KR102517005B1 - Apparatus and method for analyzing relation between mhc and peptide using artificial intelligence - Google Patents

Abstract

Disclosed is a method performed by a computing apparatus. The method includes: acquiring input data including a first set of data and a second set of data corresponding to a specific major histocompatibility complex (MHC) type, wherein the first set of data includes a plurality of peptides, one peptide includes a sequence composed of a plurality of amino acids, and the second set of data includes position-specific amino acid identifiers for the specific MHC type; and acquiring a first prediction result about the binding possibility between the plurality of peptides and the position-specific amino acid identifiers from the input data, using a first model based on artificial intelligence. Therefore, the present invention is capable of efficiently and accurately predicting the binding relation or the binding possibility between an MHC and peptides.

Description

Method and apparatus for analyzing the relationship between MHC and peptide using artificial intelligence technology

The present disclosure relates to an immunopeptidome, and more specifically to analyzing the relationship between a major histocompatibility complex (MHC) and a peptide using artificial intelligence technology.

The major histocompatibility complex is a genetic locus that encodes 'MHC molecules' that function in the immune system. MHC molecules can exist in type 1 (class I) and type 2 (class II).

Immunopeptidome refers to a set of peptides expressed on the cell surface, and for example, the immunopeptidome may refer to a combination of MHC-related peptides.

Human Leukocyte Antigen (HLA) is a glycoprotein molecule produced by the human Major Histocompatibility Complex (MHC) gene. HLA does not exist in mature erythrocytes, but is expressed in immature erythroblasts and is expressed on the surface of all tissue cells in the human body, including blood cells such as leukocytes and/or platelets. The MHC gene is present in all vertebrates, and the human MHC gene is referred to as the HLA gene, and the product expressed therefrom is referred to as HLA.

MHC genes are involved in self and non-self recognition, immune response to antigen stimulation, regulation of cellular and humoral immunity, and susceptibility to disease. HLA, a product of the MHC gene, is the second most important antigen next to the ABO blood type in survival of the transplanted organ in solid organ transplantation. HLA is known to play the most important role in transplantation success or failure in bone marrow transplantation. Therefore, recognizing the difference in HLA immunologically can be seen as the first step of the rejection action for the transplanted tissue. In addition, in transfusion therapy, HLA and antibodies play an important role in the occurrence of various side effects such as platelet transfusion refractory, febrile nonhemorrhagic transfusion side effects, acute lung injury, and graft-versus-host disease after transfusion.

HLA, like MHC, can be largely classified into Class I and Class II. Class I is classified into HLA-A, HLA-B, and HLA-C, and is expressed in most nucleated cells and platelets, and antigen recognition when cytotoxic T cells recognize and remove virus-infected cells or tumor cells. ) is essential for HLA Class II is classified into HLA-DR, HLA-DQ, and HLA-DP, and is expressed in B cells, monocytes, dendritic cells, and activated T cells. and induces a humoral immune response, and is known to be essential when recognizing antigens expressed in antigen-presenting cells. HLA is a gene that shows the largest polymorphism among genes possessed by humans, and there is a frequency difference between races and ethnic groups.

When peptides derived from proteins derived from infectious microorganisms or proteins specific to cancer cells bind to MHC and are presented on the cell surface, T cells recognize them and trigger an immune response to eliminate infected cells or cancer cells. As such, T cells are key regulators (players) that determine specific immune responses to foreign substances that do not exist in the normal human body. Therefore, the prediction of peptides that bind to MHC can be used for the development of peptide vaccines for the prevention of infectious diseases or cancer.

Korean Registered Patent No. 10-2322832

The present disclosure has been made in response to the above background art, and is to predict the binding possibility of MHC and peptides in a more efficient manner and / or more accurately.

The technical problems of the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

A method performed by a computing device is disclosed according to an embodiment of the present disclosure. The method may include obtaining input data including a first set of data and a second set of data corresponding to a specific major histocompatibility complex (MHC) type, the first set of data comprising a plurality of comprising peptides, one peptide comprising a sequence of a plurality of amino acids, and the second set of data comprising positional amino acid identifiers for the specific MHC type; and obtaining a first prediction result for a possibility of binding between the plurality of peptides and the amino acid identifiers for each position from the input data using a first artificial intelligence-based model.

In one embodiment, the first set of data and the second set of data corresponding to a second MHC type different from the first MHC type include the first set of data and the second set corresponding to the first MHC type. may be different from the data of

In one embodiment, the input data includes, as elements, a value representing the possibility of binding between the plurality of peptides included in the first set of data and the amino acid identifiers for each position included in the second set of data It may include a first matrix that

In one embodiment, the first model may be a classification model for predicting binding affinity between peptides and MHC types.

In an embodiment, the first model may include: a first sub-model for extracting a visual feature of a matrix included in the input data; and a second sub-model for analyzing time-series data corresponding to the matrix in response to an input based on the visual feature extracted from the first sub-model.

In one embodiment, the first model includes a plurality of first sub-models and a plurality of second sub-models, and a first output of a 1-1 sub-model among the plurality of first sub-models is is input to a corresponding 2-1 sub-model among the plurality of second sub-models, and a second output of the 1-1 sub-model corresponds to the 1-1 sub-model among the plurality of first sub-models. It can be input to the connected 1st-2nd sub-model.

In an embodiment, an output of each of the plurality of second sub-models may be input to a fully connected layer, and a classification result for the input data may be output.

In one embodiment, the first output includes an output passing through a flatten layer of the 1-1 sub-model, and the second output includes pooling of the 1-1 sub-model. Can include outputs that have passed through layers.

In one embodiment, the first sub-model may use a convolutional neural network-based artificial neural network, and the second sub-model may use a recurrent neural network-based artificial neural network. .

In one embodiment, the second sub-model may use an artificial neural network based on a recurrent neural network to which an attention mechanism is applied.

In an embodiment, the first model includes a plurality of first sub-models, and an output passed through a pooling layer in a 1-1 sub-model among the plurality of first sub-models is the plurality of first sub-models. Among the models, outputs input to the 1-2 sub-model connected to the 1-1 sub-model and outputs passing through the platen layer in each of the plurality of 1-1 sub-models are concatenated to obtain the second sub-model. can be input into the model.

In one embodiment, the first model may include an input layer processing the input data; N feature extraction layers for receiving the output of the input layer and outputting visual features of matrices included in the input data, where N is a natural number; a recurrent neural network layer that receives the output of the feature extraction layer and performs time-sequential learning; and an output layer receiving an output of the recurrent neural network layer and outputting a classification result corresponding to the input data, and a number of classification results corresponding to the number of feature extraction layers may be output.

In one embodiment, the input data may include a second matrix representing binding energies between amino acids of MHC and amino acids of peptides.

In one embodiment, the input data is: a first feature indicating polarity between amino acids of MHC and amino acids of peptides, a second feature indicating the size of amino acids, a third feature indicating hydrophobicity or hydrophilicity of amino acids, a charge of amino acids It may further include at least one of a fourth feature indicating whether the amino acid is present or a fifth feature indicating whether the amino acid is aromatic or aliphatic.

In one embodiment, the method includes: generating a second prediction result for the binding possibility between a peptide and MHC using a second artificial intelligence-based model that takes a plurality of prediction results including the first prediction result as input Further steps may be included.

In one embodiment, the plurality of prediction results include: the first prediction result representing binding affinity between the peptide and MHC; A binding energy value representing the strength of the binding between the peptide and MHC; a cleavage value representing the probability that a specific peptide is cleaved by proteasome and bound to the MHC in the process of MHC being bound to an antigen and being expressed on a cell; a TAP value representing the probability that the peptide cut by the proteasome passes TAP (Transporter Associated with Antigen Processing); a stability value indicating how long the binding between MHC and the peptide lasts; A value representing the degree of difference between the correct sequence of the peptide determined to bind to a specific MHC and the sequence of the specific peptide; and reliability values of immunopeptidome information stored in the database.

In one embodiment, the correct sequence of the peptide may be determined based on at least one of a mass spectrometry test and a T cell receptor (TCR) test.

In one embodiment, the value indicating the degree of difference may be determined based on whether the different location is an anchor location.

In one embodiment, the second model is based on a training data set in which the binding result between a peptide and MHC according to at least one of a mass spectrometry test and a T cell receptor (TCR) test is used as the correct answer data. It can be a classification model.

In one embodiment, the second model, in response to the same input data, outputs generated from each of a logistic regression model, a decision tree model, and a support vector machine model A classification result may be generated by applying voting.

In one embodiment, a computer program stored on a computer readable storage medium is disclosed. The computer program, when executed by a computing device, causes the computing device to perform the following operations: a first set of data and a second set of data corresponding to a particular major histocompatibility complex (MHC) type; Obtaining input data comprising: the first set of data includes a plurality of peptides, one peptide includes a sequence consisting of a plurality of amino acids, and the second set of data includes the specific MHC type Contains amino acid identifiers by position for -; and obtaining, from the input data, a first prediction result for a possibility of binding between the plurality of peptides and the amino acid identifiers for each position by using an artificial intelligence-based first model.

In one embodiment, a computing device is disclosed. The computing device includes at least one processor; and memory. The at least one processor: obtaining input data including a first set of data and a second set of data corresponding to a specific major histocompatibility complex (MHC) type - the first set of data is a plurality of peptides , wherein one peptide includes a sequence of a plurality of amino acids and the second set of data includes positional amino acid identifiers for the specific MHC type; and obtaining a first prediction result for a possibility of binding between the plurality of peptides and the positional amino acid identifiers from the input data by using the first artificial intelligence-based model.

The method and device according to one embodiment of the present disclosure, which has been devised in response to the above-described background art, can more efficiently and/or more accurately predict the binding relationship or binding possibility between MHC and peptides. there is.

1 schematically illustrates a block configuration diagram of a computing device according to an embodiment of the present disclosure.
2 illustrates an exemplary structure of an artificial intelligence-based model according to an embodiment of the present disclosure.
Figure 3 exemplarily illustrates a method for generating a prediction result for the binding possibility between MHC and a peptide according to an embodiment of the present disclosure.
4A-4C illustratively show channels of input data for a first model according to an embodiment of the present disclosure.
5 illustratively illustrates features added to input data for a first model according to one embodiment of the present disclosure.
6 shows an exemplary process for predicting the binding potential between MHC and peptides in a first model, according to an embodiment of the present disclosure.
7 shows an exemplary process for predicting the binding potential between MHC and peptides in a first model according to another embodiment of the present disclosure.
8 illustrates an example for primary extraction, secondary extraction, and tertiary extraction of a first model according to an embodiment of the present disclosure.
9 illustratively illustrates an operating manner of the second model according to an embodiment of the present disclosure.
10 exemplarily illustrates an operating manner of the second model according to an embodiment of the present disclosure.
11 is a schematic diagram of a computing environment according to one embodiment of the present disclosure.

Various embodiments are described with reference to the drawings. In this specification, various descriptions are presented to provide an understanding of the present disclosure. Prior to describing specific details for the implementation of the present disclosure, it should be noted that configurations not directly related to the technical gist of the present disclosure have been omitted within the scope of not distracting from the technical gist of the present invention. In addition, the terms or words used in this specification and claims have meanings consistent with the technical idea of the present invention based on the principle that the inventor can define the concept of appropriate terms in order to best describe his/her invention. concept should be interpreted.

As used herein, the terms "component", "module", "system", "unit", and the like refer to a computer-related entity, hardware, firmware, software, a combination of software and hardware, or an execution of software, and are interchangeable. can possibly be used. For example, a component may be, but is not limited to, a procedure, processor, object, thread of execution, program, and/or computer running on a processor. For example, both an application running on a computing device and a computing device may be components. One or more components may reside within a processor and/or thread of execution. A component can be localized within a single computer. A component may be distributed between two or more computers. Also, these components can execute from various computer readable media having various data structures stored thereon. Components may be connected, for example, via signals with one or more packets of data (e.g., data and/or signals from one component interacting with another component in a local system, distributed system) to other systems and over a network such as the Internet. data being transmitted) may communicate via local and/or remote processes.

In addition, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or”. That is, unless otherwise specified or clear from the context, “X employs A or B” is intended to mean one of the natural inclusive substitutions. That is, X uses A; X uses B; Or, if X uses both A and B, "X uses either A or B" may apply to either of these cases. Also, the term "and/or" as used herein should be understood to refer to and include all possible combinations of one or more of the listed related items.

Also, the terms "comprises" and/or "comprising" should be understood to mean that the features and/or components are present. However, it should be understood that the terms "comprises" and/or "comprising" do not exclude the presence or addition of one or more other features, elements, and/or groups thereof. Also, unless otherwise specified or where the context clearly indicates that a singular form is indicated, the singular in this specification and claims should generally be construed to mean "one or more".

In addition, the term "at least one of A or B" or "at least one of A and B" means "includes only A", "includes only B", "includes A and B in combination" should be interpreted as meaning

Those skilled in the art will further understand that the various illustrative logical components, blocks, modules, circuits, means, logics, and algorithms described in connection with the embodiments disclosed herein may be electronic hardware, computer software, or combinations of both. It should be recognized that it can be implemented as To clearly illustrate the interchangeability of hardware and software, various illustrative components, blocks, means, logics, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented in hardware or as software depends on the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application. However, such implementation decisions should not be interpreted as causing a departure from the scope of this disclosure.

The description of the presented embodiments is provided to enable any person skilled in the art to use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art of this disclosure. The general principles defined herein may be applied to other embodiments without departing from the scope of this disclosure. Thus, the present invention is not limited to the embodiments presented herein. The present invention is to be accorded the widest scope consistent with the principles and novel features set forth herein.

In the present disclosure, terms expressed as Nth, such as first, second, or third, are used to distinguish at least one entity. For example, entities represented as first and second may be the same as or different from each other.

As used in this disclosure, the term "immunopeptidome" may be used to include either the MHC peptidome or the HLA peptidome. For example, an HLA peptidome may refer to peptides bound to HLA, and an MHC peptidome may mean peptides bound to MHC.

In the present disclosure, for convenience of description, human leukocyte antigen (HLA) is exemplarily used as an example for MHC. Therefore, the description of HLA used below is an example for expressing a description of MHC, and the scope of rights of the present disclosure will be determined based on the content described in the claims, and the scope of rights will be determined through the example of HLA. It should not be construed as limited to HLA.

HLA and MHC in the present disclosure may be used interchangeably.

As used in the present disclosure, the term “Human Leukocyte Antigen (HLA)” is a glycoprotein molecule produced by a human MHC gene, and is a gene showing the largest polymorphism among genes possessed by humans. HLA typing, which determines the HLA type, can be used very actively in various fields such as organ transplantation, immunotherapy, disease-related research, paternity tests such as paternity, forensic use, and genetic research.

An HLA type in the present disclosure may include, for example, an HLA-A type, an HLA-B type, and/or an HLA-C type.

1 schematically illustrates a block configuration diagram of a computing device 100 according to one embodiment of the present disclosure.

Computing device 100 according to an embodiment of the present disclosure may include a processor 110 and a memory 130 .

The configuration of the computing device 100 shown in FIG. 1 is only a simplified example. In one embodiment of the present disclosure, the computing device 100 may include other components for performing a computing environment of the computing device 100, and only some of the disclosed components may constitute the computing device 100.

The computing device 100 in the present disclosure may mean any type of node constituting a system for implementing embodiments of the present disclosure. The computing device 100 may refer to any type of user terminal or any type of server. Components of the aforementioned computing device 100 are exemplary and some may be excluded or additional components may be included. For example, when the aforementioned computing device 100 includes a user terminal, an output unit (not shown) and an input unit (not shown) may be included within the scope of the computing device 100 .

The computing device 100 in the present disclosure may perform technical features according to embodiments of the present disclosure to be described later. For example, the computing device 100 obtains input data corresponding to a specific MHC type, and inputs the input data into an artificial intelligence-based first model to determine the possibility of binding between a plurality of peptides and positional amino acid identifiers. Can produce predictive results.

In one embodiment of the present disclosure, the computing device 100 may obtain a result of performing nucleotide sequence analysis (eg, NGS) from a server or an external entity. In another embodiment, the computing device 100 may perform nucleotide sequence analysis on genetic data (eg, DNA or RNA) obtained from a biological sample derived from a subject. The term used in the present disclosure, sequencing may be performed by any type of technique capable of analyzing the sequence of bases, for example, whole genome sequencing, whole exome It may include, but is not limited to, whole exome sequencing or whole transcriptome sequencing. As used in the present disclosure, a subject may refer to a subject or individual for obtaining a biological sample containing a major histocompatibility complex (MHC)-peptide complex. As used in the present disclosure, the term, sample, can be used without limitation as long as it is obtained from an individual or subject whose MHC type is to be determined, and for example, cells or tissues obtained by biopsy, blood, whole blood, serum, plasma, saliva, It may be cerebrospinal fluid, various secretions, urine and/or feces, and the like. Preferably, the sample may be selected from the group consisting of blood, plasma, serum, saliva, nasal fluid, sputum, ascites, vaginal secretion and/or urine, and more preferably blood, plasma or serum. The sample may be pretreated prior to use for detection or diagnosis. For example, pretreatment methods may include homogenization, filtration, distillation, extraction, concentration, inactivation of interfering components, and/or addition of reagents, and the like. In the present disclosure, a biological sample may be tissue, cell, whole blood, and/or blood, but is not limited thereto.

In one embodiment, the processor 110 may include at least one core, and may include a central processing unit (CPU) or a general purpose graphics processing unit (GPGPU) of the computing device 100 . , a processor for data analysis and/or processing, such as a tensor processing unit (TPU).

The processor 110 may read the computer program stored in the memory 130 and generate a prediction result for the binding possibility between MHC and the peptide according to an embodiment of the present disclosure. In addition, the processor 110 reads the computer program stored in the memory 130 and, according to an embodiment of the present disclosure, uses a second model using a plurality of features as inputs to determine the possibility of binding between the peptide and MHC. Can produce predictive results.

According to an embodiment of the present disclosure, the processor 110 may perform an operation for learning a neural network. The processor 110 is used for neural network learning, such as processing input data for learning in deep learning (DL), extracting features from input data, calculating errors, and updating neural network weights using backpropagation. calculations can be performed. At least one of the CPU, GPGPU, and TPU of the processor 110 may process learning of the network function. For example, the CPU and GPGPU can process learning of network functions and data classification using network functions. In addition, in one embodiment of the present disclosure, learning of a network function and data classification using a network function may be processed by using processors of a plurality of computing devices together. In addition, a computer program executed in a computing device according to an embodiment of the present disclosure may be a CPU, GPGPU or TPU executable program.

Additionally, the processor 110 may typically handle overall operations of the computing device 100 . For example, the processor 110 processes data, information, signals, etc. input or output through components included in the computing device 100 or drives an application program stored in a storage unit, thereby providing appropriate information or information to the user. A function can be provided or processed.

According to one embodiment of the present disclosure, memory 130 may store any type of information generated or determined by processor 110 and any type of information received by computing device 100 . According to one embodiment of the present disclosure, the memory 130 may be a storage medium that stores computer software that causes the processor 110 to perform operations according to embodiments of the present disclosure. Accordingly, the memory 130 may refer to computer readable media for storing software codes necessary for performing embodiments of the present disclosure, data subject to execution of the codes, and results of execution of the codes.

According to an embodiment of the present disclosure, the memory 130 may refer to any type of storage medium. For example, the memory 130 may be a flash memory type, a hard disk type ), multimedia card micro type, card type memory (eg SD or XD memory, etc.), RAM (Random Access Memory, RAM), SRAM (Static Random Access Memory), ROM (Read-Only memory, ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, and optical disk. The computing device 100 may operate in relation to a web storage that performs a storage function of the memory 130 on the Internet. The description of the above memory is only an example, and the memory 130 used in the present disclosure is not limited to the above example.

The communication unit (not shown) in the present disclosure may be configured regardless of its communication mode, such as wired and wireless, and may be configured in various communication networks such as a personal area network (PAN) and a wide area network (WAN). can be configured. In addition, the network unit 150 may operate based on the known World Wide Web (WWW), and a wireless transmission technology used for short-range communication such as Infrared Data Association (IrDA) or Bluetooth. can also be used.

The computing device 100 in the present disclosure may include any type of user terminal and/or any type of server. Accordingly, embodiments of the present disclosure may be performed by a server and/or a user terminal.

A user terminal may include any type of terminal capable of interacting with a server or other computing device. User terminals include, for example, mobile phones, smart phones, laptop computers, personal digital assistants (PDAs), slate PCs, tablet PCs, and ultrabooks. can include

A server may include any type of computing system or computing device, such as, for example, microprocessors, mainframe computers, digital processors, portable devices and device controllers, and the like.

In a further embodiment, the aforementioned server may refer to an entity that stores and manages immunopeptidome information, peptide sequence information, base sequence information, or gene information. The server may include a storage unit (not shown) for storing immunopeptidome information, peptide sequence information, positional amino acid identifier information, nucleotide sequence information, gene information, or reliability information of a database, and the storage unit may include a server It may be contained within or exist under the management of the server. As another example, the storage unit may exist outside the server and may be implemented in a form capable of communicating with the server. In this case, the storage unit may be managed and controlled by another external server different from the server.

2 illustrates an exemplary structure of an artificial intelligence-based model according to an embodiment of the present disclosure.

Throughout this specification, artificial intelligence models, artificial intelligence-based models, computational models, neural networks, network functions, and neural networks may be used interchangeably. For example, the first model and the second model in the present disclosure may be artificial intelligence-based models.

A neural network may consist of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network includes one or more nodes. Nodes (or neurons) constituting neural networks may be interconnected by one or more links.

In a neural network, one or more nodes connected through a link may form a relative relationship of an input node and an output node. The concept of an input node and an output node is relative, and any node in an output node relationship with one node may have an input node relationship with another node, and vice versa. As described above, an input node to output node relationship may be created around a link. More than one output node can be connected to one input node through a link, and vice versa.

In a relationship between an input node and an output node connected through one link, the value of data of the output node may be determined based on data input to the input node. Here, a link interconnecting an input node and an output node may have a weight. The weight may be variable, and may be changed by a user or an algorithm in order to perform a function desired by the neural network. For example, when one or more input nodes are interconnected by respective links to one output node, the output node is set to a link corresponding to values input to input nodes connected to the output node and respective input nodes. An output node value may be determined based on the weight.

As described above, in the neural network, one or more nodes are interconnected through one or more links to form an input node and output node relationship in the neural network. Characteristics of the neural network may be determined according to the number of nodes and links in the neural network, an association between the nodes and links, and a weight value assigned to each link. For example, when there are two neural networks having the same number of nodes and links and different weight values of the links, the two neural networks may be recognized as different from each other.

A neural network may be composed of a set of one or more nodes. A subset of nodes constituting a neural network may constitute a layer. Some of the nodes constituting the neural network may form one layer based on distances from the first input node. For example, a set of nodes having a distance of n from the first input node may constitute n layers. The distance from the first input node may be defined by the minimum number of links that must be passed through to reach the corresponding node from the first input node. However, the definition of such a layer is arbitrary for explanation, and the order of a layer in a neural network may be defined in a method different from the above. For example, a layer of nodes may be defined by a distance from a final output node.

In one embodiment of the present disclosure, a set of neurons or nodes may be defined as a layer.

An initial input node may refer to one or more nodes to which data is directly input without going through a link in relation to other nodes among nodes in the neural network. Alternatively, in a relationship between nodes based on a link in a neural network, it may mean nodes that do not have other input nodes connected by a link. Similarly, the final output node may refer to one or more nodes that do not have an output node in relation to other nodes among nodes in the neural network. Also, the hidden node may refer to nodes constituting the neural network other than the first input node and the last output node.

In the neural network according to an embodiment of the present disclosure, the number of nodes in the input layer may be the same as the number of nodes in the output layer, and the number of nodes decreases and then increases again as the number of nodes progresses from the input layer to the hidden layer. can In addition, the neural network according to another embodiment of the present disclosure may be a neural network in which the number of nodes of the input layer may be less than the number of nodes of the output layer and the number of nodes decreases as the number of nodes increases from the input layer to the hidden layer. there is. In addition, the neural network according to another embodiment of the present disclosure is a neural network in which the number of nodes in the input layer may be greater than the number of nodes in the output layer, and the number of nodes increases as the number of nodes increases from the input layer to the hidden layer. can A neural network according to another embodiment of the present disclosure may be a neural network in the form of a combination of the aforementioned neural networks.

A deep neural network (DNN) may refer to a neural network including a plurality of hidden layers in addition to an input layer and an output layer. Deep neural networks can reveal latent structures in data. That is, the structure of a photograph, text, video, audio, protein sequence structure, gene sequence structure, peptide sequence structure, music potential structure (e.g., what objects are in the picture, what the content and emotion of the text are, audio What is the content and emotion of , etc.), and/or the binding affinity between the peptide and MHC can be grasped. Deep neural networks include convolutional neural networks (CNNs), recurrent neural networks (RNNs), auto encoders, generative adversarial networks (GANs), and restricted boltzmann machines (RBMs). machine), a deep belief network (DBN), a Q network, a U network, a Siamese network, a Generative Adversarial Network (GAN), and the like. The description of the deep neural network described above is only an example, and the present disclosure is not limited thereto.

The artificial intelligence-based model (eg, the first model and/or the clustering model, etc.) of the present disclosure may be represented by a network structure of any of the foregoing structures including an input layer, a hidden layer, and an output layer.

A neural network that can be used in the artificial intelligence-based model of the present disclosure is at least one of supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning. can be learned in this way. Learning of the neural network may be a process of applying knowledge for the neural network to perform a specific operation to the neural network.

A neural network can be trained in a way that minimizes output errors. In the learning of the neural network, the learning data is repeatedly input into the neural network, the output of the neural network for the training data and the error of the target are calculated, and the error of the neural network is transferred from the output layer of the neural network to the input layer in the direction of reducing the error. It is a process of updating the weight of each node of the neural network by backpropagating in the same direction. In the case of supervised learning, each learning data is labeled with the correct answer (ie, labeled learning data), and in the case of unsupervised learning, the correct answer may not be labeled in each learning data. That is, for example, learning data in the case of supervised learning related to data classification may be data in which each learning data is labeled with a category. Labeled training data is input to a neural network, and an error may be calculated by comparing an output (category) of the neural network and a label of the training data. As another example, in the case of unsupervised learning for data classification, an error may be calculated by comparing input learning data with a neural network output. The calculated error is back-propagated in a reverse direction (ie, from the output layer to the input layer) in the neural network, and the connection weight of each node of each layer of the neural network may be updated according to the back-propagation. The amount of change in the connection weight of each updated node may be determined according to a learning rate. The neural network's computation of input data and backpropagation of errors can constitute a learning cycle (epoch). The learning rate may be applied differently according to the number of iterations of the learning cycle of the neural network. For example, a high learning rate may be used in the early stage of neural network training to increase efficiency by allowing the neural network to quickly obtain a certain level of performance, and a low learning rate may be used in the late stage to increase accuracy.

In neural network learning, generally, training data can be a subset of real data (ie, data to be processed using the trained neural network), and therefore, errors for training data are reduced, but errors for real data are reduced. There may be incremental learning cycles. Overfitting is a phenomenon in which errors for actual data increase due to excessive learning on training data. For example, a phenomenon in which a neural network that has learned a cat by showing a yellow cat does not recognize that it is a cat when it sees a cat other than yellow may be a type of overfitting. Overfitting can act as a cause of increasing the error of machine learning algorithms. Various optimization methods can be used to prevent such overfitting. In order to prevent overfitting, methods such as increasing the training data, regularization, inactivating some nodes in the network during learning, and using a batch normalization layer are methods. can be applied

A computer readable medium storing a data structure according to an embodiment of the present disclosure is disclosed. The above-described data structure may be stored in a storage unit in the present disclosure, executed by a processor, and transmitted and received by a communication unit.

Data structure can refer to the organization, management, and storage of data that enables efficient access and modification of data. Data structure may refer to the organization of data to solve a specific problem (eg, data retrieval, data storage, data modification in the shortest time). A data structure may be defined as a physical or logical relationship between data elements designed to support a specific data processing function. A logical relationship between data elements may include a connection relationship between user-defined data elements. A physical relationship between data elements may include an actual relationship between data elements physically stored in a computer-readable storage medium (eg, a persistent storage device). The data structure may specifically include a set of data, a relationship between data, and a function or command applicable to the data. Through an effectively designed data structure, a computing device can perform calculations while using minimal resources of the computing device. Specifically, the computing device can increase the efficiency of operation, reading, insertion, deletion, comparison, exchange, and search through an effectively designed data structure.

The data structure can be divided into a linear data structure and a non-linear data structure according to the shape of the data structure. A linear data structure may be a structure in which only one data is connected after one data. Linear data structures may include lists, stacks, queues, and decks. A list may refer to a series of data sets in which order exists internally. The list may include a linked list. A linked list may be a data structure in which data are connected in such a way that each data is connected in a single line with a pointer. In a linked list, a pointer can contain information about connection to the next or previous data. A linked list can be expressed as a singly linked list, a doubly linked list, or a circular linked list depending on the form. A stack can be a data enumeration structure that allows limited access to data. A stack can be a linear data structure in which data can be processed (eg, inserted or deleted) at only one end of the data structure. The data stored in the stack may be a LIFO-Last in First Out (Last in First Out) data structure. A queue is a data listing structure that allows limited access to data, and unlike a stack, it can be a data structure (FIFO-First in First Out) in which data stored later comes out later. A deck can be a data structure that can handle data from either end of the data structure.

The nonlinear data structure may be a structure in which a plurality of data are connected after one data. The non-linear data structure may include a graph data structure. A graph data structure can be defined as a vertex and an edge, and an edge can include a line connecting two different vertices. A graph data structure may include a tree data structure. The tree data structure may be a data structure in which one path connects two different vertices among a plurality of vertices included in the tree. That is, it may be a data structure that does not form a loop in a graph data structure.

Throughout this specification, artificial intelligence-based models, computational models, neural networks, network functions, and neural networks may be used interchangeably. Hereinafter, a neural network is unified and described. The data structure may include a neural network. And the data structure including the neural network may be stored in a computer readable medium. The data structure including the neural network may also include preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation function associated with each node or layer of the neural network, and neural network It may include a loss function for learning of . A data structure including a neural network may include any of the components described above. That is, the data structure including the neural network includes preprocessed data for processing by the neural network, data input to the neural network, weights of the neural network, hyperparameters of the neural network, data obtained from the neural network, activation function associated with each node or layer of the neural network, and neural network. It may be configured to include all or any combination thereof, such as a loss function for learning of . In addition to the foregoing configurations, the data structure comprising the neural network may include any other information that determines the characteristics of the neural network. In addition, the data structure may include all types of data used or generated in the computational process of the neural network, but is not limited to the above. A computer readable medium may include a computer readable recording medium and/or a computer readable transmission medium. A neural network may consist of a set of interconnected computational units, which may generally be referred to as nodes. These nodes may also be referred to as neurons. A neural network includes one or more nodes.

The data structure may include data input to the neural network. A data structure including data input to the neural network may be stored in a computer readable medium. Data input to the neural network may include training data input during a neural network learning process and/or input data input to a neural network that has been trained. Data input to the neural network may include pre-processed data and/or data subject to pre-processing. Pre-processing may include a data processing process for inputting data to a neural network. Accordingly, the data structure may include data subject to pre-processing and data generated by pre-processing. The foregoing data structure is only an example, and the present disclosure is not limited thereto.

The data structure may include the weights of the neural network. (In this specification, weights and parameters may be used in the same meaning.) Also, a data structure including weights of a neural network may be stored in a computer readable medium. A neural network may include a plurality of weights. The weight may be variable, and may be changed by a user or an algorithm in order to perform a function desired by the neural network. For example, when one or more input nodes are interconnected by respective links to one output node, the output node is set to a link corresponding to values input to input nodes connected to the output node and respective input nodes. A data value output from an output node may be determined based on the weight. The foregoing data structure is only an example, and the present disclosure is not limited thereto.

As a non-limiting example, the weights may include weights that are varied during neural network training and/or weights for which neural network training has been completed. The variable weight in the neural network learning process may include a weight at the time the learning cycle starts and/or a variable weight during the learning cycle. The weights for which neural network learning has been completed may include weights for which learning cycles have been completed. Accordingly, the data structure including the weights of the neural network may include a data structure including weights that are variable during the neural network learning process and/or weights for which neural network learning is completed. Therefore, it is assumed that the above-described weights and/or combinations of weights are included in the data structure including the weights of the neural network. The foregoing data structure is only an example, and the present disclosure is not limited thereto.

The data structure including the weights of the neural network may be stored in a computer readable storage medium (eg, a memory or a hard disk) after going through a serialization process. Serialization can be the process of converting a data structure into a form that can be stored on the same or another computing device and later reconstructed and used. A computing device may serialize data structures to transmit and receive data over a network. The data structure including the weights of the serialized neural network may be reconstructed on the same computing device or another computing device through deserialization. The data structure including the weights of the neural network is not limited to serialization. Furthermore, the data structure including the weights of the neural network is a data structure for increasing the efficiency of operation while minimizing the resource of the computing device (for example, B-Tree, R-Tree, Trie, m-way search tree in a nonlinear data structure) , AVL tree, Red-Black Tree). The foregoing is only an example, and the present disclosure is not limited thereto.

The data structure may include hyper-parameters of the neural network. Also, the data structure including the hyperparameters of the neural network may be stored in a computer readable medium. A hyperparameter may be a variable variable by a user. Hyperparameters include, for example, learning rate, cost function, number of learning cycle iterations, weight initialization (eg, setting the range of weight values to be targeted for weight initialization), hidden unit number (eg, the number of hidden layers and the number of nodes in the hidden layer). The foregoing data structure is only an example, and the present disclosure is not limited thereto.

A transformer may be considered as a network function for at least one of the first model, the clustering model, or the second model according to an embodiment of the present disclosure. For example, the second sub-model of the first model may operate based on a transformer. This second sub-model may be operated using, for example, a recurrent neural network to which an attention algorithm is applied or a transformer to which an attention algorithm is applied.

In one embodiment, a transformer may consist of an encoder that encodes the embedded data and a decoder that decodes the encoded data. The transformer may have a structure that receives a series of data and outputs a series of data of different types through encoding and decoding steps. In one embodiment, the series of data can be processed into a form operable by a transformer. A process of processing a series of data into a form in which a transformer can operate may include an embedding process. Expressions such as data token, embedding vector, and embedding token may refer to embedded data in a form that can be processed by a transformer.

In order for the transformer to encode and decode a series of data, encoders and decoders within the transformer may be processed using an attention algorithm. The Attention Algorithm is to obtain the similarity of one or more keys for a given query, reflect the given similarity to each key and corresponding value, and then reflect the similarity to the value ( Values) may mean an algorithm that calculates an attention value by weighting.

Depending on how to set the query, key, and value, various types of attention algorithms can be classified. For example, when attention is obtained by setting the same query, key, and value, this may mean a self-attention algorithm. In order to process a series of input data in parallel, when the dimension of the embedding vector is reduced and individual attention heads are obtained for each divided embedding vector to obtain attention, this means a multi-head attention algorithm. can do.

In one embodiment, a transformer may consist of modules that perform a plurality of multi-head self-attention algorithms or multi-head encoder-decoder algorithms. In one embodiment, the transformer may also include additional elements other than the attention algorithm, such as an embedding layer, a normalization layer, and a softmax layer. Methods for constructing transformers using the attention algorithm may include methods disclosed in Vaswani et al., Attention Is All You Need, 2017 NIPS, which is incorporated herein by reference.

Transformers can be applied to various data domains such as embedded natural language, embedded sequence information, segmented image data, and audio waveforms to convert a series of input data into a series of output data. In order to convert data having various data domains into a series of data that can be input to the transformer, the transformer can embed the data. Transformers can process additional data representing relative positional or phase relationships between a set of input data. Alternatively, a series of input data may be embedded by additionally reflecting vectors representing a relative positional relationship or phase relationship between input data to the series of input data. In one example, the relative positional relationship between a series of input data may include, but is not limited to, word order in a natural language sentence, relative positional relationship of each segmented image, temporal sequence of segmented audio waveforms, and the like. . A process of adding information representing a relative positional relationship or phase relationship between a series of input data may be referred to as positional encoding.

One example of how to embed image data and transform it into a transformer is disclosed in Dosovitskiy, et al., AN IMAGE IS WORTH 16X16 WORDS: TRANSFORMERS FOR IMAGE RECOGNITION AT SCALE, which document is incorporated herein by reference.

Figure 3 shows an exemplary method for determining the binding potential between MHC and peptides according to an embodiment of the present disclosure.

In one embodiment, the steps shown in FIG. 3 may be performed by computing device 100 . In a further embodiment, the steps in FIG. 2 may be implemented by a plurality of entities, such that some of the steps shown in FIG. 3 are performed in a user terminal and others are performed in a server.

As shown in FIG. 3 , the computing device 100 may obtain input data including a first set of data and a second set of data corresponding to a specific major histocompatibility complex (MHC) type (310). .

In one embodiment, in terms of inference, the computing device 100 performs a combination between the amino acid identifier of the MHC and the amino acid identifier of the peptide in the input data including the first set of data and the second set of data. It is possible to output a prediction result that predicts the possibility. In addition, the computing device 100 may generate a prediction result of a binding possibility between a peptide and amino acids of MHC in response to an input of a peptide sequence.

As another example, in terms of inference, the computing device 100 outputs an MHC type matching the input data in response to input data including the first set of data and the second set of data. can In this embodiment, input data corresponding to a specific MHC type herein may refer to input data composed of peptide and amino acid identifiers that do not know what the MHC type is. For example, when a specific MHC type is the first MHC type, the computing device 100 may output a prediction result of the first MHC type in response to input data using a pre-learned classification model. In this embodiment, input data corresponding to a specific type of MHC may mean data that is a target of predicting what type of MHC is. Additionally, the computing device 100 may generate a prediction result including an MHC type matched to the peptide in response to an input of the peptide sequence.

In one embodiment, in terms of learning, the computing device 100 may obtain input data (eg, a training data set) labeled with an answer MHC type that is the first MHC type. In the process of learning the classification model, the computing device 100 provides positive feedback when the output of the classification model is the first MHC type and provides negative feedback when the output of the classification model is not the first MHC type. Classification models can be trained in this way. Hereinafter, descriptions of embodiments of the present disclosure are described in a form encompassing both an inference process using a pretrained model and a learning process for learning a model.

In one embodiment, the computing device 100 may obtain input data for each type of MHC. For example, input data corresponding to the first type of MHC may be different from input data corresponding to the second type of MHC. That is, the first set of data and the second set of data corresponding to a second MHC type different from the first MHC type include the first set of data and the second set of data corresponding to the first MHC type. may differ from

In one embodiment, input data may be obtained from public databases and/or experimental results. As another example, the input data may be obtained as a result of processing data obtained from a public database and/or experimental results.

In one embodiment, the input data may be obtained in units of a specific MHC type and the first set of data in this input data may include sequences of peptides. For example, the sequence of a peptide may consist of a set of a plurality of amino acids. A second set of data in the input data may include allelic sequences capable of constituting MHC (eg, HLA allele sequences). For example, the second set of data may include positional amino acid identifiers of a specific type of MHC. In this example, the second set of data may include a plurality of pieces of data paired with positional information and amino acid identifier information.

In one embodiment, the input data may have a data format in the form of at least one matrix. Input data may have a plurality of channels, and may have a matrix corresponding to each of the plurality of channels. For example, when input data has three channels, three matrices may be included in the input data. For example, input data may include a first channel representing a first matrix and a second channel representing a second matrix. As such, input data including a plurality of channels may be processed to correspond to a plurality of channels in an artificial intelligence-based network model for processing images such as CNN. An example related to adding a channel for such input data is described in FIG. 4 .

In one embodiment, the input data may express, for example, a possibility of combining, a binding affinity, or a possibility of mutation for each of the first set of data and the second set of data. In this example, the input data may be generated for each MHC type and may be in the form of a matrix including the first set of data and the second set of data in rows and columns or columns and rows. The value of each element in the input data is a quantitative value (e.g., between 0 and 1) indicating the possibility of combination or mutation between the first set of data and the second set of data based on a predetermined combination threshold criterion. value) can be expressed as As another example, the value of each element in the input data may have a value of 0 or 1.

In one embodiment, the binding potential, binding affinity and/or variability potential considered within the input data may be obtained from a public database or from experimental results. In a further embodiment, binding potential, binding affinity and/or mutability may be obtained based on processed data from public databases or based on experimental results. The input data may be generated based on a determination of whether amino acids at positions of a specific HLA type and a specific peptide sequence have high binding potential or high mutation potential.

In one embodiment, the input data is a value expressing the possibility of binding or substitution between the plurality of peptides included in the first set of data and the positional amino acid identifiers included in the second set of data as elements It may include a first matrix including. An example of such a first matrix is a blosum matrix.

In a further embodiment, the input data may include a second matrix representing binding energies between amino acids of MHC and amino acids of peptides. An example of such a second matrix is an energy matrix.

In a further embodiment of the present disclosure, at least some of the input data may include a zero value. The importance may be determined based on the position of the peptide sequence or the position of MHC, and a value of 0 may be assigned to an element corresponding to a position having a relatively low importance.

In one embodiment, the input data is a first characteristic representing the polarity between amino acids of MHC and amino acids of the peptide, a second characteristic representing the size of amino acids, a third characteristic representing hydrophobicity or hydrophilicity of amino acids, the charge of amino acids It may further include at least one of a fourth feature indicating whether the amino acid is present or a fifth feature indicating whether the amino acid is aromatic or aliphatic. These characteristics may be converted to 0 or 1 during encoding (eg, one-hot encoding) of input data, for example. These features may be added (eg, concatenated) to at least one of the above-described input data corresponding to the first matrix or input data corresponding to the second matrix. As an example, a value of 1 for polarity with respect to the first feature and 0 for non-polarity may be assigned to the additional feature. As another example, a value of 1 otherwise 0 may be assigned as an additional feature if the amino acid corresponds to Charged, a value of 1 otherwise 0 may be assigned as an additional feature if the amino acid corresponds to Hydrophobic, and the size of the amino acid ( A value of 1 if Size) is greater than a predetermined size threshold, otherwise a value of 0 may be assigned to the additional feature. The foregoing descriptions are listed for illustrative purposes, and values to be assigned to additional features may be determined variably depending on implementation aspects. Additional features for this input data are illustrated in FIG. 5 .

As shown in FIG. 3, the computing device 100 obtains a first prediction result for the possibility of binding between a plurality of peptides and positional amino acid identifiers from input data using a first artificial intelligence-based model. can (320).

For example, the artificial intelligence-based first model may refer to a classification model for predicting binding affinity between a peptide and an MHC type or for outputting an MHC type for input data.

In one embodiment, the first model analyzes time-series data in response to a first sub-model for extracting visual features of a matrix included in input data and an input based on the visual features extracted from the first sub-model. It may include a second sub-model for For example, the first sub-model may use a convolutional neural network-based artificial neural network. As an example, the second sub-model may use a recurrent neural network-based artificial neural network. An example of an RNN-based artificial neural network is LSTM (Long Short Term Memory).

In one embodiment of the present disclosure, the first prediction result may include binding possibilities between amino acid identifiers generated in response to input data corresponding to the first MHC type and peptides. The first prediction result may include a value quantitatively indicating binding potential for each of the amino acid identifiers and peptides included in the input data.

In one embodiment of the present disclosure, the first prediction result may include an MHC type corresponding to the input data.

In one embodiment, the computing device 100 uses a second artificial intelligence-based model that takes a plurality of prediction results including the first prediction result as an input and generates a second prediction result for the binding possibility between the peptide and the MHC You can (330).

In one embodiment, the second model may refer to a model that combines (eg, ensemble) outputs of a plurality of models including the first model. For example, the second model may output a binding possibility between a peptide and MHC by applying a weight to outputs of a plurality of models. The data indicating the binding potential between the peptide and MHC may include data indicating what type of MHC corresponds to the input data.

In one embodiment, the second model may also be an AI-based model. In this case, the second model may be pretrained to output a second prediction result in response to a plurality of inputs.

4A-4C illustratively show channels of input data for a first model according to an embodiment of the present disclosure. As illustrated in FIGS. 4A to 4C , input data input to the classification model may have various forms.

As shown in FIG. 4A , the input data 410 is exemplified by data in the form of a matrix. A row in the input data 410 may represent amino acid identifiers of peptides, and a column in the input data 410 may represent an MHC type or amino acid identifiers of MHC. In one embodiment, computing device 100 uses a classification model (eg, a first model and/or a second model) to predict or correspond to values of elements in input data 410 . It is possible to predict what the MHC type will be.

4B shows a Blosum Matrix as an example of the input data 410 . The input data 420 may express substitution possibilities between amino acid identifiers of MHC and amino acid identifiers of peptides as element values. For example, a large value of an element means that the possibility of substitution between amino acid identifiers in a row and column corresponding to the corresponding element is high, and a small value of an element means that amino acid identifiers in a row and column corresponding to the corresponding element have a high substitution probability. It can mean that the possibility of substitution is low. A value representing the possibility of such substitution may be interpreted as a value representing the degree of closeness between amino acids, for example.

In one embodiment, a blosum matrix may refer to a matrix representing the degree of physical and chemical similarity or substitution possibility between amino acid sequences.

4C shows an amino acid energy matrix as an example of the input data 410. The input data 430 may represent energy values binding between amino acid identifiers of MHC and amino acid identifiers of peptides as values of elements. For example, a large value of an element means that the binding energy between amino acid identifiers in a row and column corresponding to the corresponding element is high, and a small value of an element means that the binding energy between amino acid identifiers in a row and column corresponding to the corresponding element is high. It can mean that the binding energy is low. A value representing such binding energy may be interpreted as a value representing the degree of closeness between amino acids, for example.

5 illustratively illustrates features added to input data for a classification model according to one embodiment of the present disclosure.

As shown in FIG. 5, the input data may include additional features 530 added to the matrix 510 related to peptides and additional features 540 added to the matrix 520 related to MHC. can Although input data related to peptides and MHC are illustrated as matrices 510 and 520 in FIG. 5, it will be clear to those skilled in the art that they may correspond to either rows or columns.

As shown in FIG. 5 , a classification model for input data may include, for example, a CNN/RNN 550 and a Fully Connected Layer (FCL) 560. The classification model may extract features of input data through CNN/RNN 550, learn the extracted features, and generate classification results for features through FCL 560. For example, the CNN/RNN 550 extracts a visual feature of a matrix included in the input data and generates time-series data corresponding to the matrix in response to an input based on the extracted visual feature. can be analyzed. The FCL 560 flattens the output of the CNN/RNN 550, converts it into a single vector, and generates probability information for each of predetermined classes (or labels). The FCL 560 may generate a classification result based on the analysis result of the CNN/RNN 550.

In one embodiment of the present disclosure, features 530 and 540 added to the input data may include physical and/or chemical properties of amino acids. For example, the features 530 and 540 added to the input data include information about polarity of amino acids, charged information about amino acids, information indicating hydrophobicity or hydrophilicity of amino acids, and aromaticity of amino acids. Alternatively, information indicating aliphaticity and/or information indicating the size of amino acids may be included.

In one embodiment of the present disclosure, the added features may be encoded as binary values of 1 and 0, for example. For example, a value of 1 may be assigned when polarity of an amino acid exists, and a value of 0 may be assigned when polarity does not exist. In addition, a value of 1 may be assigned if the amino acid is hydrophilic, and a value of 0 may be assigned if the amino acid is hydrophobic. In addition, a value of 1 may be assigned if the amino acid is aromatic and a value of 0 may be assigned if the amino acid is aliphatic.

As such, the technique according to an embodiment of the present disclosure utilizes information between amino acids of MHC and peptides as well as additional information inherent in each amino acid, so that an artificial intelligence-based model can more efficiently and accurately determine the binding possibility between peptides and MHC. can be allowed to learn.

6 shows an exemplary process for predicting the binding potential between MHC and peptides in a first model, according to an embodiment of the present disclosure. Descriptions of the size of the input data 610 and the filters in FIG. 6 and the dimensions of the data input to each layer are described for illustrative purposes, and it is also obvious to those skilled in the art that they can be implemented in a variable size depending on the implementation mode. something to do.

As shown in FIG. 6 , the first model may include a plurality of first sub-models 620a, 620b, and 620c and a plurality of second sub-models 630a, 630b, and 630c. For example, the first sub-models 620a, 620b, and 620c are illustrated as CNN-based models, and the second sub-models 630a, 630b, and 630c are RNN-based models (eg, LSTM). can be exemplified. Additionally, the second sub-models 630a, 630b, and 630c may use an artificial neural network based on a recurrent neural network to which an attention algorithm is applied. For example, the second sub-models 630a, 630b, and 630c may use a transformer.

Regarding the exemplary structure of the first model shown in FIG. 6, the first model includes an input layer processing input data 610, and an output of the input layer received and included in the input data. N feature extraction layers for outputting visual features of the matrix, where N is a natural number, a recurrent neural network layer that performs time-series learning by receiving the output of the feature extraction layer, and receiving the output of the recurrent neural network layer. An output layer outputting a classification result corresponding to the input data may be included. Here, the number of classification results corresponding to the number of feature extraction layers may be output.

The structure of the first model according to an embodiment of the present disclosure illustrated in FIG. 6 may be verified and learned in each section of a plurality of layers in order to accurately determine the optimized number of CNN layers and Maxpooling layers. Since it is difficult to know in advance which layer of the CNN optimizes the features optimized for the input data 610, when the structure of the first model illustrated in FIG. 6 is used, the features optimized for the input data can be extracted efficiently.

In one embodiment, the first output of the 1-1 sub-model 620a among the plurality of first sub-models 620a, 620b, and 620c is the first output of the plurality of second sub-models 630a, 630b, and 630c. It is input to the corresponding 2-1st sub-model 630a and the 2nd output of the 1-1st sub-model 620a is the 1st- out of the plurality of 1st sub-models 620a, 620b and 620c. It may be input to the 1st-2nd sub-model 620b connected to the 1st sub-model 620a.

Here, the first output includes the output passing through the flatten layer of the 1-1 sub-model 620a, and the second output includes the pooling of the 1-1 sub-model 620a. ) may contain the output passed through the layer.

In an embodiment, an output of the 1-1 sub-model 620a passing through a flatten layer may be pre-processed to be used as an input of the 2-1 sub-model 630a. That is, the computing device 100 may pre-process the output of a specific layer of the CNN to be compatible with the input of the RNN, for example. As an example of preprocessing, the Reshape technique can be used. For example, the computing device 100 outputs the flatten layer of the 1-1 sub-model 620a (for example, the output format of the flatten layer (Data size, time step * feature) format). ) can be changed to the format of (Data size, time step, feature), which is the same input format as RNN. According to this change, the format of (Data size, time step * feature), which is the output format of the flatten layer that passed through the flatten layer of the 1-1 sub-model 620a, is the format of the input shape of the RNN. It can be changed to (Data size, time step * feature, 1) so that it changes to .

As illustrated in FIG. 6 , the input data 610 is input to the 1-1 sub-model 620a and accordingly, features of the input data 610 may be primarily extracted. The first output of the 1-1 sub-model 620a may mean an output that has passed through the platen layer, and this first output is a 2-1 sub-model corresponding to the 1-1 sub-model 620a. It can be used as an input of 630a. The second output of the 1-1 sub-model 620a may mean an output that has passed through a pooling layer (eg, Maxpooling layer), and this second output is used as an input of the 1-2 sub-model 620b. Accordingly, features of the input data 610 may be extracted secondarily. As the second output of the 1-2 sub-model 620b is transferred to the 1-3 sub-model 620c in the same manner, the characteristics of the input data 610 are generated by the 1-3 sub-model 620c. This can be extracted tertiary.

In one embodiment, the first outputs of the 1-1 sub-model 620a, 1-2 sub-model 620b and 1-3 sub-model 620c are respectively the 2-1 sub-model 630a, It is transferred to the 2-2nd sub-model 630b and the 2-3rd sub-model 630c, and the 2-1st sub-model 630a, the 2-2nd sub-model 630b and the 2-3rd sub-model ( 630c) Each may analyze time-series data for the input data 610 based on visual features in the first output. The outputs of the 2-1st sub-model 630a, 2-2nd sub-model 630b and 2-3rd sub-model 630c are transmitted to FCLs 640a, 640b and 640c, respectively, and analyzed as time-series data. Vectorized or flattened information for can be generated. An output of each of the FCLs 640a, 640b, and 640c may be transferred to classification layers 650a, 650b, and 650c to generate a plurality of classification results for the input data 610. A predicted value 660 may be generated based on a plurality of classification results. In one embodiment, the prediction value 660 may correspond to the first prediction result in FIG. 3 of the present disclosure.

In one embodiment of the present disclosure, the predicted value 660 may be used as an input to a classification model (eg, a second model) 670 along with the predicted values of other models. For example, the predicted value 660 may be ensembled with other predicted values from other models to generate a classification result for the input data 610 in the classification model 670 corresponding to the second model. For example, a classification result in the classification model 670 in FIG. 6 may correspond to a second prediction result in FIG. 3 in the present disclosure.

7 shows an exemplary process for predicting the binding potential between MHC and peptides in a first model according to another embodiment of the present disclosure.

The descriptions of the size of the input data 710 and the filters in FIG. 7 and the dimensions of the data input to each layer are described for illustrative purposes, and it is also obvious to those skilled in the art that the size can be implemented in a variable size depending on the implementation mode. something to do.

The structure of the first model according to an embodiment of the present disclosure illustrated in FIG. 7 may be verified and learned in each section of a plurality of layers in order to accurately determine the optimized number of CNN layers and Maxpooling layers. Since it is difficult to know in advance which layer of the CNN optimizes the features optimized for the input data 710, when the structure of the first model illustrated in FIG. 7 is used, the features optimized for the input data can be extracted efficiently.

In one embodiment, the first output of the 1-1st sub-model 720 among the plurality of 1st sub-models 720, 730 and 740 is input to the 2nd sub-model 750 and the 1-1 The second output of the sub-model 720 is input to the 1-2 sub-model 730 connected to the 1-1 sub-model 720 among the plurality of first sub-models 720, 730 and 740. can

Here, the first output includes the output passing through the flatten layer of the 1-1 sub-model 720, and the second output includes pooling of the 1-1 sub-model 720. ) may contain the output passed through the layer.

As illustrated in FIG. 7 , the input data 710 is input to the 1-1 sub-model 720 and accordingly, features of the input data 710 may be primarily extracted. The first output of the 1-1 sub-model 720 may refer to an output that has passed through the platen layer, and this first output is the 1-2 sub-model 730 and the 1-3 sub-model 740 respectively. It can be combined (eg concatenated) with the first outputs of and used as an input of the second sub-model 750. The second output of the 1-1 sub-model 720 may refer to an output that has passed through a pooling layer (eg, Maxpooling layer), and this second output may be used as an input of the 1-2 sub-model 730. there is. As the second output of the 1-2 sub-model 730 is transmitted to the 1-3 sub-model 740 in the same manner, the characteristics of the input data 710 are obtained by the 1-3 sub-model 740. This can be extracted tertiary.

In one embodiment, the first outputs of the 1-1 sub-model 720, the 1-2 sub-model 730 and the 1-3 sub-model 740 are respectively combined to form the second sub-model 750. and the second sub-model 750 may analyze the time-sequential data of the input data 710 based on the visual characteristics of the first output. The output of the second sub-model 750 may be delivered to the FCL 760 to generate vectorized or flattened information on the analyzed time-series data. An output of the FCL 760 may be passed to the classification layer 770 to generate a classification result (eg, a first prediction result) for the input data 710 .

In one embodiment of the present disclosure, the classification result may be used as an input of another classification model (eg, the second model) together with the predicted values of the other models.

In one embodiment, FIG. 8 schematically illustrates examples of primary extraction 810 , secondary extraction 820 , and tertiary extraction 830 of features for input data.

As shown in FIG. 8, the first model includes an input layer for processing input data, N numbers for receiving the output of the input layer and outputting visual characteristics of matrices included in the input data (N is a natural number). ) of the feature extraction layer, a recurrent neural network layer (e.g., LSTM) that receives the output of the feature extraction layer and performs time-series learning, and receives the output of the recurrent neural network layer and outputs a classification result corresponding to the input data It may contain an output layer. For example, classification results corresponding to the number of feature extraction layers may be output.

According to this network structure, since the features extracted from all layers can be verified, more specific features can be extracted for the input data, and the learning performance in the second sub-model (e.g., LSTM) using these features this can be improved.

In one embodiment, a learning process for generating a classification result for the possibility of binding between peptides and MHC in the second sub-model may be performed. In an additional embodiment of the present invention, the second sub-model may use an artificial neural network based on a recurrent neural network to which an attention algorithm is applied.

In one embodiment, an output of each of the plurality of second sub-models may be input to a fully connected layer, and a classification result for the input data may be output. For example, the classification result may include a first class indicating a case in which combination is possible and a second class indicating a case in which combination is impossible. As another example, the classification result may include a quantitative numerical value between 0 and 1.

Returning to FIG. 3, the computing device 100 uses a second artificial intelligence-based model that takes a plurality of prediction results including the first prediction result as an input, and obtains a second prediction result for the binding possibility between the peptide and the MHC. It can be created (330).

A description of the operation of the second model 930 may be illustrated in FIG. 9 .

In one embodiment, the second model 930 may be trained by taking a plurality of values indicated by reference number 920 as inputs. For example, the second model 930 may be pretrained based on the correct answer data set 940 generated based on at least one of a mass spectrometry (MS) test and a T cell receptor (TCR) test. . The second model 930 is a classification model that outputs a result of binding or nonbinding, and outputs 920 of a plurality of other models that take the peptide sequence 910 for one MHC as an input. can be used as input.

In one embodiment, mass spectrometry testing and/or TCR testing may refer to the use of predictive results using ongoing experiments or computational techniques to validate an unsubstantiated theory or technique.

In one embodiment, the input data for the second model 930 or the second model 930 includes the first prediction result that is the output of the first model, but uses other additional factors 920 as features. features can be created. In this case, the first prediction result corresponding to the output of the first model may indicate binding affinity between the peptide and MHC. The second model 930 may refer to a classification model for generating a prediction result (eg, a second prediction result) including a binding possibility between a peptide and MHC in consideration of other factors including binding affinity. Alternatively, the second model 930 determines whether the corresponding peptide sequence 910 binds to the one MHC type with respect to the peptide sequence 910 corresponding to one MHC sequence in consideration of the plurality of factors 920. results can be output.

In one embodiment, the plurality of prediction results 920 used as inputs of the second model 930 are: a binding energy value representing the strength of the binding between the peptide and MHC, MHC is bound to the antigen, and the cell A cleavage value representing the probability that a specific peptide is cleaved by the proteasome and bound to the MHC in the process represented on the image, a TAP value representing the probability that the peptide cleaved by the proteasome passes TAP (Transporter Associated with Antigen Processing), A stability value indicating how long the binding between MHC and the peptide lasts, a value indicating the degree of difference between the correct sequence of the peptide determined to bind to a specific MHC and the sequence of the specific peptide, and immunopeptidome information stored in the database may include a reliability value of

In one embodiment, at least some of the plurality of prediction results 920 may include outputs generated by separate models in response to input data that takes as input a peptide sequence or a matrix including a peptide and an MHC. . As for the methodology for computing each of the plurality of prediction results described above, any known methodology may be considered. As examples of such a methodology, a logistic regression model, a decision tree model, and/or a support vector machine model may be considered.

In one embodiment, the binding affinity, which is an output from the first model, may mean a value representing the possibility of mutation between amino acids and/or the probability of binding between amino acids.

In one embodiment, the binding energy value may mean quantitatively representing the binding energy of one amino acid to another amino acid in terms of energy in MHC.

In one embodiment, in the process of expressing MHC on a cell after being bound to an antigen, a portion of the MHC sequence may be cut into fragments from a structure called a proteasome. In this situation, peptides with sizes such as 7 to 30 mer can also be produced. A specific part of the proteasome is recognized as a singularity and is cleaved. A calculation of the probability that a certain peptide sequence is actually cleaved from the proteasome in the patient and bound to MHC may correspond to the clevage score.

In one embodiment, only a specific peptide among the truncated peptides enters the Endoplasmic Reticulum (ER) through a transporter called TAP, and the calculated probability of the peptide passing through TAP may correspond to the TAP value.

In one embodiment, the persistence value may be exemplified by a value indicating how long the binding between MHC and the peptide lasts.

In one embodiment, the confidence value may correspond to the degree of contamination of immune peptidome information in the database. Regarding this reliability value, the description in Korean Patent Application No. 10-2022-0038736 referred to in this patent application may be referred to.

In one embodiment, a value representing the degree of difference between the correct sequence of a peptide determined to bind to a specific MHC and the sequence of a specific peptide may be collectively referred to as a mutation score (Alteration score or Mutation score). For example, it is assumed that the sequence of a peptide that binds to a specific MHC type stored through a mass spectrometry test or a T cell receptor (TCR) test is stored as MAAPASTAT. Under this assumption, when the mutation score of the peptide of the sequence MAAPAVTAT is calculated, it is confirmed that S and V at the 6th position are different from each other. For this difference, for example, a value corresponding to S->V in the Blosum matrix representing the possibility of substitution between amino acids may be referred to. Based on the value, the mutation value for the peptide of the sequence MAAPAVTAT can be calculated. Additionally, these disparity values may be assigned different weights depending on whether the disparity position is an anchor position. For example, different weights may be assigned depending on whether the disparity position is an anchor position or a non-anchor position.

That is, the correct sequence of the peptide is determined based on at least one of a mass spectrometry test and a T cell receptor (TCR) test, and the values representing the different degrees depend on whether the different positions are anchor positions. can be determined based on

In one embodiment, the second model 930 uses a plurality of features corresponding to the plurality of prediction results 920 described above to predict the binding possibility between MHC and the peptide. It may mean a classification model that generates a prediction result. there is. In the process of learning the second model, a learning data set 940 in which a binding result between a peptide and MHC according to at least one of a mass spectrometry test and a T cell receptor (TCR) test as correct answer data can be used. .

In one embodiment, as shown in FIG. 10 , the second model 1060 is, in response to the same input data or input features 1010, a logistic regression model 1020, a decision tree ) model 1030 and the support vector machine model 1040, respectively, a classification result may be generated by applying voting 1050 to the outputs generated.

In a further embodiment of the present disclosure, an algorithm for voting weights for outputs generated from each of a plurality of models may also be implemented by an artificial intelligence-based model. The corresponding model may determine a weight to be assigned to each of output values from a plurality of models based on machine learning.

11 is a schematic diagram of a computing environment according to one embodiment of the present disclosure.

A component, module or unit in this disclosure includes routines, procedures, programs, components, data structures, etc. that perform particular tasks or implement particular abstract data types. In addition, those skilled in the art will understand that the methods presented in this disclosure can be used in single-processor or multiprocessor computing devices, minicomputers, mainframe computers as well as personal computers, handheld computing devices, microprocessor-based or programmable consumer electronics, and the like ( It will be fully appreciated that each of these may be implemented with other computer system configurations, including those that may be operative in connection with one or more associated devices.

Embodiments described in this disclosure may also be practiced in distributed computing environments where certain tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.

A computing device typically includes a variety of computer readable media. Computer readable media can be any medium that can be accessed by a computer, including volatile and nonvolatile media, transitory and non-transitory media, removable and non-transitory media. Includes removable media. By way of example, and not limitation, computer readable media may include computer readable storage media and computer readable transmission media.

Computer readable storage media are volatile and nonvolatile media, transitory and non-transitory, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. includes media Computer readable storage media may include RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital video disk (DVD) or other optical disk storage device, magnetic cassette, magnetic tape, magnetic disk storage device or other magnetic storage device. device, or any other medium that can be accessed by a computer and used to store desired information.

A computer readable transmission medium typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism. Including all information delivery media. The term modulated data signal means a signal that has one or more of its characteristics set or changed so as to encode information within the signal. By way of example, and not limitation, computer readable transmission media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above are also intended to be included within the scope of computer readable transmission media.

An exemplary environment 2000 implementing various aspects of the present invention is shown comprising a computer 2002, which includes a processing unit 2004, a system memory 2006 and a system bus 2008. do. The computer 200 in this specification may be used interchangeably with a computing device. System bus 2008 couples system components, including but not limited to system memory 2006, to processing unit 2004. Processing unit 2004 may be any of a variety of commercially available processors. Dual processors and other multiprocessor architectures may also be used as the processing unit 2004.

System bus 2008 may be any of several types of bus structures that may additionally be interconnected to a memory bus, a peripheral bus, and a local bus using any of a variety of commercial bus architectures. System memory 2006 includes read only memory (ROM) 2010 and random access memory (RAM) 2012 . The basic input/output system (BIOS) is stored in non-volatile memory 2010, such as ROM, EPROM, EEPROM, etc. BIOS is a basic set of information that helps transfer information between components within the computer 2002, such as during startup. contains routines. RAM 2012 may also include high-speed RAM, such as static RAM, for caching data.

The computer 2002 may also read from an internal hard disk drive (HDD) 2014 (eg EIDE, SATA), a magnetic floppy disk drive (FDD) 2016 (eg a removable diskette 2018), or for writing to them), SSDs and optical disk drives 2020 (for example, for reading CD-ROM disks 2022 or reading from or writing to other high capacity optical media such as DVDs) include The hard disk drive 2014, magnetic disk drive 2016, and optical disk drive 2020 are connected to the system bus 2008 by the hard disk drive interface 2024, magnetic disk drive interface 2026, and optical drive interface 2028, respectively. ) can be connected to The interface 2024 for external drive implementation includes, for example, at least one or both of Universal Serial Bus (USB) and IEEE 1394 interface technologies.

These drives and their associated computer readable media provide non-volatile storage of data, data structures, computer executable instructions, and the like. In the case of computer 2002, drives and media correspond to storing any data in a suitable digital format. Although the description of computer-readable storage media above refers to HDDs, removable magnetic disks, and removable optical media such as CDs or DVDs, those skilled in the art can use zip drives, magnetic cassettes, flash memory cards, cartridges, It will be appreciated that other types of computer-readable storage media, such as those of other types, may also be used in the exemplary operating environment and that any such media may contain computer-executable instructions for performing the methods of the present invention. .

A number of program modules may be stored on the drive and RAM 2012, including an operating system 2030, one or more application programs 2032, other program modules 2034, and program data 2036. All or portions of the operating system, applications, modules and/or data may also be cached in RAM 2012. It will be appreciated that the present invention may be implemented in a variety of commercially available operating systems or combinations of operating systems.

A user may enter commands and information into the computer 2002 through one or more wired/wireless input devices, such as a keyboard 2038 and a pointing device such as a mouse 2040. Other input devices (not shown) may include a microphone, IR remote control, joystick, game pad, stylus pen, touch screen, and the like. Although these and other input devices are often connected to the processing unit 2004 through an input device interface 2042 that is connected to the system bus 2008, a parallel port, IEEE 1394 serial port, game port, USB port, IR interface, may be connected by other interfaces such as the like.

A monitor 2044 or other type of display device is also connected to the system bus 2008 through an interface such as a video adapter 2046. In addition to the monitor 2044, computers typically include other peripheral output devices (not shown) such as speakers, printers, and the like.

Computer 2002 may operate in a networked environment using logical connections to one or more remote computers, such as remote computer(s) 2048 via wired and/or wireless communications. Remote computer(s) 2048 may be workstations, server computers, routers, personal computers, handheld computers, microprocessor-based entertainment devices, peer devices, or other common network nodes, and generally relate to computer 2002. Although many or all of the described components are included, for brevity, only memory storage device 2050 is shown. The logical connections shown include wired/wireless connections to a local area network (LAN) 2052 and/or a larger network, such as a wide area network (WAN) 2054 . Such LAN and WAN networking environments are common in offices and corporations and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to worldwide computer networks, such as the Internet.

When used in a LAN networking environment, computer 2002 is connected to local network 2052 through wired and/or wireless communication network interfaces or adapters 2056. Adapter 2056 may facilitate wired or wireless communications to LAN 2052, which also includes a wireless access point installed therein to communicate with wireless adapter 2056. When used in a WAN networking environment, computer 2002 may include a modem 2058, be connected to a communications server on WAN 2054, or other means of establishing communications over WAN 2054, such as over the Internet. have Modem 2058, which can be internal or external and wired or wireless, is connected to system bus 2008 through serial port interface 2042. In a networked environment, program modules described for computer 2002, or portions thereof, may be stored in remote memory/storage device 2050. It will be appreciated that the network connections shown are exemplary and other means of establishing a communication link between computers may be used.

Computer 1602 is any wireless device or entity that is deployed and operating in wireless communication, such as printers, scanners, desktop and/or portable computers, portable data assistants (PDAs), communication satellites, and associated with wireless detectable tags. It operates to communicate with arbitrary equipment or places and telephones. This includes at least Wi-Fi and Bluetooth wireless technologies. Thus, the communication may be a predefined structure as in conventional networks or simply an ad hoc communication between at least two devices.

It is to be understood that the specific order or hierarchy of steps in the processes presented is an example of example approaches. Based upon design priorities, it is to be understood that the specific order or hierarchy of steps in the processes may be rearranged within the scope of the present disclosure. The method claims of this disclosure present elements of the various steps in a sample order, but are not meant to be limited to the specific order or hierarchy presented.

Claims (21)

A method performed by a computing device, comprising:
Obtaining input data including a first set of data and a second set of data corresponding to a specific major histocompatibility complex (MHC) type, wherein the first set of data includes a plurality of peptides , one peptide includes a sequence of a plurality of amino acids and the second set of data includes positional amino acid identifiers for the specific MHC type; and
Using the artificial intelligence-based first model, the possibility of binding between the plurality of peptides and the amino acid identifiers for each position from the input data, the result of what the MHC type corresponds to the input data, or the peptide and MHC obtaining a first prediction result including at least one of binding affinities between types;
Including,
The first model includes a plurality of first sub-models and a plurality of second sub-models, and
The first output of the 1-1 submodel among the plurality of first sub-models is input to the corresponding 2-1 sub-model among the plurality of second sub-models, and A second output is input to a 1-2 sub-model connected to the 1-1 sub-model among the plurality of first sub-models;
method. According to claim 1,
The specific MHC type includes a first MHC type and a second MHC type, and the first set of data and the second set of data corresponding to the second MHC type correspond to the first MHC type. different from the set of data and the second set of data,
method. According to claim 1,
The input data is
A first matrix including, as elements, values representing the possibility of binding between a plurality of peptides included in the first set of data and amino acid identifiers for each position included in the second set of data,
method delete According to claim 1,
The first model is:
a first sub-model for extracting a visual feature of a matrix included in the input data; and
a second sub-model for analyzing time-series data corresponding to the matrix in response to an input based on the visual feature extracted from the first sub-model;
including,
method delete According to claim 1,
The output of each of the plurality of second sub-models is input to a fully connected layer, and a classification result for the input data is output.
method. According to claim 1,
The first output includes an output that has passed through a flatten layer of the 1-1 sub-model, and
The second output includes an output that has passed through a pooling layer of the 1-1 submodel.
method. According to claim 5,
The first sub-model uses an artificial neural network based on a convolutional neural network, and
The second sub-model uses an artificial neural network based on a recurrent neural network,
method. According to claim 9,
The second sub-model uses an artificial neural network based on a recurrent neural network to which an attention algorithm is applied.
method. According to claim 1,
The first model includes a plurality of first sub-models,
An output passed through a pooling layer in a 1-1 sub-model among the plurality of first sub-models is input to a 1-2 sub-model connected to the 1-1 sub-model among the plurality of first sub-models, , and
The outputs passing through the platen layer in each of the plurality of first sub-models are concatenated and input to the second sub-model.
method. According to claim 1,
The first model,
an input layer processing the input data;
N feature extraction layers for receiving the output of the input layer and outputting visual features of matrices included in the input data, where N is a natural number;
a recurrent neural network layer that receives the output of the feature extraction layer and performs time-sequential learning; and
an output layer receiving the output of the recurrent neural network layer and outputting a classification result corresponding to the input data;
Including,
Classification results corresponding to the number of feature extraction layers are output,
method. According to claim 1,
The input data is
A second matrix representing the binding energy between amino acids of MHC and amino acids of peptides,
method. According to claim 1,
The input data is
The first characteristic indicates the polarity between the amino acid of MHC and the amino acid of the peptide, the second characteristic indicates the size of the amino acid, the third characteristic indicates whether the amino acid is hydrophobic or hydrophilic, the fourth characteristic indicates the presence or absence of an amino acid charge, or an amino acid Further comprising at least one of the fifth characteristics indicating whether is aromatic or aliphatic,
method. According to claim 1,
Generating a second prediction result for the binding possibility between the peptide and MHC using a second artificial intelligence-based model that takes a plurality of prediction results including the first prediction result as input;
Including more,
method. According to claim 15,
The plurality of prediction results are:
the first predicted result representing binding affinity between the peptide and MHC;
A binding energy value representing the strength of the binding between the peptide and MHC;
a cleavage value representing the probability that a specific peptide is cleaved by proteasome and bound to the MHC in the process of MHC being bound to an antigen and being expressed on a cell;
a TAP value representing the probability that the peptide cut by the proteasome passes TAP (Transporter Associated with Antigen Processing);
a stability value indicating how long the binding between MHC and the peptide lasts;
A value representing the degree of difference between the correct sequence of the peptide determined to bind to a specific MHC and the sequence of the specific peptide; and
Reliability value of immunopeptidome information stored in the database;
including at least two of
method. 17. The method of claim 16,
The value representing the degree of difference is determined based on whether the different position is an anchor position,
method. According to claim 15,
The second model,
A classification model that is learned based on a learning data set in which the binding result between a peptide and MHC according to at least one of a mass spectrometry test and a T cell receptor (TCR) test is used as correct answer data,
method. According to claim 15,
The second model,
In response to the same input data, a classification result is generated by applying voting to outputs generated from each of a logistic regression model, a decision tree model, and a support vector machine model. doing,
method. A computer program stored on a computer-readable storage medium, which, when executed by a computing device, causes the computing device to perform the following operations:
Obtaining input data including a first set of data and a second set of data corresponding to a specific major histocompatibility complex (MHC) type, wherein the first set of data includes a plurality of peptides, and one peptide contains a sequence of a plurality of amino acids and the second set of data includes positional amino acid identifiers for the particular MHC type; and
Using the artificial intelligence-based first model, the possibility of binding between the plurality of peptides and the amino acid identifiers for each position from the input data, the result of what the MHC type corresponds to the input data, or the peptide and MHC obtaining a first prediction result including at least one of binding affinities between types;
Including,
The first model includes a plurality of first sub-models and a plurality of second sub-models, and
The first output of the 1-1 submodel among the plurality of first sub-models is input to the corresponding 2-1 sub-model among the plurality of second sub-models, and A second output is input to a 1-2 sub-model connected to the 1-1 sub-model among the plurality of first sub-models;
A computer program stored on a computer readable storage medium. As a computing device,
at least one processor; and
Memory;
Including,
The at least one processor is:
Obtaining input data including a first set of data and a second set of data corresponding to a specific major histocompatibility complex (MHC) type, wherein the first set of data includes a plurality of peptides, and one peptide contains a sequence of a plurality of amino acids and the second set of data includes positional amino acid identifiers for the particular MHC type; and
Using the artificial intelligence-based first model, the possibility of binding between the plurality of peptides and the amino acid identifiers for each position from the input data, the result of what the MHC type corresponds to the input data, or the peptide and MHC obtaining a first prediction result including at least one of binding affinities between types;
to perform,
The first model includes a plurality of first sub-models and a plurality of second sub-models, and
The first output of the 1-1 submodel among the plurality of first sub-models is input to the corresponding 2-1 sub-model among the plurality of second sub-models, and A second output is input to a 1-2 sub-model connected to the 1-1 sub-model among the plurality of first sub-models;
computing device.

Images