Method for predicting the structure of three-dimensional complexes between proteins and ligands using co-folding
The integrated Co-Folding platform automates protein-ligand binding prediction by integrating sequence processing, binding site and PTM detection, and result visualization, enhancing accuracy and reliability through automated detection and visualization of complex structures.
Patent Information
- Application Number
- KR1020250199898
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2025-12-16
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2045-12-16
AI Technical Summary
Conventional protein-ligand binding prediction methods require manual identification of binding sites and lack integration of Post-Translational Modifications (PTMs, leading to reduced accuracy and inefficiency in data processing and analysis.
An integrated Co-Folding platform that automates sequence processing, binding site and PTM integration, and result visualization to predict the binding structure between proteins and ligands, using a processor to specify target proteins and ligands, identify binding and PTM sites, and calculate complex structures in three dimensions.
Enhances the accuracy of protein-ligand binding prediction by objectively evaluating hyperbinding, quantifying structural changes, and improving reliability through automated detection of PTMs and sequence alignment, enabling real-time tracking and intuitive visualization of results.
Smart Images

Figure 112025141903975-PAT00002_ABST
Abstract
Description
Technology Field
[0001] The technical concept of the present disclosure is to a method for predicting the structure of a three-dimensional complex between a protein and a ligand, and more specifically, to a method for predicting the structure of a three-dimensional protein-ligand complex between proteins and ligands that bind to each other using cofolding. Background Technology
[0002] Protein-ligand binding prediction technology has played a pivotal role in the fields of new drug development and protein function analysis. Conventional binding prediction methods have primarily involved performing unidirectional binding simulations based on the protein's three-dimensional structure or individually exploring ligand docking sites. However, these approaches have limitations, such as requiring the user to clearly identify the protein's binding sites in advance and making it difficult to comprehensively consider protein modification information, such as Post-Translational Modifications (PTMs). Furthermore, there are aspects where it is difficult to predict structural changes in a protein when a ligand binds to it.
[0005] In particular, although PTM is a biological phenomenon that significantly affects protein function regulation and ligand binding affinity, conventional technologies often manually identified it at a separate step or managed it without linking it to protein structure prediction. This problem resulted not only in reduced accuracy of binding prediction but also inefficiency in subsequent analysis processes.
[0006] Meanwhile, the data processing for protein-ligand binding prediction was also composed of several independent modules, leading to the problem that each step—such as sequence search, preprocessing, binding structure prediction, and result visualization—was performed individually. Consequently, from the user's perspective, it was difficult to establish an integrated workflow at the project level, and there was a high likelihood of input data inconsistencies or errors occurring between processes.
[0007] Accordingly, the need for an integrated Co-Folding platform to predict ligand binding structures from protein sequences has emerged. In particular, the ability to visually verify and modify automatically suggested binding sites based on an internal database, and the capability to automatically detect PTMs to distinguish important amino acids, can significantly enhance user work efficiency. The problem to be solved
[0008] The technical background of the problem that the technical concept of the present disclosure aims to solve is to provide a system and a method of operation in which automated sequence processing, binding site and PTM integration setup, Co-Folding calculation, and result visualization are organically linked in order to predict the binding structure between a protein and a ligand more accurately and efficiently. means of solving the problem
[0009] According to an embodiment of the present disclosure, a method for predicting a complex binding structure between a protein and a ligand by a processor may include the steps of specifying a target protein and a target ligand for which the complex binding structure is to be predicted, predicting a complex binding structure between the target protein and the target ligand, and outputting a protein-ligand complex structure in three dimensions according to the predicted complex binding structure.
[0010] According to one embodiment, the step of specifying the target protein and target ligand may include the step of specifying the target protein by searching for the protein name from any one of Gene Name, Uniprot, Organism, or Protein Name.
[0011] According to one embodiment, the method for predicting the complex binding structure may further include the step of identifying at least one binding site and at least one PTM site among the target protein and the target ligand, and may be characterized by indicating the sequence code corresponding to at least one of the PTM site and the binding site in a distinguishable manner in a protein sequence listed as a series of sequence codes.
[0012] According to one embodiment, the step of identifying at least one of the at least one binding site and the PTM site may include the step of modifying at least one of the PTM site and the binding site according to a user request.
[0013] According to one embodiment, the method further includes the step of searching a protein similar to the target protein from a protein database for a protein similar to the target protein, and the step of predicting the complex binding structure may be characterized by calculating the complex binding structure for the target protein by reflecting the degree of cofolding between the protein similar and the target ligand.
[0014] According to one embodiment, the step of searching for the similar protein in a protein DB includes the step of extracting a difference between the first sequence of the target protein and the second sequence of the similar protein, and the step of recognizing a pattern of a sequence index and a sequence code for the sequence difference, and the step of predicting the complex binding structure may be characterized by calculating the complex binding structure of the target protein by reflecting the pattern recognized for the similar protein.
[0015] According to one embodiment, the step of outputting the protein-ligand complex structure in three dimensions may be characterized by outputting a hyperbinding result in three dimensions, including the binding strength, binding location, and binding method at the binding site where the target protein and the target ligand are bound.
[0016] According to one embodiment, the step of outputting the protein-ligand complex structure in three dimensions may include the step of displaying the protein-ligand complex structure in a manner separated by sequence segments according to the predicted implementation accuracy for each sequence segment. Effects of the invention
[0017] According to the present invention, by predicting the binding structure between a protein and a ligand, the level of hyperbinding, which was difficult to obtain with existing simple binding prediction techniques, can be objectively evaluated.
[0018] Furthermore, by quantifying and visually representing hyperbinding results, researchers can compare and analyze the impact of each candidate ligand on protein structural changes. For instance, since it is possible to quantitatively distinguish between cases where a specific ligand partially induces the protein folding pathway and cases where it also rearranges the protein's major active site, experimental efficiency in drug design and binding optimization processes is dramatically improved.
[0019] In addition, through the integrated setting function of binding sites and Post-Translational Modifications (PTMs), users can set binding sites by simultaneously considering the structural characteristics and functional modification information of the protein. Since the presence of PTMs is automatically detected and important amino acids are visually distinguished, the accuracy of identifying sites that directly affect binding affinity can be improved. This has the effect of increasing the reliability of structure prediction compared to conventional manual setting methods.
[0020] Through sequence-based automatic preprocessing and integration with external MSA databases, the accuracy of the prediction model can be enhanced by automatically searching for and aligning sequences similar to the input protein sequence. Users do not need to perform complex sequence alignment procedures separately, and the computational quality of complex binding structures is improved as the system automatically incorporates meaningful sequence data.
[0021] Through automated status tracking and condition-linked workflows, the preprocessing, calculation, and visualization stages are organically interconnected, enabling real-time tracking of the task status. Consequently, structural prediction can be reliably performed without errors or data omissions at each stage, allowing users to consistently manage the entire process.
[0022] In addition, through result visualization and iterative editing functions, predicted combination structures can be intuitively checked on a 3D interface, and immediate modifications and recalculations can be performed if necessary. This provides high flexibility in the process of experimentally verifying model training results or performing iterative analysis by assuming various combination conditions.
[0023] The effects obtainable from the exemplary embodiments of the present disclosure are not limited to those mentioned above, and other unmentioned effects can be clearly derived and understood by those skilled in the art to which the exemplary embodiments of the present disclosure belong from the description below. That is, unintended effects resulting from the implementation of the exemplary embodiments of the present disclosure can also be derived by those skilled in the art from the exemplary embodiments of the present disclosure. Brief explanation of the drawing
[0024] FIG. 1 is a drawing illustrating an electronic device according to an embodiment of the present disclosure. FIG. 2 is a flowchart illustrating a method for outputting a protein-ligand complex structure according to an embodiment of the present disclosure. FIG. 3 is a diagram illustrating a method for calculating the complex binding structure of a target protein based on a similar protein according to one embodiment. FIG. 4 is a diagram illustrating sequence information showing binding sites and PTM sites according to one embodiment. FIG. 5 is a diagram illustrating an example in which a hyperbinding result according to one embodiment is calculated. FIG. 6 is a diagram illustrating the structure of a protein-ligand complex drawn in three dimensions according to one embodiment. FIG. 7 is a diagram illustrating binding sites between a protein and a ligand in three dimensions according to one embodiment. FIG. 8 is a diagram illustrating the structure of a protein-ligand complex shown in different ways for each sequence segment based on implementation accuracy according to one embodiment. Specific details for implementing the invention
[0025] Hereinafter, various embodiments of the present disclosure are described in conjunction with the accompanying drawings. As various embodiments of the present disclosure may be subject to various modifications and may have various forms, specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the various embodiments of the present disclosure to specific forms, and it should be understood that they include all modifications and / or equivalents and substitutions that fall within the spirit and scope of the various embodiments of the present disclosure. In relation to the description of the drawings, similar reference numerals have been used for similar components.
[0026] In various embodiments of the present disclosure, terms such as “comprising” or “having” are intended to indicate 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.
[0027] In various embodiments of the present disclosure, expressions such as “or” include any and all combinations of the words listed together. For example, “A or B” may include A, may include B, or may include both A and B.
[0028] Expressions such as "first," "second," "first," or "second" used in various embodiments of the present disclosure may modify various components of the various embodiments, but do not limit such components. For example, such expressions do not limit the order and / or importance of such components and may be used to distinguish one component from another.
[0029] When it is mentioned that a component is "connected" or "joined" to another component, it should be understood that the component may be directly connected or joined to the other component, but that a new component may also exist between the component and the other component.
[0030] In the embodiments of the present disclosure, terms such as "module," "unit," "part," etc. are used to refer to a component that performs at least one function or operation, and such component may be implemented in hardware or software, or in a combination of hardware and software. Additionally, a plurality of "modules," "units," "parts," etc. may be integrated into at least one module or chip and implemented as at least one processor, except where each needs to be implemented in specific individual hardware.
[0031] Terms such as those defined in commonly used dictionaries should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in the various embodiments of the present disclosure.
[0032] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.
[0034] FIG. 1 is a drawing illustrating an electronic device (10) according to an embodiment of the present disclosure.
[0035] Referring to FIG. 1, the electronic device (10) may include a processor (100), RAM (Random Access memory) (200), storage (300), and a communication unit (400). The electronic device (10) may be implemented as various types of devices, for example, a laptop computer, a mobile phone, a smartphone, a tablet PC, a mobile internet device (MID), a personal digital assistant (PDA), an enterprise digital assistant (EDA), a digital still camera, a digital video camera, a portable multimedia player (PMP), a personal navigation device or portable navigation device (PND), a handheld game console, an e-book, or a smart device. The smart device may be implemented as a smart watch, a smart band, or a smart ring.
[0036] The processor (100) can control the overall operation of the electronic device (10). The processor (100) may include a single processor core or a CPU (Central Processing Unit) that includes multiple processor cores. The electronic device (10) may include one or more processors (100).
[0037] For example, the processor (100) may include a microprocessor, a central processing unit, a processor core, a multi-core processor, a multiprocessor, an ASIC (Application-Specific Integrated Circuit), and an FPGA (Field Programmable Gate Array).
[0038] The processor (100) can process or execute programs, data, or instructions stored in storage (300). For example, the processor (100) can identify binding sites or PTM sites of proteins or ligands by executing programs stored in storage (300), and can predict the complex binding structure between the ligand and the protein to output a protein-ligand complex structure in three dimensions.
[0039] RAM (200) can temporarily store programs, data, or instructions. For example, programs and / or data stored in storage (300) may be temporarily stored in RAM (200) under the control of the processor (100) or boot code. For example, RAM (200) includes DRAM (Dynamic RAM), SRAM (Static RAM), SDRAM (Synchronous DRAM), etc.
[0040] Storage (300) is a storage location for storing data and can store an OS (Operating System), various programs, and various data. Storage (300) includes ROM (Read Only Memory), flash memory, PRAM (Phase-change RAM), MRAM (Magnetic RAM), RRAM (Resistive RAM), FRAM (Ferroelectric RAM), etc. In an embodiment, storage (300) can be implemented as an HDD (Hard Disk Drive), SSD (Solid State Drive), etc.
[0041] The communication unit (400) can transmit and / or receive data from the electronic device (10). For example, the communication unit (400) may be configured to communicate with various types of external devices according to various types of communication methods. The communication unit (400) may include at least one of a network chip, a Wi-Fi chip, a Bluetooth chip, a wireless communication chip, and an NFC chip. For example, each of the Wi-Fi chip and the Bluetooth chip can perform communication in a Wi-Fi method and a Bluetooth method, respectively. When using a Wi-Fi chip or a Bluetooth chip, various connection information such as SSID and session key may be transmitted and received first, and then various information may be transmitted and received after establishing a communication connection using this information. A wireless communication chip refers to a chip that performs communication according to various communication standards such as IEEE, Zigbee, 3G (3rd Generation), 3GPP (3rd Generation Partnership Project), LTE (Long Term Evolution), 5G (5th Generation), etc. Wireless communication chips according to 5G communication standards can use not only frequency bands such as 3.5 GHz (Below 6 GHz), but also millimeter wave (mmWave) bands such as 26, 28, 38, 39, and 60 GHz, i.e., ultra-high frequency bands (Above 6 GHz). An NFC chip refers to a chip that operates using the NFC (Near Field Communication) method, which uses the 13.56 MHz band among various RF-ID frequency bands such as 135 kHz, 13.56 MHz, 433 MHz, 860~960 MHz, and 2.45 GHz.
[0042] Accordingly, the electronic device (10) of the present disclosure can perform operations by means of the above configurations, temporarily store data or commands, or perform data transmission and / or reception with other electronic devices.
[0043] The following description in this specification will be explained with reference to the components of FIG. 1.
[0045] FIG. 2 is a flowchart illustrating a method for outputting a protein-ligand complex structure according to an embodiment of the present disclosure.
[0046] Referring to FIG. 2, the electronic device (10) can output a complex structure between a target protein and a target ligand selected by a user in three dimensions by sequentially performing a series of steps. At this time, the shape of the complex structure can be determined according to the complex binding structure.
[0047] In step (S100), the processor (100) can identify the target protein and target ligand for which the complex binding structure is to be predicted. In this step, the user can identify the target protein by searching the protein DB using one of Gene Name, Uniprot, Organism, or Protein Name, and selecting the searched protein. The ligand can be selected from an internal library or an external compound DB. The process of identifying the target protein and target ligand is not limited to this, and the user can designate the structure as the target protein or target ligand by directly drawing or editing and inputting the molecular structure.
[0048] In step (S200), the processor (100) identifies at least one binding site and at least one Post-Translational Modification (PTM) site among the target protein and the target ligand. A binding site refers to a location on the sequence of the target protein where binding with the target ligand is possible, and a PTM site may refer to a location of an amino acid residue where protein modifications, such as phosphorylation or acetylation, occur. In this step, sequence indices corresponding to binding or PTM sites can be automatically detected from a series of sequence codes of the protein and displayed in a visually distinguishable manner. Additionally, the user can modify or add to the automatically detected results to supplement the accuracy of the prediction.
[0049] In step (S300), the processor (100) can predict the complex binding structure for the target ligand of the target protein based on the identified information. The complex binding structure is a result indicating the level of structural stabilization where the protein and the ligand fold while mutually influencing each other, and is predicted by comprehensively considering the sequence characteristics of the protein, the binding energy of the ligand, structural complementarity, etc. The degree of cofolding can be predicted using a hyperbinding value.
[0050] According to one embodiment, the processor (100) can search for a similar protein having a protein sequence similar to the target protein from an external DB and improve the reliability of the calculation by reflecting the sequence pattern of the similar protein. A method for predicting the complex binding structure of the target protein by reflecting the sequence pattern of the similar protein will be described later through FIG. 3.
[0051] In step (S400), the processor (100) outputs the protein-ligand complex structure in a three-dimensional form according to the predicted complex binding structure. In this step, the hyperbinding results, including binding strength, binding location, and binding method, are visualized so that the user can intuitively check the cofolding results. Additionally, since the predicted implementation accuracy can be visually distinguished and displayed for each sequence segment, the binding reliability and structural stability of a specific segment can be clearly determined.
[0052] Thus, the complex binding structure prediction method of the present invention can provide the effect of quantitatively and visually predicting complex interactions between biomolecules by automatically analyzing the binding structure between a protein and a ligand and evaluating structural mutual adaptability in combination with PTM information.
[0054] FIG. 3 is a diagram illustrating a method for calculating the complex binding structure of a target protein based on a similar protein according to one embodiment.
[0055] Referring to FIG. 3, a method for calculating a complex binding structure according to one embodiment can calculate the complex binding structure of a target protein by reflecting the complex binding structure of a similar protein having a sequence similar to that of the target protein. The similar protein may be a protein with a sequence similar to that of the target protein. The similar protein may be identified as the protein with the most similar sequence, but may also be identified as a plurality of proteins whose similarity to the target protein is greater than or equal to a certain threshold.
[0056] In step (S310), the processor (100) can extract the sequence difference between the first sequence of the target protein and the second sequence of the similar protein. In this step, the major sequence segments of the proteins are compared to identify the different parts, and the effect of the difference on the structural characteristics of the protein or ligand binding can be analyzed.
[0057] In step (S320), the processor (100) recognizes the pattern of the sequence index and the sequence code for the extracted sequence difference. Here, the sequence index refers to a specific location on the protein sequence, and the sequence code indicates the type of amino acid at that location. In this step, the characteristics of the protein's binding site or modification site can be identified more specifically by not only determining whether a difference exists, but also by identifying the modification pattern that appears repeatedly in a specific section or the continuity of the preserved sequence.
[0058] In step (S330), the processor (100) calculates the complex binding structure of the target protein based on the pattern recognized in the similar protein. For example, if the sequence pattern of the similar protein shows high similarity at the whole sequence level, it can be determined that the target protein is also likely to form a similar binding structure. Conversely, if the difference in the whole sequence is large, a low weight can be given to the calculation of the degree of cofolding of the target protein, and if the sequence conservation is low, the degree of cofolding can be calculated as low. In this way, the ligand binding stability of the target protein can be predicted more realistically by reflecting the structural characteristics and modification patterns of the similar protein.
[0059] As a result, the process illustrated in Fig. 3 can increase the reliability of cofolding calculations and simultaneously improve the precision of structural similarity analysis between proteins and ligand binding prediction by estimating the structural adaptability of the target protein based on sequence information of similar proteins.
[0061] FIG. 4 is a diagram illustrating sequence information showing binding sites (41) and PTM sites (42) according to one embodiment.
[0062] Referring to FIG. 4, the processor (100) can identify at least one of a binding site (41) and a PTM site (42) in the sequence of a target protein, and can output at least one of the identified binding site (41) and PTM site (42) by representing it with a separate display method.
[0063] FIG. 4(a) is an example in which a region capable of ligand binding within the sequence of a target protein is automatically detected and marked as a binding site (41). Each alphabet represents an amino acid sequence constituting the protein, and the part marked in green corresponds to the binding site (41). The system identifies sequence segments with a high probability of binding through an internal database or a prediction algorithm and presents the sequence index of the corresponding segment (e.g., 718, 719, 790, 791, etc.). Through this marking, the user can intuitively identify the binding site and, if necessary, modify the corresponding segment or add a new site.
[0064] According to one embodiment, the processor (100) may represent different sequence portions in the same target protein as binding sites (41) or PTM sites (42) depending on the type of target ligand. For example, if a specific ligand primarily forms hydrophobic bonds, a region where hydrophobic amino acids are concentrated may be indicated as a binding site (41), and if a different ligand interacts with a phosphorylation site, phosphorylated amino acid residues within the same protein may be highlighted as PTM sites (42). In this way, the processor (100) can dynamically identify candidate binding regions within the protein sequence by considering the chemical properties and binding mechanisms of the ligands, and intuitively visualize different binding patterns for each ligand.
[0065] FIG. 4(b) shows an example in which a PTM site (42) is displayed together in a protein sequence. Here, PTM refers to a phenomenon in which chemical modification occurs at specific amino acid residues after the protein has been synthesized, such as phosphorylation and acetylation. In the figure, phosphorylated threonine located at the 288th sequence position is indicated as the PTM site (42) and is visualized in orange. Additionally, a binding site (41) (green) is displayed together within the same frame, making it easy to identify areas where the PTM site and the binding site are adjacent or overlapping.
[0066] As a result, the configuration illustrated in Fig. 4 visually distinguishes and displays binding sites and modification sites within the protein sequence, thereby clearly understanding the functional core regions of the protein structure and effectively reflecting the factors affecting binding strength and structural stability when predicting co-folding.
[0068] FIG. 5 is a drawing showing the result of calculating the hyperbinding result (51) according to one embodiment.
[0069] Referring to FIG. 5, the processor (100) can calculate the hyperbinding result (51) between the target protein and the target ligand and output it as a quantified value. The hyperbinding result (51) may represent a result quantifying the binding stability between the protein and the ligand in units of energy. The lower the value of the hyperbinding result (51) (the larger the negative absolute value), the more stable the mutual binding between the protein and the ligand. For example, in the figure, the hyperbinding value is shown as -8.9 kcal / mol, which may mean that the corresponding protein-ligand complex has a relatively high level of binding stability.
[0070] The system of the present invention can present these hyperbinding results (51) not merely as simple binding energy values, but as a comprehensive indicator calculated by reflecting various biological factors such as PTM sites, the location of binding sites, and sequence patterns of similar proteins. Through this value, the user can structurally visualize how strongly the ligand binds within the protein structure and how it induces structural deformation.
[0071] According to one embodiment, the processor (100) can predict the complex binding structure between a target ligand and a target protein using a cofolding prediction model. In this case, the processor (100) calculates the complex binding structure through a pre-trained cofolding prediction model using sequence information, structural features, physicochemical properties, etc., of the protein and the ligand as input data.
[0072] Cofolding prediction models learn binding stability, interaction patterns, and structural modification patterns from large-scale protein-ligand complex datasets, and can quantitatively predict expected structural adaptability and binding strength based on the characteristics of the input proteins and ligands. For example, by learning the tendency of specific regions of a protein sequence to interact with specific functional groups of a ligand, the model can infer similar binding patterns even in new combinations.
[0073] In addition, since the cofolding prediction model can simultaneously process multiple forms of data, such as 3D coordinate data, molecular graphs, or binding energy distributions, in addition to sequence-based input, it is possible to make high-precision predictions that reflect the fine structural changes occurring during the protein folding and ligand binding processes.
[0074] Consequently, by applying an AI model to the cofolding prediction method of the present invention, prediction accuracy and generalization performance can be improved through data-based learning effects, and fast and reliable structure prediction can be performed for various protein-ligand combinations.
[0075] In addition, in the embodiment of FIG. 5, the hyperbinding results (51) are displayed in parallel along with the molecular characteristics (molecular weight, polar surface area (TPSA), lipophilicity (LogP), etc.) of each ligand, thereby allowing for an integrated comparison of protein-ligand binding efficiency and physicochemical stability. For example, among several candidate ligands, molecules with high hyperbinding results (51) and low energy values are selected preferentially, thereby enabling the rapid selection of efficient candidates during actual experiments or drug design phases.
[0076] Therefore, the configuration illustrated in Fig. 5 allows for the intuitive evaluation of binding strength and structural interdependence by quantifying and displaying the predicted results of cofolding between proteins and ligands as hyperbinding values, thereby enabling researchers to quantitatively compare and analyze the binding characteristics and cofolding levels of each ligand.
[0078] FIG. 6 is a diagram illustrating the structure of a protein-ligand complex drawn in three dimensions according to one embodiment.
[0079] Referring to FIG. 6, the processor (100) can implement the complex structure between a specified target protein and a target ligand as a three-dimensional object. The target protein (61) has a complex three-dimensional structure composed of multiple helical structures and parallel structures, and the target ligand (62) can be located inside the target protein (61). In the present invention, by expressing this protein-ligand complex structure in three dimensions according to the degree of co-folding, the structural deformation of the protein and the binding location of the ligand can be simultaneously visually confirmed.
[0080] In particular, the target ligand (62) is stabilized by forming hydrogen bonds, hydrophobic interactions, or π-π bonds with various residues near the active site of the protein. This binding state can be indicated in conjunction with the hyperbinding value calculated during the cofolding calculation process, and the higher the binding strength, the more deeply and stably the ligand is inserted into the protein structure can be visualized.
[0081] In addition, the chemical structural formula (63) of the target ligand is displayed on one side of the display, making it easy to identify the correspondence between the arrangement in three-dimensional space and the molecular structure. Through this visualization screen, the user can intuitively check the binding type between a specific functional group of the target ligand and a specific amino acid residue of the protein, and can precisely analyze the distance and interaction direction with the binding site or PTM site.
[0082] Accordingly, Figure 6 specifically illustrates the spatial binding state of a protein-ligand complex generated based on the hyperbinding results predicted in the present invention, and can provide a visual basis for comprehensively verifying the structural adaptability of the protein and the binding efficiency of the ligand.
[0084] FIG. 7 is a diagram illustrating binding sites between a protein and a ligand in three dimensions according to one embodiment.
[0085] According to FIG. 7, the target protein and the target ligand can be bound at at least one functional group, and the binding portion can be indicated by a dotted line. Specifically, different functional groups of the ligand can be bound to specific amino acid residues of the protein, and the bound amino acid residues and functional groups can be represented by connecting objects (71a, 71b, 71c). The ligand forms chemical interactions with various residues within the protein's binding pocket, and these bindings are indicated by dotted connecting objects (71a, 71b, 71c), allowing the location and direction of the binding to be clearly visualized. The connecting objects (71a, 71b, 71c) generally represent hydrogen bonds, van der Waals interactions, or electrostatic interactions, and can be distinguished by thickness or color depending on differences in binding strength or distance.
[0086] This visualization method intuitively illustrates how the ligand's functional group is positioned at specific sites on the protein (e.g., R76, L92, C75, etc.) and can be directly correlated with the binding stability results obtained through co-folding calculations. For instance, if the nitrogen atom of the ligand forms a hydrogen bond with the oxygen atom of the protein, that bond is reflected as a factor that increases binding stability when predicting the complex binding structure.
[0087] Consequently, the configuration illustrated in Fig. 7 visually represents the binding interactions within the protein-ligand complex, thereby enabling specific verification of the structural adaptability of the protein and the binding efficiency of the ligand, and provides a useful schematic basis for analyzing the cofolding results in conjunction with the actual binding form.
[0089] FIG. 8 is a diagram illustrating the structure of a protein-ligand complex shown in different ways for each sequence segment based on implementation accuracy according to one embodiment.
[0090] Referring to FIG. 8, depending on the complex binding structure, the protein-ligand complex structure can be represented in different ways depending on the implementation accuracy of the target protein's three-dimensional structure. Implementation accuracy may be referred to as a confidence value or confidence level. For example, a sequence segment with very high implementation accuracy may be implemented as a first object (81a), a sequence segment with high implementation accuracy may be implemented as a second object (81b), a sequence segment with low implementation accuracy may be implemented as a third object (81c), and a sequence segment with very low implementation accuracy may be implemented as a fourth object. Sequence segments may be divided into a predetermined number of sequence indices, but the number of sequence indices in which sequence segments are divided may not be limited to a specific number.
[0091] According to one embodiment, implementation accuracy corresponds to a prediction reliability index calculated during the cofolding calculation process and can be evaluated by comprehensively considering factors such as binding stability between the protein and the ligand, sequence conservation, and structural match rate.
[0092] In this embodiment, the first object (81a) is indicated in darker blue as the implementation accuracy is very high, which means that the predicted structure has a high degree of agreement with the stable binding form of the actual protein. The high accuracy range can be represented by the second object (81b) in turquoise, the low accuracy range by the third object (81c) in yellow, and the very low accuracy range by the fourth object (81d) in orange. This color distinction allows for intuitive differentiation of the reliability of each sequence region within the structure, enabling the researcher to evaluate the predicted stability around the binding site at a glance.
[0093] For example, if the region adjacent to the ligand binding site is indicated with high confidence while the outer region is indicated with low confidence, this means that the cofolding calculation was performed accurately at the protein's core binding site, but there is relative prediction uncertainty at the terminals or flexible regions.
[0094] Therefore, the configuration illustrated in Fig. 8 goes beyond simply presenting the predicted results of the complex binding structure as numerical values; by visually distinguishing the implementation accuracy by position within the three-dimensional structure, it can provide the effect of allowing the user to intuitively grasp the reliability distribution of the protein structure and adjust the focus of the analysis around the ligand binding site.
[0096] Meanwhile, the methods according to the various embodiments of the present invention described above can be implemented in the form of an application or software program that can be installed on an existing electronic device.
[0097] In addition, the whole or part of the method may be composed of multiple software function modules and implemented on an operating system (OS). Alternatively, each step may be composed of a single software function module, or each step may be combined to form a single software function module and implemented on an operating system. Therefore, even if all of the embodiments of the present disclosure are not implemented as a single software function module, if multiple software function modules implement each step of the present disclosure and multiple software function modules are implemented on a single operating system, it can be understood that the method of the present disclosure has been implemented.
[0098] In addition, the methods according to the various embodiments of the present invention described above can be implemented solely through software upgrades or hardware upgrades of existing electronic devices. Furthermore, the various embodiments of the present invention described above can also be performed through an embedded server equipped in an electronic device or an external server of the electronic device.
[0099] Meanwhile, according to one embodiment of the present invention, the various embodiments described above may be implemented as software comprising instructions stored on a computer-readable recording medium using software, hardware, or a combination thereof. In some cases, the embodiments described herein may be implemented as the processor itself. According to the software implementation, embodiments such as the procedures and functions described herein may be implemented as separate software modules. Each of the software modules may perform one or more functions and operations described herein.
[0100] Meanwhile, a computer or a similar device may include a device according to the disclosed embodiments, which is capable of calling instructions stored from a storage medium and operating according to the called instructions. When said instructions are executed by a processor, the processor may perform a function corresponding to said instructions directly or by using other components under the control of said processor. The instructions may include code generated or executed by a compiler or an interpreter.
[0101] A computer-readable recording medium may be provided in the form of a non-transitory computer-readable recording medium. Here, "non-transitory" simply means that the storage medium does not contain a signal and is tangible, without distinguishing whether data is stored semi-permanently or temporarily on the storage medium. In this context, a non-transitory computer-readable medium refers to a medium that stores data semi-permanently and is readable by a device, rather than a medium that stores data for a short moment, such as registers, caches, or memory. Specific examples of non-transitory computer-readable media may include CDs, DVDs, hard disks, Blu-ray discs, USBs, memory cards, and ROMs.
[0102] As described above, exemplary embodiments have been disclosed in the drawings and specification. Although specific terms have been used to describe the embodiments in this specification, they are used only for the purpose of explaining the technical concept of this disclosure and are not intended to limit the meaning or the scope of this disclosure as defined in the claims. Therefore, those skilled in the art will understand that various modifications and equivalent alternative embodiments are possible therefrom. Accordingly, the true technical scope of protection of this disclosure should be determined by the technical concept of the appended claims.
Claims
Claim 1 A method for predicting a complex binding structure between a protein and a ligand by a processor, comprising: a step of specifying a target protein and a target ligand for which the complex binding structure is to be predicted; a step of searching a similar protein having a protein sequence similar to the target protein from a protein DB; a step of predicting a complex binding structure between the target protein and the target ligand; and a step of outputting a protein-ligand complex structure in three dimensions according to the predicted complex binding structure, wherein the step of predicting the complex binding structure is characterized by calculating the complex binding structure for the target protein by reflecting the degree of cofolding between the similar protein and the target ligand. Claim 2 A method for predicting a complex binding structure according to claim 1, wherein the step of specifying the target protein and target ligand includes the step of specifying the target protein by searching for the protein name from any one of Gene Name, Uniprot, Organism, or Protein Name. Claim 3 The method for predicting a complex binding structure according to claim 1 further comprises the step of identifying at least one binding site and at least one PTM site among the target protein and the target ligand, and is characterized by indicating a sequence code corresponding to at least one of the PTM site and the binding site in a protein sequence listed in a series of sequence codes in a distinguishable manner. Claim 4 A method for predicting a composite binding structure according to claim 3, wherein the step of identifying at least one of the at least one binding site and the PTM site includes the step of modifying at least one of the PTM site and the binding site according to a user request. Claim 5 delete Claim 6 A method for predicting a complex binding structure according to claim 1, wherein the step of searching for the similar protein in a protein DB comprises: a step of extracting a difference between the first sequence of the target protein and the second sequence of the similar protein; and a step of recognizing a pattern of a sequence index and a sequence code for the sequence difference, and the step of predicting the complex binding structure comprises calculating the complex binding structure of the target protein by reflecting the pattern recognized for the similar protein. Claim 7 A method for predicting a complex binding structure according to claim 1, wherein the step of outputting the protein-ligand complex structure in three dimensions is characterized by outputting a hyperbinding result in three dimensions, including binding strength, binding location, and binding mode at a binding site where the target protein and the target ligand are bound. Claim 8 A method for predicting a complex binding structure according to claim 1, wherein the step of outputting the protein-ligand complex structure in three dimensions includes the step of displaying the protein-ligand complex structure in a manner classified by sequence segment according to the predicted implementation accuracy for each sequence segment.
Citation Information
Patent Citations
Method and apparatus for protein-ligand docking using information of given reference complex structure
KR1020240073785A
