Latent variable-based imputation method for missing RNA data

Abstract

The present invention relates to a system and method for predicting a phenotype from a data set including genetic information of an organism. Specifically, the present invention relates to a system and method with improved performance, capable of efficiently predicting a phenotype by restoring missing RNA data in a data set including missing RNA values.

Description

A Latent Variable Imputation Method for Missing RNA Data

[0001] The present invention relates to a system and method for predicting a phenotype from a dataset containing genetic information of an organism. Specifically, the present invention relates to an improved system and method capable of efficiently predicting a phenotype by restoring missing RNA data from a dataset containing missing RNA values.

[0002] As artificial intelligence technology advances, techniques for predicting the phenotypes of living organisms using vast amounts of genetic information are improving; however, there remains an unmet need for methods to restore missing values ​​within datasets containing genetic information.

[0003] In the case of SNP information, all instances within a dataset often have measured values, whereas in the case of RNA information, it is often difficult to conduct experiments on every instance, resulting in a high number of missing values.

[0004] Conventionally, attempts have been made to solve this missing data problem using traditional statistical approaches or machine learning-based restoration methods utilizing unimodal data; however, there is a problem in that it is difficult to solve the issue solely with machine learning-based restoration techniques because the dimensionality of missing RNA features is often too high to be directly restored.

[0005] Accordingly, the inventors of the present invention intend to solve the problem of missing RNA in various downstream tasks by compressing RNA features into latent expressions based on a deep learning model, and then generating latent expressions of missing RNA features from SNP features based on a generative model.

[0006] The inventors of the present invention aim to provide a system and method with improved performance capable of efficiently predicting phenotypes by learning phenotype prediction from a set containing RNA data through the present invention, and to provide a system and method capable of efficiently learning a dataset containing RNA data in which data for one or more instances is missing.

[0007] To solve the above problem, the present invention provides a phenotype prediction system.

[0008] The above-mentioned phenotype prediction system includes a memory that stores one or more instructions; and

[0009] At least one that executes the one or more instructions stored in the memory.

[0010] Includes a processor,

[0011] The operation performed by the above one or more instructions is

[0012] The method includes a step of learning how an artificial intelligence model can restore missing RNA data from a dataset containing SNP information and RNA information, and

[0013] The step here of learning how to restore the aforementioned missing RNA data is

[0014] A step of extracting SNP features and RNA features from a training dataset containing SNP information and RNA information,

[0015] A step of learning the process of an artificial intelligence model to generate SNP latent variables from the above SNP features, generate RNA latent variables from RNA features, and reconstruct SNP features from an integrated latent variable formed by integrating the SNP latent variables and the RNA latent variables, and

[0016] It includes a step of learning the process of reconstructing RNA features from SNP features extracted from the above-mentioned training data by an artificial intelligence model.

[0017] The present invention provides a phenotype prediction system characterized by an artificial intelligence model reconstructing SNP features from the same input value and generating RNA latent variables in the step of reconstructing RNA features from SNP features extracted from the training data.

[0018] The present invention provides a phenotype prediction system in which, in the step of reconstructing RNA features from SNP features extracted from the training data, the artificial intelligence model includes a generative model based on a variational autoencoder.

[0019] The present invention provides a phenotype prediction system comprising the step of an artificial intelligence model that has completed training restoring missing RNA data and predicting a phenotype using the restored RNA data.

[0020] The present invention provides a phenotype prediction system comprising a step of learning the process of reconstructing SNP features from the integrated latent variable, wherein the artificial intelligence model learns using a mean squared error loss function.

[0021] The present invention provides a phenotype prediction system comprising the step of reconstructing RNA features from SNP features extracted from the training data, wherein an artificial intelligence model generates RNA latent variables from SNP features.

[0022] The present invention provides a phenotype prediction system that further includes a step of an artificial intelligence model learning a process of reconstructing RNA features from RNA features extracted from the training data.

[0023] The present invention provides a phenotype prediction system comprising an artificial intelligence model having a multilayer perceptron structure in the step of reconstructing RNA features from RNA features extracted from the training data.

[0024] The present invention also provides a phenotype prediction method performed by at least one processor.

[0025] The above phenotype prediction method includes the step of an artificial intelligence model learning how to restore missing RNA data from a dataset containing SNP information and RNA information, and

[0026] The step here of learning how to restore the aforementioned missing RNA data is

[0027] A step of extracting SNP features and RNA features from a training dataset containing SNP information and RNA information,

[0028] A step of learning the process of an artificial intelligence model to generate SNP latent variables from the above SNP features, generate RNA latent variables from RNA features, and reconstruct SNP features from an integrated latent variable formed by integrating the SNP latent variables and the RNA latent variables, and

[0029] It includes a step of learning the process of reconstructing RNA features from SNP features extracted from the above-mentioned training data by an artificial intelligence model.

[0030] The present invention provides a phenotype prediction method characterized by, in the step of reconstructing RNA features from SNP features extracted from the training data, an artificial intelligence model reconstructing SNP features from the same input value and generating RNA latent variables.

[0031] The present invention provides a phenotype prediction method in which, in the step of reconstructing RNA features from SNP features extracted from the training data, the artificial intelligence model includes a generative model based on a variational autoencoder.

[0032] The present invention provides a phenotype prediction method comprising the step of an artificial intelligence model that has completed training restoring missing RNA data and predicting a phenotype using the restored RNA data.

[0033] The present invention provides a phenotype prediction method comprising the step of learning the process of reconstructing SNP features from the integrated latent variable, wherein the artificial intelligence model learns using a mean squared error loss function.

[0034] The present invention provides a phenotype prediction method comprising the step of reconstructing RNA features from SNP features extracted from the training data, wherein an artificial intelligence model generates RNA latent variables from SNP features.

[0035] The present invention provides a phenotype prediction method that further includes the step of an artificial intelligence model learning a process of reconstructing RNA features from RNA features extracted from the training data.

[0036] The present invention provides a phenotype prediction method comprising an artificial intelligence model having a multilayer perceptron structure in the step of reconstructing RNA features from RNA features extracted from the training data.

[0037] In addition, the present invention provides a program stored on a computer-readable recording medium to execute the above method on a computer.

[0038] By using the phenotype prediction system and method of the present invention, missing RNA values ​​in a dataset can be restored more accurately compared to conventional technology.

[0039] By using the phenotype prediction system and method of the present invention, missing RNA information within a dataset can be restored with high accuracy and fully utilized for phenotype prediction learning, thereby enabling phenotype prediction with superior performance compared to conventional technology.

[0040] The phenotype prediction system and method of the present invention can be utilized to obtain information for diagnosing the disease state of an individual, and can also be utilized to obtain information for evaluating the disease risk of an individual, and can also be utilized to discover biomarkers for disease diagnosis or to discover targets for new drug development.

[0041] FIG. 1 is a block diagram of a device capable of implementing a phenotype prediction system and method according to one embodiment of the present invention.

[0042] FIG. 2 is a flowchart illustrating a system and method for an artificial intelligence model to learn how to restore missing RNA data in a dataset containing SNP information and RNA information according to an embodiment of the present invention.

[0043] FIG. 3 is a flowchart illustrating a system and method for learning the process of reconstructing RNA features from RNA features extracted from artificial intelligence learning data according to an embodiment of the present invention.

[0044] FIG. 4 is a flowchart illustrating a system and method for learning the process of reconstructing SNP features from SNP features extracted from training data according to an embodiment of the present invention.

[0045] FIG. 5 is a flowchart illustrating a system and method for an artificial intelligence model to learn the process of reconstructing RNA features from SNP features extracted from training data according to an embodiment of the present invention.

[0046] FIG. 6 illustrates a phenotype prediction system and method according to one embodiment of the present invention.

[0047] Figure 7 is an exemplary on-premise full stack structure.

[0048] To clarify the technical concept of the present invention, embodiments of the present invention will be described in detail with reference to the attached drawings. In describing the present invention, detailed descriptions of related known functions or components will be omitted if it is determined that such detailed descriptions may unnecessarily obscure the essence of the present invention. Components having substantially the same function or configuration among the drawings have been assigned the same reference numbers and symbols as much as possible, even if they are shown in different drawings. For convenience of explanation, devices and methods will be described together where necessary. Each operation of the present invention does not necessarily have to be performed in the order described, but may be performed in parallel, selectively, or individually.

[0049] The terms used in the embodiments of the present invention have been selected to be as widely used as possible, taking into account the functions of the invention; however, these terms may vary depending on the intent of those skilled in the art, case law, the emergence of new technologies, etc. Additionally, in specific cases, terms have been arbitrarily selected by the applicant, and in such cases, their meanings will be described in detail in the description section of the relevant embodiments. Therefore, terms used in this specification should be defined not merely by their names, but based on their meanings and the overall content of the invention.

[0050] Throughout the invention, singular expressions may include plural expressions unless the context clearly indicates otherwise. Terms such as "include" or "have" are intended to specify the existence of features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. That is, throughout the invention, when a part is described as "includes" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0051] Expressions such as "at least one" modify the entire list of components and do not modify the components of the list individually. For example, "at least one of A, B, and C" and "at least one of A, B, or C" refer to only A, only B, only C, both A and B, both B and C, both A and C, all of A, B, and C, or any combination thereof.

[0052] Additionally, terms such as "... part," "... module," etc., as described in the present invention refer to a unit that processes at least one function or operation, and this may be implemented in hardware or software, or a combination of hardware and software.

[0053] Throughout the invention, when a part is described as being "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "electrically connected" with other elements interposed between them. Furthermore, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0054] Throughout the invention, the expression “configured to” may be replaced, depending on the context, with, for example, “suitable for,” “having the capacity to,” “designed to,” “adapted to,” “made to,” or “capable of.” The term “configured to” does not necessarily mean only “specifically designed to” in hardware. Instead, in some situations, the expression “system configured to” may mean that the system is “capable of” in conjunction with other devices or components. For example, the phrase “processor configured to perform A, B, and C” may mean a dedicated processor for performing the corresponding operations (e.g., an embedded processor), or a generic-purpose processor (e.g., a CPU or an application processor) capable of performing the corresponding operations by executing one or more software programs stored in memory.

[0055] The artificial intelligence-related functions according to the present invention are operated through a processor and memory. The processor may be composed of one or more processors. In this case, the one or more processors may be general-purpose processors such as CPUs, APs, and DSPs (Digital Signal Processors), graphics-dedicated processors such as GPUs and VPUs (Vision Processing Units), or artificial intelligence-dedicated processors such as NPUs. The one or more processors control the processing of input data according to predefined operation rules or artificial intelligence models stored in memory. Alternatively, if the one or more processors are artificial intelligence-dedicated processors, the artificial intelligence-dedicated processors may be designed with a hardware structure specialized for processing a specific artificial intelligence model.

[0056] Artificial intelligence is a field of computer science and information technology that studies methods to enable computers to perform thinking, learning, and self-development tasks typically accomplished by human intelligence; it refers to the ability of computers to mimic intelligent human behavior. Furthermore, artificial intelligence does not exist in isolation but is closely related, directly or indirectly, to many other areas of computer science. Particularly in the modern era, there is active research being conducted across various fields of information technology to introduce AI elements and utilize them for problem-solving.

[0057] Machine learning is a field of artificial intelligence that enables computers to learn without explicit programming. Specifically, machine learning can be defined as a technology that studies and builds systems and algorithms capable of learning, making predictions, and improving their own performance based on empirical data. Rather than executing strictly defined static program commands, machine learning algorithms adopt an approach of constructing specific models to derive predictions or decisions based on input data. The term 'machine learning' may be used interchangeably with 'machine learning'.

[0058] Many machine learning algorithms have been developed to address how to classify data in machine learning. Representative examples of these algorithms include Decision Trees, Bayesian Networks, Support Vector Machines (SVM), and Artificial Neural Networks (ANN).

[0059] A decision tree is an analytical method that performs classification and prediction by plotting decision rules in a tree structure. A Bayesian network is a model that represents the probabilistic relationships (conditional independence) between multiple variables in a graph structure. Bayesian networks are suitable for data mining through unsupervised learning.

[0060] Support Vector Machines are supervised learning models for pattern recognition and data analysis, primarily used for classification and regression analysis. Artificial neural networks model the operating principles of biological neurons and the relationships between them; they are information processing systems in which multiple neurons, referred to as nodes or processing elements, are connected in a layered structure.

[0061] Artificial neural networks are models used in machine learning, serving as statistical learning algorithms in machine learning and cognitive science that draw inspiration from biological neural networks (particularly the brain within the animal central nervous system). Specifically, an artificial neural network can refer to a model in which artificial neurons (nodes), forming a network through the connection of synapses, change the strength of these connections through learning to possess problem-solving capabilities. The term 'artificial neural network' may be used interchangeably with 'neural network'.

[0062] An artificial neural network may include multiple layers, and each layer may include multiple neurons. Additionally, an artificial neural network may include synapses connecting neurons. An artificial neural network can generally be defined by the following three factors: the connection patterns between neurons in different layers, a learning process that updates the weights of the connections, and an activation function that generates an output value from a weighted sum of inputs received from the previous layer.

[0063] Artificial neural networks may include, but are not limited to, network models such as Deep Neural Networks (DNN), Recurrent Neural Networks (RNN), Bidirectional Recurrent Deep Neural Networks (BRDNN), Multilayer Perceptrons (MLP), and Convolutional Neural Networks (CNN). In this specification, the term 'layer' may be used interchangeably with the term 'layer'.

[0064] Artificial neural networks are classified into single-layer neural networks and multi-layer neural networks depending on the number of layers. A typical single-layer neural network consists of an input layer and an output layer. Additionally, a typical multi-layer neural network consists of an input layer, one or more hidden layers, and an output layer.

[0065] The input layer is a layer that receives external data, and the number of neurons in the input layer is equal to the number of input variables. The hidden layer is located between the input layer and the output layer, receives signals from the input layer, extracts features, and transmits them to the output layer. The output layer receives signals from the hidden layer and outputs an output value based on the received signals. Input signals between neurons are multiplied by their respective connection strengths (weights) and then summed; if this sum is greater than the neuron's threshold, the neuron is activated and outputs the value obtained through the activation function.

[0066] Meanwhile, a deep neural network containing multiple hidden layers between the input layer and the output layer can be a representative artificial neural network that implements deep learning, a type of machine learning technique. Meanwhile, the term 'deep learning' may be used interchangeably with the term 'deep learning'.

[0067] The machine learning workflow consists of a series of processes involving collecting data for learning and validation, modeling, and training the model, and may include the processes of collecting training data, checking and exploring data, data preprocessing and cleaning, modeling, and training.

[0068] A phenotype refers to the actual traits that manifest according to genetic information as a result of genotype expression. This includes not only physical appearance, biochemical, and physiological traits, but also characteristics such as behavioral traits and pathological conditions. Generally, the phenotype can be used as a concept encompassing disease diagnosis, prognosis prediction, and treatment responsiveness; it is a subject that can be predicted or classified through the analysis of various omics data, such as genomics, transcriptomics, and proteomics.

[0069] A disease-related phenotype refers to the observable manifestation in an individual carrying a disease, and can include various aspects such as symptoms, changes in appearance, behavior, and biochemical characteristics. For example, the decline in memory observed in individuals with dementia is a disease-related phenotype.

[0070] Phenotype prediction refers to the process of using biological data to predict an individual's observable characteristics, disease status, response, or other biological attributes. By utilizing multi-omics data, phenotypes such as disease diagnosis, prognosis, and treatment responsiveness can be predicted.

[0071] The present invention relates to a performance-improved phenotype system and method capable of efficiently predicting phenotypes, and more specifically, to a phenotype system and method capable of predicting phenotypes with excellent performance from a dataset containing missing values ​​for RNA information.

[0072] When a dataset containing missing RNA information is input to an artificial intelligence model that has completed training according to the system and method of the present invention, the artificial intelligence model restores the missing RNA information and predicts the phenotype based on the restored data, so superior performance compared to conventional artificial intelligence models can be expected.

[0073] 1. Data Acquisition

[0074] The phenotype prediction system and method of the present invention can learn a method for predicting a phenotype using at least one dataset containing genetic information of an organism.

[0075] The genetic information included in the dataset used by the phenotype prediction system and method of the present invention includes single nucleotide polymorphism (SNP) and RNA (Ribonucleic acid) information.

[0076] SNP information indicates the location and type of base sequence substitution when compared to a reference DNA base sequence, and may be a binary value representing, for example, the presence or absence of a variant. The RNA information is information regarding messenger RNA (mRNA) expression, and may be a quantitatively expressed value representing, for example, the amount of messenger RNA expression.

[0077] The phenotype prediction system and method of the present invention can predict phenotypes with excellent performance from a dataset containing missing RNA information by learning a method to accurately restore RNA information from a dataset containing missing RNA information.

[0078] 2. Inspection and Exploration of Training Data

[0079] Once training data for learning an artificial intelligence model is collected, the collected training data can be examined and explored regarding its structure, noise data, and data cleaning methods for machine learning applications.

[0080] This stage of data inspection and exploration is referred to as the Exploratory Data Analysis (EDA) phase, which can be described as the process of observing and understanding collected data from various angles. Before training the data, independent variables, dependent variables, variable types, and data types are examined using visualizations such as graphs and statistical tests, allowing the characteristics of the data and inherent structural relationships to be identified in advance. Through this EDA, examining the distribution and values ​​of the data enables a better understanding of the phenomena represented by the data and the discovery of potential issues. Furthermore, by examining the data from various angles, diverse patterns that might have been overlooked during the problem definition stage can be discovered, allowing for the modification of existing hypotheses or the formulation of new ones. Exploratory data analysis can broadly encompass the process of searching for data outliers and analyzing the relationships between data attributes.

[0081] The process of detecting outliers involves verifying whether they exist in the data and can include sampling methods, statistical methods, and visualization methods. Sampling methods involve drawing random samples from the data to identify overall trends and anomalies in the data values. Statistical methods may utilize summary statistics, such as the mean, median, and mode to identify the center of the data, or range and variance to check the dispersion. Visualization methods utilize probability density functions, histograms, dot plots, word clouds, time series charts, and maps to determine which statistical indicators are appropriate for the individual attributes of the collected data. However, when using statistical indicators, caution should be exercised as the mean reflects all data values ​​within a set and is therefore affected by outliers, whereas the median uses only the single value in the middle, allowing for representative results even in the presence of outliers.

[0082] The process of analyzing relationships between data attributes involves identifying combinations of attributes within the data that possess meaningful correlations. Relationship analysis can be conducted differently depending on the combination of attributes between qualitative attributes (Categorical Variables; Qualitative), which cannot be expressed numerically but can be arbitrarily quantified, and quantitative attributes (Numeric Variables; Quantitative), which can be quantified. Categorical-categorical relationships can display the number of values ​​corresponding to each pair of attribute values ​​using cross-tabulation tables or mosaic plots; Numeric-categorical relationships can be visually represented through box plots or by observing statistical values ​​by category (mean, median, etc.); and Numeric-numeric relationships can analyze the association between two attributes using correlation coefficients. It can be confirmed that a correlation coefficient of -1 indicates a negative correlation where the two attributes change in opposite directions, 0 indicates no correlation, and 1 indicates a positive correlation where the two attributes always change in the same direction. The relationship between two attributes with a correlation coefficient can also exhibit various aspects, which can be visually represented using a scatter plot.

[0083] 3. Preprocessing of Training Data

[0084] Data that has completed inspection and exploration undergoes data preprocessing to transform it into a format suitable for machine learning training models. Data preprocessing involves cleaning the data and converting it into a form that the model can understand; it generally includes handling missing data, outlier removal, data scaling, categorical data encoding, feature selection and extraction, and data transformation. The detailed processes of data preprocessing may be performed in whole or in part selectively, and a separate machine learning model may be used for this purpose.

[0085] Handling Missing Data is the process of handling missing values ​​when they exist in the data; these values ​​can be displayed as NaN (Not a Number) or empty, or deleted. Filling in or deleting missing values ​​improves data completeness, and values ​​such as the mean, median, or mode may be used when filling in missing values.

[0086] Outlier removal is the process of eliminating outliers, which are values ​​that deviate from typical data patterns. Since outliers can degrade model performance, they must be removed or replaced; this involves identifying outliers and deleting the corresponding rows or columns or replacing them with other values.

[0087] Data scaling is the process of adjusting the size of data; through data scaling, the range of the data is adjusted, which can improve model performance or accelerate convergence. Data scaling allows data characteristics to be aligned within a similar range, and generally, standardization and normalization can be applied.

[0088] Categorical Data Encoding is the process of converting categorical variables, which are represented as string or integer values ​​and cannot be directly input into a model, into a numeric type that can be input. Generally, one-hot encoding or label encoding can be used to convert categorical variables into numeric types.

[0089] Feature selection and extraction is intended to improve the performance of a model by selecting the most useful features for model training or extracting new features. Through this process, the complexity of the model can be reduced and overfitting can be prevented.

[0090] Data transformation involves converting data to extract new information or enable a model to understand it better, and may include the tokenization of text data or the preprocessing of image data. Through data transformation, model performance can be improved by extracting useful features from original data or converting data into an appropriate format.

[0091] Through data preprocessing as described above, it is possible to achieve the effects of improving the performance and ensuring the stability of machine learning models.

[0092] Meanwhile, if the collected data has not been preprocessed according to the requirements, tokenization, cleaning, and normalization can be performed to suit the intended use of the data.

[0093] In order for a computer to understand and process text, it must be appropriately converted into numbers. Since the performance of natural language processing varies significantly depending on how words are represented, many techniques have been proposed to quantify words. Currently, word embedding methods, which vectorize each word through artificial neural network learning, are frequently used.

[0094] 4. Training of the Phenotype Prediction Model

[0095] The present invention relates to a phenotype system and method capable of predicting a phenotype with excellent performance by restoring missing values ​​in a dataset containing missing RNA information. An embodiment of the phenotype prediction system and method of the present invention is illustrated in FIG. 2.

[0096] The present invention includes the steps of: an artificial intelligence model learning a method for restoring missing RNA data in a dataset containing SNP information and RNA information; and the artificial intelligence model, having completed the learning, restoring the missing RNA data and predicting a phenotype using the restored RNA data.

[0097] The step of learning how to restore missing RNA data may include a step of learning the process of reconstructing RNA features from RNA features extracted from training data (201), a step of learning the process of reconstructing SNP features from SNP features extracted from training data (202), and a step of learning the process of reconstructing RNA features from SNP features extracted from training data by an artificial intelligence model (203).

[0098] The step (201) of learning the process of reconstructing RNA features from RNA features extracted from training data includes the step (301) of the first encoder generating RNA latent variables from RNA features extracted from training data, the step (302) of the first decoder reconstructing RNA features from RNA latent variables, and the step (303) of the first encoder and the first decoder learning the method of reconstructing RNA features from RNA features extracted from training data.

[0099] In the present invention, a latent variable is an internal state variable used for feature reconstruction and missing value restoration within an artificial intelligence model, representing state information that is not directly observed from observed input features.

[0100] The step (202) of learning the process of reconstructing SNP features from SNP features extracted from training data includes the step (401) of generating RNA latent variables from RNA features extracted from training data by the first encoder, the step (402) of generating SNP latent variables from SNP features by the second encoder, the step (403) of generating integrated latent variables by integrating RNA latent variables and SNP latent variables, the step (404) of reconstructing SNP features from integrated latent variables by the second decoder, and the step (405) of learning the process of reconstructing SNP features from SNP features extracted from training data by the second encoder and the second decoder, wherein the first encoder does not need to perform learning in the above step (202).

[0101] The step (202) of learning the process of reconstructing SNP features from SNP features extracted from the above learning data involves learning the second encoder and the second decoder through the step (404) of reconstructing SNP features from integrated latent variables, and by learning in this way, it is possible to learn to reconstruct SNP features from SNP information and RNA information with high accuracy.

[0102] In the step (405) where the second encoder and the second decoder learn the process of reconstructing SNP features from SNP features extracted from the training data, the learning method is not particularly limited, but can be learned using a loss function, and preferably can be learned using a mean squared error loss function.

[0103] The step (203) of learning the process of reconstructing RNA features from SNP features extracted from training data by the artificial intelligence model includes a step (502, 503) in which a second encoder and a second decoder reconstruct SNP features from SNP features extracted from training data, a step (504) in which a third encoder first learns information necessary to reconstruct SNP features from RNA latent variables by repeating the above steps, a step (505) in which a third encoder generates RNA latent variables from SNP features, a step (506) in which a first decoder reconstructs RNA features from RNA latent variables, and a step (507) in which information necessary to generate RNA latent variables from SNP features is secondarily learned by repeating the above steps (505, 506).

[0104] In the step of reconstructing RNA features from SNP features extracted from the above training data, the artificial intelligence model is characterized by reconstructing SNP features from the same input value and generating RNA latent variables.

[0105] Here, the third encoder learns the information necessary to reconstruct SNP features from RNA latent variables, thereby enabling the generation of RNA latent variables from SNP features with superior performance, and using the RNA latent variables generated by the third encoder, RNA features can be restored with high accuracy.

[0106] There are no particular restrictions on the model that can be used as the third encoder, but a generative model may be used, and preferably, a generative model based on a variational autoencoder may be used.

[0107] In predicting a phenotype using the system or method of the present invention, if RNA features exist for a specific instance, a first encoder generates RNA latent variables from RNA features and a second encoder generates SNP latent variables from SNP features, and then the said latent variables can be used as input values ​​for a phenotype prediction model; and if RNA features are lacking for a specific instance, a second encoder generates SNP latent variables from SNP features and a third encoder generates RNA latent variables from SNP features, and then the said latent variables can be used as input values ​​for a phenotype prediction model.

[0108] MoE (Mixture of Experts) Architecture

[0109] In one embodiment of the present invention, a system for predicting phenotypes may be performed by utilizing a model architecture such as MoE. Here, MoE may refer to an architecture of a machine learning model that solves complex problems by combining multiple expert models.

[0110] Such MoE may include expert models, which are multiple small networks designed to learn different parts and / or different features of a given data and perform data processing operations accordingly, and a gating network that evaluates the performance of each expert model and determines which expert model is most suitable for assigning a specific task based on the given data based on this evaluation.

[0111] Thus, according to the MoE architecture, a gating network that acquires predetermined input data determines probabilistic or deterministic task assignments for each expert model, and the selected expert models perform their respective tasks and return the results, thereby enabling data processing for a specific task.

[0112] According to one embodiment of the present invention, the MoE model used may refer to a specific MoE model implemented according to a universal method known in the art. For example, the MoE model may include a Switch Transformer, Conditional Computation in Neural Networks, Sparse Mixture of Experts, and / or a Megatron-LM.

[0113] In addition, in one embodiment of the present invention, a MoE model based on the combination of a plurality of Specialized Models (SM) and Routers (Gating Network, RT) may be included, and a MoE model based on domain-specific Specialized Models may also be included.

[0114] By utilizing such MoE, the overall efficiency and performance of the phenotype prediction system of the present invention can be enhanced by activating only specific parts and concentrating computational resources in cases such as handling complex tasks or large datasets.

[0115] 5. Computing System Device

[0116] Embodiments of the present invention may be implemented as application-specific integrated circuits (ASICs) designed to suit specific application fields and special functions of devices.

[0117] Custom integrated circuits are also referred to as custom semiconductors. Unlike standard semiconductors, which have fixed specifications and can be applied to any electronic product or application as long as certain requirements are met, custom semiconductors are used for specific products or functions and are integrated circuits manufactured by semiconductor companies to meet specific orders. In other words, custom semiconductors are designed and manufactured to perform only the functions necessary for a specific device or feature. Custom semiconductors are broadly classified according to their design method into Full Custom ICs, which design and manufacture circuits from scratch to meet user requirements, and Semi-Custom ICs, which design and manufacture circuits using parts of a standardized design.

[0118] Application-specific semiconductors are primarily used in communication systems, high-performance computing systems, consumer electronics, automobiles, industrial automation, medical devices, the military, and the aerospace industry; recently, they are being applied to AI semiconductors that execute the large-scale computations required for AI implementation with high performance and power efficiency.

[0119] Application-specific semiconductors (ASICs) are used as core components in communication systems, such as network routers, switches, and modems, performing data packet processing, protocol conversion, and signal processing to provide high throughput and low latency. In high-performance computing systems, ASICs serve as key components for high-speed and parallel processing, while in consumer electronics—including digital cameras, smartphones, tablets, and game consoles—ASICs provide high-performance and low-power solutions required to perform specific functions. In the automotive industry, ASICs are used to control various electronic systems within vehicles, and in industrial automation systems, they provide solutions for high-precision control and high-performance processing.

[0120] The application-specific integrated circuit to which the embodiment of the present invention is applied includes a memory in which an individual memory interface (I / F) is implemented, and may include a plurality of function blocks that request memory access. Each function block may be a Direct Memory Access (DMA) function block, a processor, a video processor, a cache controller, a decompression block, or a data path block. The basic configuration of the application-specific integrated circuit may include a transistor that amplifies or switches an electrical signal, a logic gate which is a circuit that performs a logical function by combining transistors, a memory cell that stores data, an analog circuit which is a circuit that processes continuous voltage or current by combining transistors, and an Intellectual Property Core (IP Core) such as a microprocessor, DSP, or graphics core that is pre-designed to perform a specific function.

[0121] The ASIC may include an individual memory I / F that interfaces with individual memory and an embedded memory I / F that interfaces with embedded memory. The individual memory I / F is connected to each function block to receive memory access signals (e.g., control signals, address signals, and data signals) and, based on these input signals, can generate signals to control the individual memory. The embedded memory I / F is connected to each function block to receive memory access signals (e.g., control signals, address signals, and data signals) and, based on these input signals, can generate modified memory access signals to control the embedded memory. The individual memory I / F and the embedded memory I / F are designed within the memory control block of the ASIC to provide a memory control structure that can be flexibly applied to both the individual memory and the embedded memory.

[0122] Additionally, an application-specific integrated circuit (ASIC) for an artificial neural network is composed of multiple neurons arranged in an array and multiple synapse circuits, each neuron being composed of a register, a microprocessor, and at least one input, and each synapse circuit being configured to include memory for storing synapse weights. Here, each neuron of the ASIC may be connected to at least one other neuron through one of the multiple synapse circuits.

[0123] Although it has been described that the present invention can generally be implemented by a computing device, a person skilled in the art will be well aware that the present invention can be implemented by combining computer-executable instructions and / or other program modules that can be executed on one or more computers and / or as a combination of hardware and software.

[0124] Those skilled in the art will understand that information and signals may be represented using any various different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced in the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0125] Those skilled in the art will understand that the various exemplary logic blocks, modules, processors, means, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented by electronic hardware, various forms of programs or design code (referred to herein as software for convenience), or a combination of all such. To clearly illustrate this interoperability between hardware and software, various exemplary components, blocks, modules, circuits, and steps have been generally described above in relation to their functions. Whether such functions are implemented as hardware or software depends on the design constraints imposed on the specific application and the overall system. Those skilled in the art may implement the functions described in various ways for each specific application, but such implementation decisions should not be interpreted as being outside the scope of the invention.

[0126] The various embodiments presented herein may be implemented as methods, devices, or manufactured articles using standard programming and / or engineering techniques. The term manufactured article includes a computer program, carrier, or medium accessible from any computer-readable storage device. For example, computer-readable storage media include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical discs (e.g., CDs, DVDs, etc.), smart cards, and flash memory devices (e.g., EEPROMs, cards, sticks, key drives, etc.). Additionally, the various storage media presented herein include one or more devices and / or other machine-readable media for storing information.

[0127] It should be understood that the specific order or hierarchy of steps in the presented processes is merely an example of exemplary approaches. It should be understood that, based on design priorities, the specific order or hierarchy of steps in the processes may be rearranged within the scope of the invention. The appended method claims provide various step elements in a sample order, but do not imply limitation to the specific order or hierarchy presented.

[0128] FIG. 1 illustrates an example of a block diagram of a computing system device (100) of the present invention.

[0129] Various operations of the system and method of the present invention may be performed by any suitable means capable of performing corresponding functions. Such means may include various hardware, software components, modules, and combinations thereof, including but not limited to circuits, processors, application-specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs).

[0130] Referring to FIG. 1, a computing system (100) implementing an artificial neural network of the present invention may include a transceiver (110), memory (120), a database (130), and a processor (140). However, not all components shown in FIG. 1 are essential components of the computing system device (100). The computing system device (100) may be implemented with more components than those shown in FIG. 1, or with fewer components than those shown in FIG. 1. Furthermore, the transceiver (110), memory (120), and processor (140) may be implemented in the form of a single chip.

[0131] In one embodiment, the transceiver (110) can communicate with a terminal or other electronic device connected to the computing system device (100) via wired or wireless connection.

[0132] Various types of data, such as programs and files, such as applications, can be installed and stored in the memory (120). The processor (140) may access and use the data stored in the memory (120) or store new data in the memory (120). Additionally, one or more instructions may be stored in the memory (120). The processor (140) may execute one or more instructions stored in the memory.

[0133] The processor (140) controls the overall operation of the computing system device (100). Here, the processor (140) may be composed of at least one of a central processing unit (CPU), a graphics processing unit (GPU), an application integrated circuit, a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field programmable gate array, controllers, microcontrollers, microprocessors, and / or electrical units for performing other functions, or a plurality of electrically connected processors.

[0134] The processor (140) can control other components included in the computing system device (100) to perform operations for operating the computing system device (100).

[0135] The database (130) may store various training data for training a learning model. Additionally, the database (130) may store protein amino acid sequence information, protein structure information, simulation result information, etc., and in various embodiments, output data produced by the learning model may be stored. Although FIG. 1 is illustrated as including a database (130) in a computing system device (100), the database (130) may be provided outside the device. In this case, the database (130) may be connected to the computing system device (100) via a wired or wireless connection.

[0136] Additionally, the learning model of the present invention may be implemented outside the computing system device (100) (e.g., cloud-based) or included inside the computing system device (100).

[0137] One embodiment of the present invention may also be implemented in the form of a recording medium comprising computer-executable instructions, such as program modules executed by a computer. A computer-readable medium may be any available medium accessible by a computer and includes both volatile and non-volatile media, and both removable and non-removable media. Additionally, a computer-readable medium may include both computer storage media and communication media. A computer storage medium includes both volatile and non-volatile, removable and non-removable media implemented by any method or technique for storing information, such as computer-readable instructions, data structures, program modules, or other data. A communication medium typically includes computer-readable instructions, data structures, or program modules and includes any information transmission medium.

[0138] Throughout the invention, the device may include, but is not limited to, a server, smartphone, tablet PC, PC, TV, smart TV, mobile phone, PDA (personal digital assistant), speaker, laptop, media player, micro server, e-book object recognition device, digital broadcasting object recognition device, kiosk, MP3 player, digital camera, robot vacuum cleaner, home appliance, other mobile or non-mobile computing device, a watch equipped with communication functions and data processing functions, glasses, hair band and ring, etc.

[0139] 6. Applications of Phenotype Prediction Systems and Methods

[0140] The phenotype prediction system and method of the present invention can be used for various purposes, such as diagnosing a disease in an individual, providing information for evaluating the risk or prognosis of a disease in an individual, searching for biomarkers for diagnosing, evaluating risk of, or predicting the prognosis of a specific disease, or searching for biomolecules associated with the disease or potential drug targets for new drug development.

[0141] The phenotypic prediction system of the present invention can identify characteristics highly associated with disease, and from this, it is possible to explore potential drug targets, disease prognostic biomarkers, differentially expressed genes in the disease, features important to the disease, and features involved in the disease's pathogenesis.

[0142] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will understand that other specific forms can be easily modified without altering the technical concept or essential features of the present invention. For example, each component described as a single unit may be implemented in a distributed manner, and components described as distributed may also be implemented in a combined form.

[0143] The scope of the present invention is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present invention.

Claims

1. In a phenotype prediction system, Memory for storing one or more instructions; and At least one that executes the one or more instructions stored in the memory. Includes a processor, The operation performed by the above one or more instructions is The method includes a step of learning how an artificial intelligence model can restore missing RNA data from a dataset containing SNP information and RNA information, and The step here of learning how to restore the aforementioned missing RNA data is A step of extracting SNP features and RNA features from a training dataset containing SNP information and RNA information, A step of learning the process of an artificial intelligence model to generate SNP latent variables from the above SNP features, generate RNA latent variables from RNA features, and reconstruct SNP features from an integrated latent variable formed by integrating the SNP latent variables and the RNA latent variables, and A phenotype prediction system comprising a step of an artificial intelligence model learning to reconstruct RNA features from SNP features extracted from the above-mentioned training data.

2. A phenotype prediction system according to claim 1, characterized in that, in the step of reconstructing RNA features from SNP features extracted from the training data, an artificial intelligence model reconstructs SNP features from the same input value and generates RNA latent variables.

3. A phenotype prediction system according to claim 1, comprising the step of an artificial intelligence model that has completed training restoring missing RNA data and predicting a phenotype using the restored RNA data.

4. In claim 1, the step of learning the process of reconstructing SNP features from the integrated latent variable includes a step in which an artificial intelligence model learns using a mean squared error loss function, in a phenotype prediction system.

5. The phenotype prediction system according to claim 1, wherein the step of reconstructing RNA features from SNP features extracted from the training data includes the step of an artificial intelligence model generating RNA latent variables from SNP features.

6. A phenotype prediction system according to claim 1, further comprising a step of an artificial intelligence model learning a process of reconstructing RNA features from RNA features extracted from the training data.

7. A phenotype prediction system according to claim 6, wherein the artificial intelligence model in the step of reconstructing RNA features from RNA features extracted from the training data includes a model having a multilayer perceptron structure.

8. A phenotype prediction method performed by at least one processor, The method includes a step of learning how an artificial intelligence model can restore missing RNA data from a dataset containing SNP information and RNA information, and The step here of learning how to restore the aforementioned missing RNA data is A step of extracting SNP features and RNA features from a training dataset containing SNP information and RNA information, A step of learning the process of an artificial intelligence model to generate SNP latent variables from the above SNP features, generate RNA latent variables from RNA features, and reconstruct SNP features from an integrated latent variable formed by integrating the SNP latent variables and the RNA latent variables, and A phenotype prediction method comprising the step of an artificial intelligence model learning a process of reconstructing RNA features from SNP features extracted from the above-mentioned training data.

9. A phenotype prediction method according to claim 8, characterized in that, in the step of reconstructing RNA features from SNP features extracted from the above-mentioned training data, an artificial intelligence model reconstructs SNP features from the same input value and generates RNA latent variables.

10. A phenotype prediction method according to claim 8, comprising the step of an artificial intelligence model that has completed training restoring missing RNA data and predicting a phenotype using the restored RNA data.

11. A phenotype prediction method according to claim 8, wherein the step of learning the process of reconstructing SNP features from the integrated latent variable includes the step of an artificial intelligence model learning using a mean squared error loss function.

12. In claim 8, the step of reconstructing RNA features from SNP features extracted from the above-mentioned training data includes the step of an artificial intelligence model generating RNA latent variables from SNP features, in a phenotype prediction method.

13. A phenotype prediction method according to claim 8, further comprising the step of an artificial intelligence model learning a process of reconstructing RNA features from RNA features extracted from the above-mentioned training data.

14. A phenotype prediction method according to claim 13, wherein the artificial intelligence model in the step of reconstructing RNA features from RNA features extracted from the above-mentioned training data includes a model having a multilayer perceptron structure.

15. A program stored on a computer-readable recording medium to execute the method of any one of paragraphs 8 through 14 on a computer.

Images