-axis of the graph represents the remaining number, the y-axis represents the RMSF value, and the values ​​of the graph represent the RMSF value over time.

[0086] Root Mean Square Fluctuation (RMSF) can be an indicator used to measure the variability over time of individual atoms or residues. Additionally, RMSF can be used to analyze structural flexibility.

[0087] Referring to Fig. 5c, regions with high RMSF values ​​according to residue number may indicate high flexibility, suggesting that the residue at that residue number interacts relatively weakly with the lipid membrane. Regions with low RMSF values ​​according to residue number may indicate low flexibility, suggesting that the residue at that residue number interacts relatively strongly with the lipid membrane. For example, referring to Fig. 5c, the two shaded regions correspond to regions with high RMSF values, indicating that the residue at that residue number interacts relatively weakly with the lipid membrane.

[0088] The RMSF value according to residue number can indicate the interaction state between the residue of the corresponding residue number and the lipid membrane. Therefore, by referring to the RMSF value according to residue number, a contact map indicating the contact between the peptide and the lipid membrane or between the peptide and water molecules can be effectively understood.

[0089] FIG. 5d is an exemplary diagram illustrating the electrostatic energy of a peptide and a lipid membrane corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment.

[0090] Referring to Fig. 5d, the x-axis of the graph represents time, the y-axis represents electrostatic energy, and the values ​​of the graph represent electrostatic energy over time. For example, the blue line may represent the change in electrostatic energy between the peptide and the lower layer of the lipid membrane, and the red line may represent the change in electrostatic energy between the upper and lower layers of the lipid membrane.

[0091] Information on the electrical interactions of a peptide, including the electrostatic energy of the peptide and the lipid membrane, can be defined through electrostatic attraction or repulsion between the charge characteristics of the lipid membrane and the charge distribution of the peptide. Additionally, information on the electrical interactions of a peptide can be used to predict the dynamic behavior and structural characteristics of the peptide.

[0092] Changes in electrostatic energy between a peptide and a lipid membrane can indicate that the peptide interacts with the lipid membrane and penetrates the lipid membrane. Therefore, by referring to changes in electrostatic energy between a peptide and a lipid membrane, a contact map representing the contact between the peptide and the lipid membrane can be effectively understood.

[0093] For example, referring to Fig. 5d, the change of the blue line from a positive (+) value to a negative (-) value over time may indicate that the peptide is moving closer to the layer beneath the lipid membrane as it penetrates the lipid membrane. Additionally, the change of the blue line from a positive (+) value to a negative (-) value over time may indicate that an attractive force is acting between the peptide and the layer beneath the lipid membrane. Therefore, it can be expected that the number of contacts between the peptide and the layer beneath the lipid membrane increases. In this case, the interaction between the peptide and the layer beneath the lipid membrane may be manifested as a change in the contact map.

[0094] FIG. 5e is an exemplary diagram illustrating changes in hydrogen bonding between a peptide and a lipid membrane and between a peptide and a water molecule, corresponding to an example of interaction information between a peptide and a lipid membrane according to one embodiment.

[0095] Referring to Fig. 5e, the x-axis of the graph represents time, the y-axis represents the number of hydrogen bonds, and the values ​​of the graph represent the number of hydrogen bonds over time. For example, the blue line represents the change in the number of hydrogen bonds between the peptide and the lipid molecule during the permeation process, and the red line represents the change in the number of hydrogen bonds between the peptide and the water molecule during the permeation process.

[0096] Hydrogen bonds can be formed between specific residues of a peptide and lipid molecules of a lipid membrane. Additionally, hydrogen bonds can be formed primarily through interactions between the hydrophilic head portion of a lipid and hydrogen-bonding atoms of a peptide.

[0097] Changes in the number of hydrogen bonds can indicate changes in the bonding state between the peptide and the lipid membrane, and between the peptide and water molecules. Additionally, changes in the number of hydrogen bonds can indicate changes in the interactions between the peptide and the lipid membrane, and between the peptide and water molecules. Furthermore, changes in the number of hydrogen bonds may be associated with the presence or absence of contact. Therefore, by referring to the number of hydrogen bonds that change over time, a contact map that changes over time can be effectively understood.

[0098] FIG. 6 is a block diagram of a device for predicting the interaction between a peptide and a lipid membrane according to one embodiment.

[0099] Referring to FIG. 6, the device (600) may include a memory (610) and a processor (620). Only the components related to the embodiment are shown in the device (600) of FIG. 6. Therefore, a person skilled in the art will understand that the device (600) may include other general-purpose components in addition to the components shown in FIG. 6.

[0100] The memory (610) is hardware that stores various data processed within the device (600) and can store programs for various operations, processing, and control of the processor (620).

[0101] The memory (610) may include RAM (Random Access Memory), such as DRAM (Dynamic Random Access Memory) and SRAM (Static Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), CD-ROM, Blu-ray or other optical disc storage, HDD (hard disk drive), SSD (solid state drive), or flash memory.

[0102] The processor (620) controls the overall operation of the device (600). For example, the processor (620) can control the memory (610), an input unit (not shown) and / or an output unit (not shown), etc., by executing programs stored in the memory (610). The processor (620) can control the operation of the device (600) by executing at least one program stored in the memory (610).

[0103] A processor (620) can control at least some of the operations of the device (600) described above with reference to FIGS. 1 to 3. For example, the processor (620) can input any first peptide and a first lipid membrane into a pre-trained artificial intelligence model and obtain first interaction information of the first peptide and the first lipid membrane, including a contact map, as the output of the pre-trained artificial intelligence model, and the first interaction information may include contact map information.

[0104] Meanwhile, a specific example of the operation of the processor (620) is the same as described above with reference to FIGS. 1 to 3. Therefore, a specific description of the operation of the processor (620) is omitted below.

[0105] The processor (620) may be implemented using at least one of ASICs (Application Specific Integrated Circuits), DSPs (Digital Signal Processors), DSPDs (Digital Signal Processing Devices), PLDs (Programmable Logic Devices), FPGAs (Field Programmable Gate Arrays), controllers, microcontrollers, microprocessors, and other electrical units for performing functions.

[0106] In one embodiment, the device (600) may be a server device. The server device may be implemented as a computer device or a plurality of computer devices that communicate through a network to provide commands, code, files, content, services, etc.

[0107] In another embodiment, the device (600) may be a mobile electronic device. For example, the device (600) may be implemented as a smartphone, tablet PC, PC, smart TV, PDA (personal digital assistant), laptop, media player, and other mobile electronic devices.

[0108] In another embodiment, the process performed in the device (600) may be performed by at least some of the server device and the electronic device having mobility.

[0109] Meanwhile, an embodiment according to the present invention may be implemented in the form of a computer program that can be executed through various components on a computer, and such a computer program may be recorded on a computer-readable medium. In this case, the medium may include, but is not limited to, magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, and flash memory.

[0110] Meanwhile, the above-mentioned computer program may be one specifically designed and configured for the present invention, or one known and available to those skilled in the art of computer software. Examples of computer programs may include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0111] According to one embodiment, the method according to various embodiments of the present invention may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices. In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created in a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0112] Unless explicitly stated or contrary to the order of the steps constituting the method according to the present invention, said steps may be performed in a suitable order. The present invention is not necessarily limited by the order in which said steps are described. The use of all examples or exemplary terms (e.g., etc.) in the present invention is merely for the purpose of describing the present invention in detail, and the scope of the present invention is not limited by said examples or exemplary terms unless limited by the claims. Furthermore, those skilled in the art will understand that various modifications, combinations, and changes may be made according to design conditions and factors within the scope of the claims or equivalents to which they are added.

[0113] Accordingly, the scope of the present invention should not be limited to the embodiments described above, and all scopes equivalent to or equivalently modified from the claims set forth below, as well as the claims set forth below, shall be considered to fall within the scope of the concept of the present invention.

Claims

Claim 1 A method for predicting the interaction between a peptide and a lipid membrane, comprising: a step of inputting an arbitrary first peptide and a first lipid membrane into a previously trained artificial intelligence model; and a step of obtaining first interaction information between the first peptide and the first lipid membrane as the output of the previously trained artificial intelligence model; wherein the first interaction information includes contact map information. Claim 2 A method according to claim 1, wherein the step of acquiring the first interaction information comprises: generating contact map information including distance information between the first peptide and the first lipid membrane that changes over time; and acquiring the first interaction information including the contact map information. Claim 3 A method according to claim 1, wherein the first interaction information further comprises at least one of the gradient of the peptide, structural change of the peptide (RMSD, RMSF), electrostatic energy between the peptide and the lipid membrane, change in hydrogen bonding between the peptide and the lipid membrane, and change in hydrogen bonding between the peptide and a water molecule. Claim 4 The method of claim 1 further comprises: a step of obtaining MD simulation data by applying a weighted ensemble MD (Molecular Dynamics) simulation to a second peptide and a second lipid membrane; a step of extracting second interaction information between the second peptide and the second lipid membrane from the MD simulation data; and a step of training the artificial intelligence model by inputting the second peptide and the second lipid membrane as input data of the artificial intelligence model and obtaining the second interaction information as output data of the artificial intelligence model; wherein the second interaction information includes contact map information. Claim 5 In paragraph 4, the MD simulation data is subjected to preprocessing to reduce the dimensionality of the data or to highlight key information. Claim 6 A device for predicting the interaction between a peptide and a lipid membrane, comprising: a memory in which at least one program is stored; and a processor that operates by executing the at least one program, wherein the processor inputs an arbitrary first peptide and a first lipid membrane into a previously trained artificial intelligence model, obtains first interaction information between the first peptide and the first lipid membrane as the output of the previously trained artificial intelligence model, and the first interaction information includes contact map information. Claim 7 A computer-readable recording medium having a program for executing the method according to paragraph 1 on a computer.

Images