System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating a molecular compound using a generative model; means for predicting effectiveness of the generated molecular compound; and means for integrating results of the generation and the prediction.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] The traditional drug development process requires high costs and long development times, and there is a demand for the rapid provision of new treatments, especially in an aging society. Furthermore, the process from compound discovery to efficacy evaluation is complex and often inefficient. To resolve this issue, a system is needed that can generate new molecular compounds and rapidly predict their efficacy. [Means for solving the problem]

[0005] The present invention provides a system equipped with a means for generating molecular compounds using a generative model and a means for predicting the efficacy of the generated molecular compounds. Specifically, the molecular structure of an input compound is converted into a SMILES representation, and the generative model generates a new molecular compound based on the SMILES representation. Furthermore, the efficacy of the generated molecular compound is evaluated using an algorithm for predicting its efficacy. In this way, by integrating the results of generation and prediction, a system is realized that streamlines the drug development process and enables the rapid provision of new treatments.

[0006] A "generative model" is a machine learning algorithm for generating new molecular compounds based on input molecular structures.

[0007] A "molecular compound" is a chemical molecule used as a drug, a substance with a specific chemical structure.

[0008] "SMILES notation" is a string-format chemical notation for concisely describing molecular structures, and is a method for expressing structural information about compounds in text.

[0009] "Efficacy" refers to the effect or efficacy of a molecular compound for a specific purpose (e.g., medicinal effect).

[0010] An "algorithm" refers to a set of computational procedures or rules for solving a particular problem, and is used in the present invention to predict the effectiveness of molecular compounds.

[0011] The term "integrate" refers to combining multiple different data or results into one set, and in the present invention means combining the generated compounds and the results of their efficacy. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2]1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram illustrating a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0013] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0016] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0017] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0018] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0019] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0020] [First embodiment]

[0021] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0022] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0023] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0024] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0025] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0026] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0027] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0028] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0032] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0033] The system of the present invention uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[0034] System configuration

[0035] Model initialization

[0036] The server initializes the generative AI model. The path to the model is specified, and the generative model is instantiated using this path.

[0037] Entering seed compounds

[0038] The user inputs the seed compound in SMILES notation, which is then converted into a molecular object.

[0039] Compound generation

[0040] The terminal generates new compounds using a generative AI model. First, the SMILES representation of the input seed compound is provided to the generative model, which then generates a specified number of new molecular compounds with the SMILES representation. The generated compounds are then converted back into molecular objects.

[0041] Prediction of efficacy

[0042] The server predicts the efficacy of the new molecular compounds generated. The efficacy of each compound is calculated using a specific algorithm, which basically evaluates the drug's efficacy based on the compound's structural information.

[0043] Integration of results

[0044] The server aggregates all generated molecular compounds and their predicted efficacy values, specifically, by creating a list of pairs of each compound's SMILES representation and its efficacy value.

[0045] Natural language processing explanation

[0046] The processing of the system will be explained in natural language below.

[0047] 1. Initialize the model

[0048] The server loads and initializes the generative model from the specified path. This generative model is used to generate new compounds based on the given seed compounds.

[0049] 2. Entering the species

[0050] The user inputs the SMILES notation of a seed compound, which becomes the basis for generating compounds. For example, suppose the SMILES notation of ethanol, "CCO", is input.

[0051] 3. Compound generation

[0052] The terminal passes this input SMILES notation to the generative model to generate a new molecular compound represented by SMILES notation, which is then converted back into a molecular object.

[0053] 4. Prediction of efficacy

[0054] The server has the function of predicting the efficacy of each compound generated. The efficacy of each compound is evaluated using a series of algorithms. For example, the prediction algorithm takes the compound's structure as input and outputs its efficacy as a numerical value.

[0055] 5. Synthesis of results

[0056] The server compiles the generated compounds and their predicted efficacies in a list format and presents it to the user. For example, pairs of the generated compounds' SMILES representations and their efficacies are displayed.

[0057] Specific examples

[0058] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each generated compound, the server will predict its effectiveness, resulting in a list like this:

[0059] SMILES: CCOC, Efficacy: 0.5839

[0060] SMILES: CCOCC, Efficacy: 0.7123

[0061] ...

[0062] This allows users to quickly confirm the structure and efficacy of each generated compound, making this system a powerful tool for streamlining the drug development process and accelerating the delivery of treatments.

[0063] The processing flow will be explained below.

[0064] Step 1:

[0065] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[0066] Step 2:

[0067] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[0068] Step 3:

[0069] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[0070] Step 4:

[0071] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[0072] Step 5:

[0073] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[0074] Step 6:

[0075] The server integrates the results with the molecular compounds whose efficacy is predicted, specifically by creating a list of pairs of the SMILES representation of each compound and its efficacy value.

[0076] Step 7:

[0077] The user receives the list provided by the server and checks the generated compounds and their effectiveness. For example, each entry in the list (the compound's SMILES representation and effectiveness value) is displayed on the screen. This allows the user to quickly evaluate the structure of a new compound and its effectiveness.

[0078] Example 1

[0079] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0080] In the modern drug development process, the generation and evaluation of novel active ingredients is a time-consuming and costly task. In particular, there is a need for methods to rapidly and efficiently generate a large number of candidate molecules and predict their efficacy at an early stage. Traditional methods require enormous resources, both experimentally and computationally, and new approaches are needed to solve this challenge.

[0081] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0082] In this invention, the server includes a means for loading and initializing a generative model from a specified path, a means for inputting the molecular structure of a seed compound in SMILES notation and converting it into a molecular object, a means for generating a new molecular compound in SMILES notation using the generative AI model, a means for predicting the effectiveness of the newly generated molecular compound, and a means for integrating the results of the generation and prediction, thereby enabling the rapid and efficient generation of a large number of candidate molecules and the prediction of their effectiveness.

[0083] The "specified path" is a string that indicates the directory where the generative model file is saved or the location of the file.

[0084] A "generative model" is a machine learning model for generating new molecular compounds from given input.

[0085] The "means for initialization" is a function that loads a generative model and executes a series of operations to make it usable.

[0086] A "seed compound" is an existing chemical compound from which new compounds can be made.

[0087] "SMILES notation" is a standard notation for expressing chemical structural formulas in string format.

[0088] A "molecular object" is a molecular representation generated from SMILES notation in a data format used for computational processing and structural analysis.

[0089] A "generative AI model" is a model for generating new molecular compounds using artificial intelligence technology.

[0090] The "means for predicting efficacy" is an algorithm or calculation method for calculating the pharmacological effect or other performance of the generated molecular compound.

[0091] The "means for integrating results" is a function that brings together the information on the generated compounds and the results of their effectiveness evaluation into a list or the like.

[0092] This invention relates to a system that uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[0093] The server first loads and initializes the generative AI model from the specified path. This generative model is often implemented using machine learning libraries such as TensorFlow or PyTorch. For example, the model path is specified as "models / generative_model.pth", and the model is instantiated based on this.

[0094] The user inputs the SMILES notation of the seed compound through the terminal. SMILES notation is a standard notation for expressing chemical structural formulas in string format, and the user starts generating a compound by inputting the SMILES notation of the compound. For example, the SMILES notation of ethanol, "CCO", is input.

[0095] The device receives the SMILES notation entered by the user and converts it into a molecular object using a library such as RDKit. The converted molecular object is then provided to a generative AI model to generate new molecular compounds. This generation process uses generative models such as the Variational Autoencoder (VAE) and Generative Adversarial Network (GAN).

[0096] The new SMILES-represented molecular compounds are then converted into molecular objects using the RDKit. For each compound, the server predicts its efficacy. This prediction uses algorithms such as the QED (Quantitative Estimation of Drug-likeness) score and molecular docking simulation. This allows the pharmacological effects and other performance of each compound to be evaluated numerically.

[0097] Finally, the server compiles a list of all the generated molecular compounds and their predicted efficacy, and presents it to the user. For example, the results are presented in a list format, with the SMILES representation of the generated compound paired with its efficacy value. As a concrete example, the compound generated based on ethanol and its predicted efficacy are shown below.

[0098] SMILES: CCOC, Efficacy: 0.5839

[0099] SMILES: CCOCC, Efficacy: 0.7123

[0100] ...

[0101] This system allows users to quickly confirm the structure and efficacy of each generated compound, contributing to the efficiency of the drug development process.

[0102] An example prompt sentence would be of the following format:

[0103] "Generate new molecular compounds based on the SMILES code 'CCO' and predict their efficacy."

[0104] This allows the generative AI model to automatically generate new molecular compounds based on the input SMILES notation and predict their effectiveness.

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Step 1:

[0107] The server loads and initializes the generative AI model from the specified path. The path to the generative model file is given as input. The initialized generative model is obtained as output. Specifically, the model file is loaded using TensorFlow or PyTorch.

[0108] Step 2:

[0109] The user inputs the SMILES notation of a seed compound into the terminal. The SMILES notation of the seed compound (for example, "CCO") is given as input. The input SMILES notation is used as is as output. Specifically, the user inputs the SMILES notation through an interface such as a text box.

[0110] Step 3:

[0111] The terminal converts the input SMILES notation into a molecule object. As input, a SMILES notation is given. As output, a molecule object is obtained. Specifically, a library such as RDKit is used to convert the SMILES notation into a molecule object.

[0112] Step 4:

[0113] The terminal provides a molecular object to the generative AI model to generate new molecular compounds in SMILES notation. As input, a molecular object and a generative model are given. As output, a list of new SMILES notations is obtained. Specifically, the terminal generates new molecular compounds using a generative model (VAE or GAN).

[0114] Step 5:

[0115] The terminal converts the generated SMILES representation back into molecule objects. As input, it is given a new SMILES representation. As output, it obtains a list of molecule objects. Specifically, it uses the RDKit again to convert the SMILES representation into molecule objects.

[0116] Step 6:

[0117] The server predicts the efficacy of each generated molecular compound. As input, it receives a list of generated molecular objects. As output, it obtains an efficacy score corresponding to each molecular compound. Specifically, it calculates efficacy using algorithms such as QED score and molecular docking simulation.

[0118] Step 7:

[0119] The server integrates the generated SMILES representations of compounds with their corresponding validity scores. As input, it receives pairs of the generated SMILES representations of molecular compounds and validity scores. As output, it obtains a list of the integrated results. Specifically, it compiles the results in a list format and presents them to the user.

[0120] Step 8:

[0121] The user checks the results through a terminal. As input, a list of integrated results from the server is given. As output, the results are displayed to the user. Specifically, the generated compounds and their effectiveness are viewed through a web browser or a dedicated app.

[0122] This makes the operation of the entire system more concrete and clear, and allows you to understand the detailed flow and specific operation of each processing step.

[0123] (Application example 1)

[0124] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0125] The traditional drug development process has the problem of taking a lot of time and effort to generate new molecular compounds and evaluate their effectiveness. To streamline this process, a system is needed that automates the generation of new compounds, efficacy evaluation, and integration of the results quickly and accurately. Furthermore, integration with factory automation is also necessary to operate this system effectively in pharmaceutical and chemical factories.

[0126] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0127] In this invention, the server includes a means for generating molecular compounds, a means for predicting the effectiveness of the generated molecular compounds, a means for integrating the results of the generation and prediction, a means for automating evaluation and reporting of the generated molecular compounds, a means for a user to input the molecular structure of the compound, and a means for linking with a factory automation system to enable use in chemical factories and pharmaceutical factories. This enables rapid and accurate evaluation of each generated molecular compound, and realizes efficiency and automation of the entire process.

[0128] A "molecular compound" is a chemical substance formed by bonding between molecules.

[0129] A "generating means" is a method or device for creating new molecular structures based on specific input data.

[0130] "Efficacy predictors" refer to algorithms or models for assessing the biological or pharmacological effects of generated compounds.

[0131] A "means of synthesizing results" is a method for compiling the generated molecular compounds and their predicted efficacy into a single list or report format.

[0132] The "means for automating evaluation and reporting" is a system for collecting data on generated compounds in real time, evaluating the data, and automatically presenting the results to the user.

[0133] The "means for inputting molecular structure" refers to an interface or device that allows a user to input structural information about a compound.

[0134] "Means for linking with factory automation systems to enable use in chemical and pharmaceutical factories" refers to a method or device for automating the product production and evaluation processes in a factory and linking the system with existing factory automation systems.

[0135] The present invention provides a system for producing molecular compounds and predicting their effectiveness in chemical and pharmaceutical factories. As an example, the present invention will be described using the system in a chemical factory.

[0136] System configuration and program description

[0137] 1. Initialize the model

[0138] The server first loads and initializes the generative AI model from the specified path. This generative AI model is used to generate new molecular compounds. The server manages the generative model and the efficacy prediction model.

[0139] 2. The user inputs the molecular structure

[0140] Users input the molecular structure of a compound in SMILES notation through a dedicated interface that is built into computer terminals used on-site at chemical and pharmaceutical plants.

[0141] 3. Generation of new compounds

[0142] The server passes the input SMILES notation to the generative AI model to generate a new molecular compound. Various data for the generated compound is then converted back into a molecular object.

[0143] 4. Prediction of efficacy of the resulting compounds

[0144] The server predicts the efficacy of each compound generated using a specific algorithm, and evaluates the efficacy of the compound based on its chemical structure.

[0145] 5. Synthesis and reporting of results

[0146] The server integrates the SMILES representations of all generated compounds and their predicted effectiveness. It compiles them into a list and provides immediate feedback to the user on the evaluation results. The evaluation results are also automatically compiled into a report and saved in collaboration with other systems within the factory.

[0147] Hardware and Software

[0148] Hardware used

[0149] Server (High-Performance Computing Server)

[0150] Factory terminals (user interface computers)

[0151] Factory automation systems (production equipment for chemical and pharmaceutical factories)

[0152] Software used

[0153] Generative AI model (model for generating molecular compounds)

[0154] Chemical information processing library (e.g., chemoinformatics)

[0155] Efficacy prediction algorithms (specific algorithms for predicting drug effects)

[0156] Specific examples of explanation

[0157] For example, if a user inputs the SMILES representation of ethanol ("CCO"), new molecular compounds are generated based on this SMILES representation. For each generated compound, the server predicts its effectiveness and aggregates the results in a list. The user receives a list like this:

[0158] SMILES: CCOC, Efficacy: 0.5839

[0159] SMILES: CCOCC, Efficacy: 0.7123

[0160] This allows users to quickly confirm the structure and efficacy of each compound generated. The system will significantly improve the efficiency of research and development processes in chemical and pharmaceutical factories.

[0161] Prompt Sentence Examples

[0162] "Generate new compounds based on ethanol (SMILES: "CCO") and predict their effectiveness."

[0163] In this way, this invention automates the creation of new molecular compounds and the prediction of their efficacy, greatly streamlining the research and development process. By linking with factory automation systems, it can be easily used in actual production sites, and is expected to speed up and improve the accuracy of drug efficacy evaluation.

[0164] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0165] Step 1:

[0166] The server initializes the generative model. The server loads the generative AI model from the specified path and initializes it. Specifically, the server loads the model file and instantiates it. This process makes the generative AI model ready to run. The input is the path to the model file, and the output is the initialized generative AI model.

[0167] Step 2:

[0168] A user uses a terminal to input the molecular structure of a compound in SMILES notation. This input is done through a dedicated interface. The terminal receives the SMILES notation entered by the user and sends it to the server. The input is the SMILES notation from the user, and the output is the SMILES notation received by the server.

[0169] Step 3:

[0170] The server passes the received SMILES notation to the generative AI model to generate new molecular compounds. Specifically, the server passes the input SMILES notation to the model, which generates multiple new SMILES notations. These new SMILES notations are converted back into molecular objects. The input is the SMILES notation received from the user, and the output is a list of the generated new molecular compounds.

[0171] Step 4:

[0172] The server predicts the efficacy of each generated molecular compound. This prediction is performed using a specific algorithm, which evaluates drug efficacy based on the input molecular structure information. Specifically, the server inputs the structural information of each molecular compound into the prediction algorithm and outputs a numerical value of efficacy. The input is a list of newly generated molecular compounds, and the output is a numerical value representing the efficacy of each compound.

[0173] Step 5:

[0174] The server integrates the generated SMILES representations of all compounds and their predicted efficacy. Specifically, the server pairs the SMILES representation of each compound with its efficacy value and compiles it into a list. This can be displayed or saved through an interface with the user or other systems. The input is the SMILES representation of each compound and its efficacy value, and the output is an integrated list.

[0175] Step 6:

[0176] The server generates a report from the integrated results and connects it to other automation systems in the chemical or pharmaceutical plant. The server outputs the generated report in an appropriate format and connects it to the plant's quality control system or database. The input is the integrated list, and the output is the report, along with its storage and connection.

[0177] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0178] The system of the present invention combines the basic functions of generating molecular compounds using generative models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system allows users to efficiently develop new drug compounds and consider emotional feedback on the results.

[0179] System configuration

[0180] Model initialization

[0181] The server initializes a generative AI model from a given model path, which is used to generate new molecular compounds based on the SMILES representation of the input seed compound.

[0182] Entering seed compounds

[0183] The user inputs the SMILES notation of the seed compound. For example, the SMILES notation of ethanol, "CCO," is input. A molecular object is generated based on this notation.

[0184] Compound generation

[0185] The terminal generates new compounds using a generative model. The input SMILES notation is passed to the generative model, and a specified number of new molecular compounds with SMILES notation are generated. The generated compounds are converted into molecular objects.

[0186] Prediction of efficacy

[0187] The server predicts the efficacy of each compound generated, and evaluates the efficacy of each compound numerically using a prediction algorithm.

[0188] Activating the Emotion Engine

[0189] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's emotional state based on their input and operation history. The emotion engine provides feedback based on the user's emotions and dynamically adjusts the generation and prediction process.

[0190] Results synthesis and feedback

[0191] The server integrates the generated compounds, their efficacy, and the user's emotional state, allowing the user to check the structure and efficacy of each generated compound while receiving feedback based on their own emotions.

[0192] Natural language processing explanation

[0193] The processing of the system will be specifically explained in natural language below.

[0194] 1. Initialize the model

[0195] The server loads and initializes a generative model from the specified model path. This generative model is used to generate new molecular structures based on seed compounds.

[0196] 2. Entering the species

[0197] The user inputs the SMILES notation of a seed compound (e.g., "CCO") into the terminal and generates a molecular object that will be the basis of the compound.

[0198] 3. Compound generation

[0199] The terminal passes the input SMILES notation to the generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[0200] 4. Prediction of efficacy

[0201] The server uses an algorithm to predict the efficacy of each compound, which quantifies the efficacy of each compound.

[0202] 5. Activating the Emotional Engine

[0203] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[0204] 6. Integration of results and feedback

[0205] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds, the efficacy of each compound, and emotional feedback.

[0206] Specific examples

[0207] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each compound, the server will predict its effectiveness and create a list like this:

[0208] SMILES: CCOC, Efficacy: 0.5839

[0209] SMILES: CCOCC, Efficacy: 0.7123

[0210] ...

[0211] Additionally, the emotion engine monitors the user's reactions and provides emotional feedback if the user is surprised by the generated results:

[0212] User Emotion: Surprise

[0213] Feedback: The compounds produced appear to be highly effective.

[0214] This process allows users to efficiently generate new compounds and verify their effectiveness, while receiving feedback that takes into account the user's emotions. The entire system dynamically adjusts to improve the user experience.

[0215] The processing flow will be explained below.

[0216] Step 1:

[0217] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[0218] Step 2:

[0219] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[0220] Step 3:

[0221] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[0222] Step 4:

[0223] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[0224] Step 5:

[0225] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[0226] Step 6:

[0227] The server activates an emotion engine based on the user's input history and operation patterns to analyze the user's emotional state. The emotion engine evaluates the user's emotions in real time and prepares emotion-based feedback.

[0228] Step 7:

[0229] The server integrates the molecular compounds predicted to be effective with the user's emotional state. Specifically, it compiles the SMILES representation of each generated compound, its effectiveness value, and the user's emotional state into a single list.

[0230] Step 8:

[0231] The user receives the list provided by the server, checks the generated compounds and their effectiveness, and receives feedback from the emotion engine. This allows the user to quickly evaluate the structure and effectiveness of new compounds, and also receive feedback based on their own emotions.

[0232] Example 2

[0233] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0234] Conventional molecular compound generation systems can generate new drug compounds and predict their efficacy, but they lack the ability to provide dynamic feedback based on the user's emotions. This limits the user experience and makes it difficult to obtain feedback based on the user's emotions. Furthermore, the generation and prediction processes are not fully integrated, resulting in incomplete information provided to the user.

[0235] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0236] In this invention, the server includes: a means for generating a molecular compound using a generative model; a means for predicting the effectiveness of the generated molecular compound; a means for accepting a notation of a seed compound input by a user; a prediction means including an algorithm for quantifying the effectiveness of the generated molecular compound; an emotion engine means for recognizing the emotional state of the user and providing feedback; and a means for integrating the results of the generation and prediction and generating feedback based on the user's emotions. This allows the user to receive feedback based on their own emotions while confirming the generation and effectiveness of a new molecular compound.

[0237] A "generative model" refers to an algorithm or program that generates new data based on input data.

[0238] A "molecular compound" refers to a compound made up of multiple atoms chemically bonded together.

[0239] "Efficacy" is a concept that indicates the effectiveness of a drug or compound for a specific purpose or effect.

[0240] "SMILES notation" is an abbreviation for Simplified Molecular Input Line Entry System, and refers to a method of expressing molecular structures as characters.

[0241] An "emotion engine" refers to a system or algorithm that recognizes and analyzes a user's emotional state based on their input and operation history, and provides appropriate feedback.

[0242] "Feedback" refers to information or guidance provided to a user by a system or algorithm, which changes depending on the user's behavior and emotions.

[0243] A "molecule object" is a data object that holds structural information about a chemical molecule and is used for analysis and simulation.

[0244] A "predictive algorithm" is a set of computational methods or procedures for analyzing given data and predicting future outcomes or effectiveness.

[0245] The system of the present invention combines the basic functions of generating molecular compounds using generative AI models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system is designed to enable users to efficiently develop new drug compounds and take into account emotional feedback on the results.

[0246] System configuration

[0247] The system mainly consists of three components: the server, the terminal, and the user. The operation of each component is described in detail below.

[0248] Model initialization

[0249] The server loads and initializes the generative AI model from the specified path. This generative model is used to generate new molecular compounds based on the SMILES representation of the input seed compound. For example, if the file path of the generative AI model is " / models / generative_model.pth", the server loads the model from this path and initializes the necessary parameters.

[0250] Entering seed compounds

[0251] The user uses the terminal's input form to input the SMILES representation of a seed compound. For example, the SMILES representation of ethanol is "CCO." Once this input is made, the terminal validates the input to ensure that it is in the correct format. The server then creates a molecule object based on this SMILES representation.

[0252] Compound generation

[0253] The terminal passes the user-entered SMILES notation to the generative AI model, which generates the specified number of new compounds. The generated new SMILES notation is converted into molecular objects by the server. These modified molecular objects are then analyzed for their shape and properties.

[0254] Prediction of efficacy

[0255] The server uses a specific algorithm to predict the efficacy of each compound generated, for example, by using a drug efficacy evaluation model to quantify the efficacy of the compound generated, and this quantified data is then provided to the user.

[0256] Activating the Emotion Engine

[0257] The server uses an emotion engine to analyze the user's emotional state. It recognizes the user's emotional state based on their input history and operation patterns, and generates the results as feedback. For example, it uses the "EmotionRecognitionAPI" to identify the user's emotions.

[0258] Results synthesis and feedback

[0259] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds and information about the efficacy of each compound. Feedback is also provided based on the user's emotions.

[0260] Specific examples

[0261] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. The server will generate the following list:

[0262] SMILES: CCOC, Efficacy: 0.5839

[0263] SMILES: CCOCC, Efficacy: 0.7123

[0264] ...

[0265] The emotion engine monitors the user's reactions and, for example, if the user is surprised by the generated results, provides that emotion as feedback:

[0266] User Emotion: Surprise

[0267] Feedback: The compounds produced appear to be highly effective.

[0268] This process allows users to generate new compounds and check their effectiveness while receiving feedback based on their own emotions. The entire system is dynamically adjusted to improve the user experience.

[0269] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0270] Step 1: Initialize the model

[0271] The server loads and initializes the generative AI model from the specified model path.

[0272] Input: Model file path (e.g. " / models / generative_model.pth")

[0273] Output: An initialized generative AI model

[0274] Specific operation: The server loads the model parameters based on the file path and initializes the generated AI model.

[0275] Step 2: Enter the seed compound

[0276] The user inputs the SMILES notation of the seed compound using an input form on the terminal.

[0277] Input: SMILES notation entered by the user (e.g. "CCO")

[0278] Output: Validation result of the input SMILES notation

[0279] Specific operation: The terminal validates the input SMILES notation and displays a warning if it is invalid. If it is valid, the server creates a molecule object based on the SMILES notation.

[0280] Step 3: Compound generation

[0281] The device passes the SMILES notation entered by the user to the generative AI model to generate new compounds.

[0282] Input: SMILES notation and generative AI model

[0283] Output: SMILES representation of the new compound generated

[0284] Specific operation: Based on the SMILES notation input to the generative AI model, the terminal generates a specified number of new SMILES notations. The server converts these generated SMILES notations into molecular objects.

[0285] Step 4: Predicting efficacy

[0286] The server uses a specific algorithm to predict the efficacy of each compound generated.

[0287] Input: Generated molecule object

[0288] Output: Potency score for each compound

[0289] Specific operation: The server uses a drug efficacy evaluation model to quantify and predict the efficacy of each compound generated.

[0290] Step 5: Activating the Emotion Engine

[0291] The server uses an emotion engine to analyze the user's emotional state.

[0292] Input: User input history and operation patterns

[0293] Output: User's emotional state and feedback

[0294] Specific operation: The emotion engine analyzes the user's input and operation history to recognize the current emotional state. Based on the recognized emotion, the server generates feedback.

[0295] Step 6: Consolidate and feedback results

[0296] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[0297] Input: Generated list of compounds, efficacy score for each compound, and user's emotional state

[0298] Output: Final result list and feedback to the user

[0299] Specific operation: The integrated result list and feedback information are provided to the user through the terminal, and the user decides on the next action based on this.

[0300] (Application example 2)

[0301] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0302] Conventional compound generation systems generate molecular compounds and predict their efficacy, but lack a dynamic adjustment function that takes into account the user's emotional feedback, resulting in issues with user experience and efficiency. Therefore, it is necessary to provide feedback that reflects the user's emotions about the generated compounds and adjust the generation and prediction process based on that feedback.

[0303] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating molecular compounds using a generative model, means for predicting the effectiveness of the generated molecular compounds, means for integrating the results of the generation and prediction, and means for dynamically adjusting the generation and prediction processes using an emotion engine that recognizes the user's emotional state. This makes it possible to dynamically adjust the compound generation process while providing feedback that reflects the user's emotions to the generated compounds.

[0304] A "generative model" is an algorithm or program for generating new molecular compounds based on the molecular structure of chemical substances.

[0305] A "molecular compound" is a chemical substance made up of multiple atoms bonded together.

[0306] "Efficacy" is an index used to evaluate the degree to which the produced molecular compound has the desired medicinal effect or function.

[0307] An "emotion engine" is a system or software for recognizing and analyzing a user's emotional state.

[0308] "Dynamic adjustment" means changing and optimizing processes in real time or incrementally based on user sentiment and other real-time data.

[0309] "SMILES notation" is a format for expressing molecular structures as strings of characters.

[0310] The present invention is a system that generates molecular compounds, predicts their effectiveness, and dynamically adjusts the process by taking into account the user's emotions. This system integrates a generative model, an emotion engine, and molecular generation and prediction functions. Specific embodiments for implementing the invention are described below.

[0311] System Program

[0312] The server uses a generative model to generate molecular compounds and predict their efficacy, and an emotion engine to recognize user emotions and dynamically adjust the process based on them.

[0313] Processing Description

[0314] Initializing the generative model

[0315] The server loads and initializes a generative model from a specified model path. This generative model is an algorithm that generates new molecular compounds based on the molecular structures of chemical substances.

[0316] Entering seed compounds

[0317] The user inputs the SMILES notation of a compound into the terminal and generates a molecular object that will be the basis for the compound. Specifically, the SMILES notation of ethanol, "CCO," is input.

[0318] Compound generation

[0319] The server passes the input SMILES representation to a generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[0320] Prediction of efficacy

[0321] The server predicts the efficacy of each compound generated. Using a prediction algorithm, the efficacy of each compound is evaluated as a numerical value.

[0322] Activating the Emotion Engine

[0323] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[0324] Results synthesis and feedback

[0325] The server integrates the SMILES representation of the generated compounds, the efficacy values, and the user's emotional state to finally provide an evaluation of the product and feedback to the user.

[0326] Hardware and software used

[0327] Software used:

[0328] Generative model (molecular generation algorithm)

[0329] Emotion engine (emotion recognition software)

[0330] Efficacy prediction algorithm

[0331] Hardware used:

[0332] Server (performs the generation and prediction process)

[0333] Terminal (where the user enters input)

[0334] Specific examples

[0335] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal generates a new compound. The server predicts the validity of this compound and creates a list like the following:

[0336] Compounds produced and their effectiveness:

[0337] SMILES: CCOC, Efficacy: 0.5839

[0338] SMILES: CCOCC, Efficacy: 0.7123

[0339] At the same time, the emotion engine monitors the user's reactions and provides emotional feedback if, for example, the user is surprised by the generated results.

[0340] User Emotion: Surprise

[0341] Feedback: The compound produced appears to be highly effective.

[0342] Prompt Sentence Examples

[0343] The following is a specific example of a prompt when entering information:

[0344] Generate new compounds based on the SMILES representation of medicinal compounds entered by the user. Then, predict the medicinal effects of the compounds and provide feedback based on the user's sentiment. Here is the user's input data:

[0345] SMILES: CCO (ethanol)

[0346] User operation history: ["input", "confirm"]

[0347] By following the above steps, the system can generate new molecular compounds based on the compounds entered by the user, predict their efficacy, and provide feedback that reflects the user's emotions, thereby improving the efficiency of the generation process and the user experience.

[0348] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0349] Step 1:

[0350] The server reads and initializes the generative model from a specified model path.

[0351] Specifically, the algorithm of the generative model is loaded and its environment is set up. At this stage, the generative model is ready to generate molecular structures.

[0352] The input data is a model path, and the output is an initialized generative model.

[0353] Step 2:

[0354] The user inputs the SMILES notation of the compound into the terminal.

[0355] Specifically, the user enters the SMILES representation of the compound (eg, "CCO") into a designated text field using a keyboard or other input device.

[0356] The input data is a SMILES representation, and the output is a SMILES representation of the input compound.

[0357] Step 3:

[0358] The server passes the input SMILES representation to the generative model and generates the specified number of new compounds.

[0359] Specifically, the generative model generates multiple new molecular compounds based on the input SMILES notation and outputs the SMILES notation of each compound.

[0360] The input data is the SMILES representation entered by the user, and the output is a list of new SMILES representations that are generated.

[0361] Step 4:

[0362] The server predicts the efficacy of each compound generated.

[0363] Specifically, the SMILES representation of each compound is input into a prediction algorithm, and its effectiveness is evaluated as a numerical value, which indicates the compound's efficacy.

[0364] The input data is a list of newly generated SMILES representations, and the output is a numerical value indicating the potency of each compound.

[0365] Step 5:

[0366] The server uses an emotion engine to analyze the user's emotional state.

[0367] Specifically, the emotion engine recognizes and analyzes the user's emotions based on the user's operation history and input data. The analysis results indicate the user's emotional state.

[0368] The input data is the user's operation history and input data, and the output is the analysis result of the user's emotional state.

[0369] Step 6:

[0370] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[0371] Specifically, the generated compound list, efficacy values, and the user's emotional state are provided to the user as feedback.

[0372] The input data are the SMILES representation of the generated compound, its efficacy value, and the user's emotional state, and the output is integrated feedback.

[0373] Step 7:

[0374] The user receives the feedback provided by the server.

[0375] Specifically, the user checks the feedback displayed on the device's display and makes a decision on the next step.

[0376] The input data is the consolidated feedback from the server and the output is the user's decision.

[0377] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0378] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0379] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0380] [Second embodiment]

[0381] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0382] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0383] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0384] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0385] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0386] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0387] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0388] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0389] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0390] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0391] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0392] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0393] The system of the present invention uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[0394] System configuration

[0395] Model initialization

[0396] The server initializes the generative AI model. The path to the model is specified, and the generative model is instantiated using this path.

[0397] Entering seed compounds

[0398] The user inputs the seed compound in SMILES notation, which is then converted into a molecular object.

[0399] Compound generation

[0400] The terminal generates new compounds using a generative AI model. First, the SMILES representation of the input seed compound is provided to the generative model, which then generates a specified number of new molecular compounds with the SMILES representation. The generated compounds are then converted back into molecular objects.

[0401] Prediction of efficacy

[0402] The server predicts the efficacy of the new molecular compounds generated. The efficacy of each compound is calculated using a specific algorithm, which basically evaluates the drug's efficacy based on the compound's structural information.

[0403] Integration of results

[0404] The server aggregates all generated molecular compounds and their predicted efficacy values, specifically, by creating a list of pairs of each compound's SMILES representation and its efficacy value.

[0405] Natural language processing explanation

[0406] The processing of the system will be explained in natural language below.

[0407] 1. Initialize the model

[0408] The server loads and initializes the generative model from the specified path. This generative model is used to generate new compounds based on the given seed compounds.

[0409] 2. Entering the species

[0410] The user inputs the SMILES notation of a seed compound, which becomes the basis for generating compounds. For example, suppose the SMILES notation of ethanol, "CCO", is input.

[0411] 3. Compound generation

[0412] The terminal passes this input SMILES notation to the generative model to generate a new molecular compound represented by SMILES notation, which is then converted back into a molecular object.

[0413] 4. Prediction of efficacy

[0414] The server has the function of predicting the efficacy of each compound generated. The efficacy of each compound is evaluated using a series of algorithms. For example, the prediction algorithm takes the compound's structure as input and outputs its efficacy as a numerical value.

[0415] 5. Synthesis of results

[0416] The server compiles the generated compounds and their predicted efficacies in a list format and presents it to the user. For example, pairs of the generated compounds' SMILES representations and their efficacies are displayed.

[0417] Specific examples

[0418] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each generated compound, the server will predict its effectiveness, resulting in a list like this:

[0419] SMILES: CCOC, Efficacy: 0.5839

[0420] SMILES: CCOCC, Efficacy: 0.7123

[0421] ...

[0422] This allows users to quickly confirm the structure and efficacy of each generated compound, making this system a powerful tool for streamlining the drug development process and accelerating the delivery of treatments.

[0423] The processing flow will be explained below.

[0424] Step 1:

[0425] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[0426] Step 2:

[0427] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[0428] Step 3:

[0429] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[0430] Step 4:

[0431] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[0432] Step 5:

[0433] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[0434] Step 6:

[0435] The server integrates the results with the molecular compounds whose efficacy is predicted, specifically by creating a list of pairs of the SMILES representation of each compound and its efficacy value.

[0436] Step 7:

[0437] The user receives the list provided by the server and checks the generated compounds and their effectiveness. For example, each entry in the list (the compound's SMILES representation and effectiveness value) is displayed on the screen. This allows the user to quickly evaluate the structure of a new compound and its effectiveness.

[0438] Example 1

[0439] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0440] In the modern drug development process, the generation and evaluation of novel active ingredients is a time-consuming and costly task. In particular, there is a need for methods to rapidly and efficiently generate a large number of candidate molecules and predict their efficacy at an early stage. Traditional methods require enormous resources, both experimentally and computationally, and new approaches are needed to solve this challenge.

[0441] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0442] In this invention, the server includes a means for loading and initializing a generative model from a specified path, a means for inputting the molecular structure of a seed compound in SMILES notation and converting it into a molecular object, a means for generating a new molecular compound in SMILES notation using the generative AI model, a means for predicting the effectiveness of the newly generated molecular compound, and a means for integrating the results of the generation and prediction, thereby enabling the rapid and efficient generation of a large number of candidate molecules and the prediction of their effectiveness.

[0443] The "specified path" is a string that indicates the directory where the generative model file is saved or the location of the file.

[0444] A "generative model" is a machine learning model for generating new molecular compounds from given input.

[0445] The "means for initialization" is a function that loads a generative model and executes a series of operations to make it usable.

[0446] A "seed compound" is an existing chemical compound from which new compounds can be made.

[0447] "SMILES notation" is a standard notation for expressing chemical structural formulas in string format.

[0448] A "molecular object" is a molecular representation generated from SMILES notation in a data format used for computational processing and structural analysis.

[0449] A "generative AI model" is a model for generating new molecular compounds using artificial intelligence technology.

[0450] The "means for predicting efficacy" is an algorithm or calculation method for calculating the pharmacological effect or other performance of the generated molecular compound.

[0451] The "means for integrating results" is a function that brings together the information on the generated compounds and the results of their effectiveness evaluation into a list or the like.

[0452] This invention relates to a system that uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[0453] The server first loads and initializes the generative AI model from the specified path. This generative model is often implemented using machine learning libraries such as TensorFlow or PyTorch. For example, the model path is specified as "models / generative_model.pth", and the model is instantiated based on this.

[0454] The user inputs the SMILES notation of the seed compound through the terminal. SMILES notation is a standard notation for expressing chemical structural formulas in string format, and the user starts generating a compound by inputting the SMILES notation of the compound. For example, the SMILES notation of ethanol, "CCO", is input.

[0455] The device receives the SMILES notation entered by the user and converts it into a molecular object using a library such as RDKit. The converted molecular object is then provided to a generative AI model to generate new molecular compounds. This generation process uses generative models such as the Variational Autoencoder (VAE) and Generative Adversarial Network (GAN).

[0456] The new SMILES-represented molecular compounds are then converted into molecular objects using the RDKit. For each compound, the server predicts its efficacy. This prediction uses algorithms such as the QED (Quantitative Estimation of Drug-likeness) score and molecular docking simulation. This allows the pharmacological effects and other performance of each compound to be evaluated numerically.

[0457] Finally, the server compiles a list of all the generated molecular compounds and their predicted efficacy, and presents it to the user. For example, the results are presented in a list format, with the SMILES representation of the generated compound paired with its efficacy value. As a concrete example, the compound generated based on ethanol and its predicted efficacy are shown below.

[0458] SMILES: CCOC, Efficacy: 0.5839

[0459] SMILES: CCOCC, Efficacy: 0.7123

[0460] ...

[0461] This system allows users to quickly confirm the structure and efficacy of each generated compound, contributing to the efficiency of the drug development process.

[0462] An example prompt sentence would be of the following format:

[0463] "Generate new molecular compounds based on the SMILES code 'CCO' and predict their efficacy."

[0464] This allows the generative AI model to automatically generate new molecular compounds based on the input SMILES notation and predict their effectiveness.

[0465] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0466] Step 1:

[0467] The server loads and initializes the generative AI model from the specified path. The path to the generative model file is given as input. The initialized generative model is obtained as output. Specifically, the model file is loaded using TensorFlow or PyTorch.

[0468] Step 2:

[0469] The user inputs the SMILES notation of a seed compound into the terminal. The SMILES notation of the seed compound (for example, "CCO") is given as input. The input SMILES notation is used as is as output. Specifically, the user inputs the SMILES notation through an interface such as a text box.

[0470] Step 3:

[0471] The terminal converts the input SMILES notation into a molecule object. As input, a SMILES notation is given. As output, a molecule object is obtained. Specifically, a library such as RDKit is used to convert the SMILES notation into a molecule object.

[0472] Step 4:

[0473] The terminal provides a molecular object to the generative AI model to generate new molecular compounds in SMILES notation. As input, a molecular object and a generative model are given. As output, a list of new SMILES notations is obtained. Specifically, the terminal generates new molecular compounds using a generative model (VAE or GAN).

[0474] Step 5:

[0475] The terminal converts the generated SMILES representation back into molecule objects. As input, it is given a new SMILES representation. As output, it obtains a list of molecule objects. Specifically, it uses the RDKit again to convert the SMILES representation into molecule objects.

[0476] Step 6:

[0477] The server predicts the efficacy of each generated molecular compound. As input, it receives a list of generated molecular objects. As output, it obtains an efficacy score corresponding to each molecular compound. Specifically, it calculates efficacy using algorithms such as QED score and molecular docking simulation.

[0478] Step 7:

[0479] The server integrates the generated SMILES representations of compounds with their corresponding validity scores. As input, it receives pairs of the generated SMILES representations of molecular compounds and validity scores. As output, it obtains a list of the integrated results. Specifically, it compiles the results in a list format and presents them to the user.

[0480] Step 8:

[0481] The user checks the results through a terminal. As input, a list of integrated results from the server is given. As output, the results are displayed to the user. Specifically, the generated compounds and their effectiveness are viewed through a web browser or a dedicated app.

[0482] This makes the operation of the entire system more concrete and clear, and allows you to understand the detailed flow and specific operation of each processing step.

[0483] (Application example 1)

[0484] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0485] The traditional drug development process has the problem of taking a lot of time and effort to generate new molecular compounds and evaluate their effectiveness. To streamline this process, a system is needed that automates the generation of new compounds, efficacy evaluation, and integration of the results quickly and accurately. Furthermore, integration with factory automation is also necessary to operate this system effectively in pharmaceutical and chemical factories.

[0486] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0487] In this invention, the server includes a means for generating molecular compounds, a means for predicting the effectiveness of the generated molecular compounds, a means for integrating the results of the generation and prediction, a means for automating evaluation and reporting of the generated molecular compounds, a means for a user to input the molecular structure of the compound, and a means for linking with a factory automation system to enable use in chemical factories and pharmaceutical factories. This enables rapid and accurate evaluation of each generated molecular compound, and realizes efficiency and automation of the entire process.

[0488] A "molecular compound" is a chemical substance formed by bonding between molecules.

[0489] A "generating means" is a method or device for creating new molecular structures based on specific input data.

[0490] "Efficacy predictors" refer to algorithms or models for assessing the biological or pharmacological effects of generated compounds.

[0491] A "means of synthesizing results" is a method for compiling the generated molecular compounds and their predicted efficacy into a single list or report format.

[0492] The "means for automating evaluation and reporting" is a system for collecting data on generated compounds in real time, evaluating the data, and automatically presenting the results to the user.

[0493] The "means for inputting molecular structure" refers to an interface or device that allows a user to input structural information about a compound.

[0494] "Means for linking with factory automation systems to enable use in chemical and pharmaceutical factories" refers to a method or device for automating the product production and evaluation processes in a factory and linking the system with existing factory automation systems.

[0495] The present invention provides a system for producing molecular compounds and predicting their effectiveness in chemical and pharmaceutical factories. As an example, the present invention will be described using the system in a chemical factory.

[0496] System configuration and program description

[0497] 1. Initialize the model

[0498] The server first loads and initializes the generative AI model from the specified path. This generative AI model is used to generate new molecular compounds. The server manages the generative model and the efficacy prediction model.

[0499] 2. The user inputs the molecular structure

[0500] Users input the molecular structure of a compound in SMILES notation through a dedicated interface that is built into computer terminals used on-site at chemical and pharmaceutical plants.

[0501] 3. Generation of new compounds

[0502] The server passes the input SMILES notation to the generative AI model to generate a new molecular compound. Various data for the generated compound is then converted back into a molecular object.

[0503] 4. Prediction of efficacy of the resulting compounds

[0504] The server predicts the efficacy of each compound generated using a specific algorithm, and evaluates the efficacy of the compound based on its chemical structure.

[0505] 5. Synthesis and reporting of results

[0506] The server integrates the SMILES representations of all generated compounds and their predicted effectiveness. It compiles them into a list and provides immediate feedback to the user on the evaluation results. The evaluation results are also automatically compiled into a report and saved in collaboration with other systems within the factory.

[0507] Hardware and Software

[0508] Hardware used

[0509] Server (High-Performance Computing Server)

[0510] Factory terminals (user interface computers)

[0511] Factory automation systems (production equipment for chemical and pharmaceutical factories)

[0512] Software used

[0513] Generative AI model (model for generating molecular compounds)

[0514] Chemical information processing library (e.g., chemoinformatics)

[0515] Efficacy prediction algorithms (specific algorithms for predicting drug effects)

[0516] Specific examples of explanation

[0517] For example, if a user inputs the SMILES representation of ethanol ("CCO"), new molecular compounds are generated based on this SMILES representation. For each generated compound, the server predicts its effectiveness and aggregates the results in a list. The user receives a list like this:

[0518] SMILES: CCOC, Efficacy: 0.5839

[0519] SMILES: CCOCC, Efficacy: 0.7123

[0520] This allows users to quickly confirm the structure and efficacy of each compound generated. The system will significantly improve the efficiency of research and development processes in chemical and pharmaceutical factories.

[0521] Prompt Sentence Examples

[0522] "Generate new compounds based on ethanol (SMILES: "CCO") and predict their effectiveness."

[0523] In this way, this invention automates the creation of new molecular compounds and the prediction of their efficacy, greatly streamlining the research and development process. By linking with factory automation systems, it can be easily used in actual production sites, and is expected to speed up and improve the accuracy of drug efficacy evaluation.

[0524] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0525] Step 1:

[0526] The server initializes the generative model. The server loads the generative AI model from the specified path and initializes it. Specifically, the server loads the model file and instantiates it. This process makes the generative AI model ready to run. The input is the path to the model file, and the output is the initialized generative AI model.

[0527] Step 2:

[0528] A user uses a terminal to input the molecular structure of a compound in SMILES notation. This input is done through a dedicated interface. The terminal receives the SMILES notation entered by the user and sends it to the server. The input is the SMILES notation from the user, and the output is the SMILES notation received by the server.

[0529] Step 3:

[0530] The server passes the received SMILES notation to the generative AI model to generate new molecular compounds. Specifically, the server passes the input SMILES notation to the model, which generates multiple new SMILES notations. These new SMILES notations are converted back into molecular objects. The input is the SMILES notation received from the user, and the output is a list of the generated new molecular compounds.

[0531] Step 4:

[0532] The server predicts the efficacy of each generated molecular compound. This prediction is performed using a specific algorithm, which evaluates drug efficacy based on the input molecular structure information. Specifically, the server inputs the structural information of each molecular compound into the prediction algorithm and outputs a numerical value of efficacy. The input is a list of newly generated molecular compounds, and the output is a numerical value representing the efficacy of each compound.

[0533] Step 5:

[0534] The server integrates the generated SMILES representations of all compounds and their predicted efficacy. Specifically, the server pairs the SMILES representation of each compound with its efficacy value and compiles it into a list. This can be displayed or saved through an interface with the user or other systems. The input is the SMILES representation of each compound and its efficacy value, and the output is an integrated list.

[0535] Step 6:

[0536] The server generates a report from the integrated results and connects it to other automation systems in the chemical or pharmaceutical plant. The server outputs the generated report in an appropriate format and connects it to the plant's quality control system or database. The input is the integrated list, and the output is the report, along with its storage and connection.

[0537] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0538] The system of the present invention combines the basic functions of generating molecular compounds using generative models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system allows users to efficiently develop new drug compounds and consider emotional feedback on the results.

[0539] System configuration

[0540] Model initialization

[0541] The server initializes a generative AI model from a given model path, which is used to generate new molecular compounds based on the SMILES representation of the input seed compound.

[0542] Entering seed compounds

[0543] The user inputs the SMILES notation of the seed compound. For example, the SMILES notation of ethanol, "CCO," is input. A molecular object is generated based on this notation.

[0544] Compound generation

[0545] The terminal generates new compounds using a generative model. The input SMILES notation is passed to the generative model, and a specified number of new molecular compounds with SMILES notation are generated. The generated compounds are converted into molecular objects.

[0546] Prediction of efficacy

[0547] The server predicts the efficacy of each compound generated, and evaluates the efficacy of each compound numerically using a prediction algorithm.

[0548] Activating the Emotion Engine

[0549] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's emotional state based on their input and operation history. The emotion engine provides feedback based on the user's emotions and dynamically adjusts the generation and prediction process.

[0550] Results synthesis and feedback

[0551] The server integrates the generated compounds, their efficacy, and the user's emotional state, allowing the user to check the structure and efficacy of each generated compound while receiving feedback based on their own emotions.

[0552] Natural language processing explanation

[0553] The processing of the system will be specifically explained in natural language below.

[0554] 1. Initialize the model

[0555] The server loads and initializes a generative model from the specified model path. This generative model is used to generate new molecular structures based on seed compounds.

[0556] 2. Entering the species

[0557] The user inputs the SMILES notation of a seed compound (e.g., "CCO") into the terminal and generates a molecular object that will be the basis of the compound.

[0558] 3. Compound generation

[0559] The terminal passes the input SMILES notation to the generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[0560] 4. Prediction of efficacy

[0561] The server uses an algorithm to predict the efficacy of each compound, which quantifies the efficacy of each compound.

[0562] 5. Activating the Emotional Engine

[0563] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[0564] 6. Integration of results and feedback

[0565] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds, the efficacy of each compound, and emotional feedback.

[0566] Specific examples

[0567] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each compound, the server will predict its effectiveness and create a list like this:

[0568] SMILES: CCOC, Efficacy: 0.5839

[0569] SMILES: CCOCC, Efficacy: 0.7123

[0570] ...

[0571] Additionally, the emotion engine monitors the user's reactions and provides emotional feedback if the user is surprised by the generated results:

[0572] User Emotion: Surprise

[0573] Feedback: The compounds produced appear to be highly effective.

[0574] This process allows users to efficiently generate new compounds and verify their effectiveness, while receiving feedback that takes into account the user's emotions. The entire system dynamically adjusts to improve the user experience.

[0575] The processing flow will be explained below.

[0576] Step 1:

[0577] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[0578] Step 2:

[0579] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[0580] Step 3:

[0581] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[0582] Step 4:

[0583] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[0584] Step 5:

[0585] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[0586] Step 6:

[0587] The server activates an emotion engine based on the user's input history and operation patterns to analyze the user's emotional state. The emotion engine evaluates the user's emotions in real time and prepares emotion-based feedback.

[0588] Step 7:

[0589] The server integrates the molecular compounds predicted to be effective with the user's emotional state. Specifically, it compiles the SMILES representation of each generated compound, its effectiveness value, and the user's emotional state into a single list.

[0590] Step 8:

[0591] The user receives the list provided by the server, checks the generated compounds and their effectiveness, and receives feedback from the emotion engine. This allows the user to quickly evaluate the structure and effectiveness of new compounds, and also receive feedback based on their own emotions.

[0592] Example 2

[0593] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0594] Conventional molecular compound generation systems can generate new drug compounds and predict their efficacy, but they lack the ability to provide dynamic feedback based on the user's emotions. This limits the user experience and makes it difficult to obtain feedback based on the user's emotions. Furthermore, the generation and prediction processes are not fully integrated, resulting in incomplete information provided to the user.

[0595] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0596] In this invention, the server includes: a means for generating a molecular compound using a generative model; a means for predicting the effectiveness of the generated molecular compound; a means for accepting a notation of a seed compound input by a user; a prediction means including an algorithm for quantifying the effectiveness of the generated molecular compound; an emotion engine means for recognizing the emotional state of the user and providing feedback; and a means for integrating the results of the generation and prediction and generating feedback based on the user's emotions. This allows the user to receive feedback based on their own emotions while confirming the generation and effectiveness of a new molecular compound.

[0597] A "generative model" refers to an algorithm or program that generates new data based on input data.

[0598] A "molecular compound" refers to a compound made up of multiple atoms chemically bonded together.

[0599] "Efficacy" is a concept that indicates the effectiveness of a drug or compound for a specific purpose or effect.

[0600] "SMILES notation" is an abbreviation for Simplified Molecular Input Line Entry System, and refers to a method of expressing molecular structures as characters.

[0601] An "emotion engine" refers to a system or algorithm that recognizes and analyzes a user's emotional state based on their input and operation history, and provides appropriate feedback.

[0602] "Feedback" refers to information or guidance provided to a user by a system or algorithm, which changes depending on the user's behavior and emotions.

[0603] A "molecule object" is a data object that holds structural information about a chemical molecule and is used for analysis and simulation.

[0604] A "predictive algorithm" is a set of computational methods or procedures for analyzing given data and predicting future outcomes or effectiveness.

[0605] The system of the present invention combines the basic functions of generating molecular compounds using generative AI models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system is designed to enable users to efficiently develop new drug compounds and take into account emotional feedback on the results.

[0606] System configuration

[0607] The system mainly consists of three components: the server, the terminal, and the user. The operation of each component is described in detail below.

[0608] Model initialization

[0609] The server loads and initializes the generative AI model from the specified path. This generative model is used to generate new molecular compounds based on the SMILES representation of the input seed compound. For example, if the file path of the generative AI model is " / models / generative_model.pth", the server loads the model from this path and initializes the necessary parameters.

[0610] Entering seed compounds

[0611] The user uses the terminal's input form to input the SMILES representation of a seed compound. For example, the SMILES representation of ethanol is "CCO." Once this input is made, the terminal validates the input to ensure that it is in the correct format. The server then creates a molecule object based on this SMILES representation.

[0612] Compound generation

[0613] The terminal passes the user-entered SMILES notation to the generative AI model, which generates the specified number of new compounds. The generated new SMILES notation is converted into molecular objects by the server. These modified molecular objects are then analyzed for their shape and properties.

[0614] Prediction of efficacy

[0615] The server uses a specific algorithm to predict the efficacy of each compound generated, for example, by using a drug efficacy evaluation model to quantify the efficacy of the compound generated, and this quantified data is then provided to the user.

[0616] Activating the Emotion Engine

[0617] The server uses an emotion engine to analyze the user's emotional state. It recognizes the user's emotional state based on their input history and operation patterns, and generates the results as feedback. For example, it uses the "EmotionRecognitionAPI" to identify the user's emotions.

[0618] Results synthesis and feedback

[0619] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds and information about the efficacy of each compound. Feedback is also provided based on the user's emotions.

[0620] Specific examples

[0621] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. The server will generate the following list:

[0622] SMILES: CCOC, Efficacy: 0.5839

[0623] SMILES: CCOCC, Efficacy: 0.7123

[0624] ...

[0625] The emotion engine monitors the user's reactions and, for example, if the user is surprised by the generated results, provides that emotion as feedback:

[0626] User Emotion: Surprise

[0627] Feedback: The compounds produced appear to be highly effective.

[0628] This process allows users to generate new compounds and check their effectiveness while receiving feedback based on their own emotions. The entire system is dynamically adjusted to improve the user experience.

[0629] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0630] Step 1: Initialize the model

[0631] The server loads and initializes the generative AI model from the specified model path.

[0632] Input: Model file path (e.g. " / models / generative_model.pth")

[0633] Output: An initialized generative AI model

[0634] Specific operation: The server loads the model parameters based on the file path and initializes the generated AI model.

[0635] Step 2: Enter the seed compound

[0636] The user inputs the SMILES notation of the seed compound using an input form on the terminal.

[0637] Input: SMILES notation entered by the user (e.g. "CCO")

[0638] Output: Validation result of the input SMILES notation

[0639] Specific operation: The terminal validates the input SMILES notation and displays a warning if it is invalid. If it is valid, the server creates a molecule object based on the SMILES notation.

[0640] Step 3: Compound generation

[0641] The device passes the SMILES notation entered by the user to the generative AI model to generate new compounds.

[0642] Input: SMILES notation and generative AI model

[0643] Output: SMILES representation of the new compound generated

[0644] Specific operation: Based on the SMILES notation input to the generative AI model, the terminal generates a specified number of new SMILES notations. The server converts these generated SMILES notations into molecular objects.

[0645] Step 4: Predicting efficacy

[0646] The server uses a specific algorithm to predict the efficacy of each compound generated.

[0647] Input: Generated molecule object

[0648] Output: Potency score for each compound

[0649] Specific operation: The server uses a drug efficacy evaluation model to quantify and predict the efficacy of each compound generated.

[0650] Step 5: Activating the Emotion Engine

[0651] The server uses an emotion engine to analyze the user's emotional state.

[0652] Input: User input history and operation patterns

[0653] Output: User's emotional state and feedback

[0654] Specific operation: The emotion engine analyzes the user's input and operation history to recognize the current emotional state. Based on the recognized emotion, the server generates feedback.

[0655] Step 6: Consolidate and feedback results

[0656] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[0657] Input: Generated list of compounds, efficacy score for each compound, and user's emotional state

[0658] Output: Final result list and feedback to the user

[0659] Specific operation: The integrated result list and feedback information are provided to the user through the terminal, and the user decides on the next action based on this.

[0660] (Application example 2)

[0661] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0662] Conventional compound generation systems generate molecular compounds and predict their efficacy, but lack a dynamic adjustment function that takes into account the user's emotional feedback, resulting in issues with user experience and efficiency. Therefore, it is necessary to provide feedback that reflects the user's emotions about the generated compounds and adjust the generation and prediction process based on that feedback.

[0663] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating molecular compounds using a generative model, means for predicting the effectiveness of the generated molecular compounds, means for integrating the results of the generation and prediction, and means for dynamically adjusting the generation and prediction processes using an emotion engine that recognizes the user's emotional state. This makes it possible to dynamically adjust the compound generation process while providing feedback that reflects the user's emotions to the generated compounds.

[0664] A "generative model" is an algorithm or program for generating new molecular compounds based on the molecular structure of chemical substances.

[0665] A "molecular compound" is a chemical substance made up of multiple atoms bonded together.

[0666] "Efficacy" is an index used to evaluate the degree to which the produced molecular compound has the desired medicinal effect or function.

[0667] An "emotion engine" is a system or software for recognizing and analyzing a user's emotional state.

[0668] "Dynamic adjustment" means changing and optimizing processes in real time or incrementally based on user sentiment and other real-time data.

[0669] "SMILES notation" is a format for expressing molecular structures as strings of characters.

[0670] The present invention is a system that generates molecular compounds, predicts their effectiveness, and dynamically adjusts the process by taking into account the user's emotions. This system integrates a generative model, an emotion engine, and molecular generation and prediction functions. Specific embodiments for implementing the invention are described below.

[0671] System Program

[0672] The server uses a generative model to generate molecular compounds and predict their efficacy, and an emotion engine to recognize user emotions and dynamically adjust the process based on them.

[0673] Processing Description

[0674] Initializing the generative model

[0675] The server loads and initializes a generative model from a specified model path. This generative model is an algorithm that generates new molecular compounds based on the molecular structures of chemical substances.

[0676] Entering seed compounds

[0677] The user inputs the SMILES notation of a compound into the terminal and generates a molecular object that will be the basis for the compound. Specifically, the SMILES notation of ethanol, "CCO," is input.

[0678] Compound generation

[0679] The server passes the input SMILES representation to a generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[0680] Prediction of efficacy

[0681] The server predicts the efficacy of each compound generated. Using a prediction algorithm, the efficacy of each compound is evaluated as a numerical value.

[0682] Activating the Emotion Engine

[0683] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[0684] Results synthesis and feedback

[0685] The server integrates the SMILES representation of the generated compounds, the efficacy values, and the user's emotional state to finally provide an evaluation of the product and feedback to the user.

[0686] Hardware and software used

[0687] Software used:

[0688] Generative model (molecular generation algorithm)

[0689] Emotion engine (emotion recognition software)

[0690] Efficacy prediction algorithm

[0691] Hardware used:

[0692] Server (performs the generation and prediction process)

[0693] Terminal (where the user enters input)

[0694] Specific examples

[0695] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal generates a new compound. The server predicts the validity of this compound and creates a list like the following:

[0696] Compounds produced and their effectiveness:

[0697] SMILES: CCOC, Efficacy: 0.5839

[0698] SMILES: CCOCC, Efficacy: 0.7123

[0699] At the same time, the emotion engine monitors the user's reactions and provides emotional feedback if, for example, the user is surprised by the generated results.

[0700] User Emotion: Surprise

[0701] Feedback: The compound produced appears to be highly effective.

[0702] Prompt Sentence Examples

[0703] The following is a specific example of a prompt when entering information:

[0704] Generate new compounds based on the SMILES representation of medicinal compounds entered by the user. Then, predict the medicinal effects of the compounds and provide feedback based on the user's sentiment. Here is the user's input data:

[0705] SMILES: CCO (ethanol)

[0706] User operation history: ["input", "confirm"]

[0707] By following the above steps, the system can generate new molecular compounds based on the compounds entered by the user, predict their efficacy, and provide feedback that reflects the user's emotions, thereby improving the efficiency of the generation process and the user experience.

[0708] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0709] Step 1:

[0710] The server reads and initializes the generative model from a specified model path.

[0711] Specifically, the algorithm of the generative model is loaded and its environment is set up. At this stage, the generative model is ready to generate molecular structures.

[0712] The input data is a model path, and the output is an initialized generative model.

[0713] Step 2:

[0714] The user inputs the SMILES notation of the compound into the terminal.

[0715] Specifically, the user enters the SMILES representation of the compound (eg, "CCO") into a designated text field using a keyboard or other input device.

[0716] The input data is a SMILES representation, and the output is a SMILES representation of the input compound.

[0717] Step 3:

[0718] The server passes the input SMILES representation to the generative model and generates the specified number of new compounds.

[0719] Specifically, the generative model generates multiple new molecular compounds based on the input SMILES notation and outputs the SMILES notation of each compound.

[0720] The input data is the SMILES representation entered by the user, and the output is a list of new SMILES representations that are generated.

[0721] Step 4:

[0722] The server predicts the efficacy of each compound generated.

[0723] Specifically, the SMILES representation of each compound is input into a prediction algorithm, and its effectiveness is evaluated as a numerical value, which indicates the compound's efficacy.

[0724] The input data is a list of newly generated SMILES representations, and the output is a numerical value indicating the potency of each compound.

[0725] Step 5:

[0726] The server uses an emotion engine to analyze the user's emotional state.

[0727] Specifically, the emotion engine recognizes and analyzes the user's emotions based on the user's operation history and input data. The analysis results indicate the user's emotional state.

[0728] The input data is the user's operation history and input data, and the output is the analysis result of the user's emotional state.

[0729] Step 6:

[0730] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[0731] Specifically, the generated compound list, efficacy values, and the user's emotional state are provided to the user as feedback.

[0732] The input data are the SMILES representation of the generated compound, its efficacy value, and the user's emotional state, and the output is integrated feedback.

[0733] Step 7:

[0734] The user receives the feedback provided by the server.

[0735] Specifically, the user checks the feedback displayed on the device's display and makes a decision on the next step.

[0736] The input data is the consolidated feedback from the server and the output is the user's decision.

[0737] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0738] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0739] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0740] [Third embodiment]

[0741] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0742] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0743] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0744] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0745] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0746] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0747] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0748] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0749] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0750] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0751] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0752] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0753] The system of the present invention uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[0754] System configuration

[0755] Model initialization

[0756] The server initializes the generative AI model. The path to the model is specified, and the generative model is instantiated using this path.

[0757] Entering seed compounds

[0758] The user inputs the seed compound in SMILES notation, which is then converted into a molecular object.

[0759] Compound generation

[0760] The terminal generates new compounds using a generative AI model. First, the SMILES representation of the input seed compound is provided to the generative model, which then generates a specified number of new molecular compounds with the SMILES representation. The generated compounds are then converted back into molecular objects.

[0761] Prediction of efficacy

[0762] The server predicts the efficacy of the new molecular compounds generated. The efficacy of each compound is calculated using a specific algorithm, which basically evaluates the drug's efficacy based on the compound's structural information.

[0763] Integration of results

[0764] The server aggregates all generated molecular compounds and their predicted efficacy values, specifically, by creating a list of pairs of each compound's SMILES representation and its efficacy value.

[0765] Natural language processing explanation

[0766] The processing of the system will be explained in natural language below.

[0767] 1. Initialize the model

[0768] The server loads and initializes the generative model from the specified path. This generative model is used to generate new compounds based on the given seed compounds.

[0769] 2. Entering the species

[0770] The user inputs the SMILES notation of a seed compound, which becomes the basis for generating compounds. For example, suppose the SMILES notation of ethanol, "CCO", is input.

[0771] 3. Compound generation

[0772] The terminal passes this input SMILES notation to the generative model to generate a new molecular compound represented by SMILES notation, which is then converted back into a molecular object.

[0773] 4. Prediction of efficacy

[0774] The server has the function of predicting the efficacy of each compound generated. The efficacy of each compound is evaluated using a series of algorithms. For example, the prediction algorithm takes the compound's structure as input and outputs its efficacy as a numerical value.

[0775] 5. Synthesis of results

[0776] The server compiles the generated compounds and their predicted efficacies in a list format and presents it to the user. For example, pairs of the generated compounds' SMILES representations and their efficacies are displayed.

[0777] Specific examples

[0778] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each generated compound, the server will predict its effectiveness, resulting in a list like this:

[0779] SMILES: CCOC, Efficacy: 0.5839

[0780] SMILES: CCOCC, Efficacy: 0.7123

[0781] ...

[0782] This allows users to quickly confirm the structure and efficacy of each generated compound, making this system a powerful tool for streamlining the drug development process and accelerating the delivery of treatments.

[0783] The processing flow will be explained below.

[0784] Step 1:

[0785] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[0786] Step 2:

[0787] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[0788] Step 3:

[0789] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[0790] Step 4:

[0791] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[0792] Step 5:

[0793] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[0794] Step 6:

[0795] The server integrates the results with the molecular compounds whose efficacy is predicted, specifically by creating a list of pairs of the SMILES representation of each compound and its efficacy value.

[0796] Step 7:

[0797] The user receives the list provided by the server and checks the generated compounds and their effectiveness. For example, each entry in the list (the compound's SMILES representation and effectiveness value) is displayed on the screen. This allows the user to quickly evaluate the structure of a new compound and its effectiveness.

[0798] Example 1

[0799] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0800] In the modern drug development process, the generation and evaluation of novel active ingredients is a time-consuming and costly task. In particular, there is a need for methods to rapidly and efficiently generate a large number of candidate molecules and predict their efficacy at an early stage. Traditional methods require enormous resources, both experimentally and computationally, and new approaches are needed to solve this challenge.

[0801] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0802] In this invention, the server includes a means for loading and initializing a generative model from a specified path, a means for inputting the molecular structure of a seed compound in SMILES notation and converting it into a molecular object, a means for generating a new molecular compound in SMILES notation using the generative AI model, a means for predicting the effectiveness of the newly generated molecular compound, and a means for integrating the results of the generation and prediction, thereby enabling the rapid and efficient generation of a large number of candidate molecules and the prediction of their effectiveness.

[0803] The "specified path" is a string that indicates the directory where the generative model file is saved or the location of the file.

[0804] A "generative model" is a machine learning model for generating new molecular compounds from given input.

[0805] The "means for initialization" is a function that loads a generative model and executes a series of operations to make it usable.

[0806] A "seed compound" is an existing chemical compound from which new compounds can be made.

[0807] "SMILES notation" is a standard notation for expressing chemical structural formulas in string format.

[0808] A "molecular object" is a molecular representation generated from SMILES notation in a data format used for computational processing and structural analysis.

[0809] A "generative AI model" is a model for generating new molecular compounds using artificial intelligence technology.

[0810] The "means for predicting efficacy" is an algorithm or calculation method for calculating the pharmacological effect or other performance of the generated molecular compound.

[0811] The "means for integrating results" is a function that brings together the information on the generated compounds and the results of their effectiveness evaluation into a list or the like.

[0812] This invention relates to a system that uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[0813] The server first loads and initializes the generative AI model from the specified path. This generative model is often implemented using machine learning libraries such as TensorFlow or PyTorch. For example, the model path is specified as "models / generative_model.pth", and the model is instantiated based on this.

[0814] The user inputs the SMILES notation of the seed compound through the terminal. SMILES notation is a standard notation for expressing chemical structural formulas in string format, and the user starts generating a compound by inputting the SMILES notation of the compound. For example, the SMILES notation of ethanol, "CCO", is input.

[0815] The device receives the SMILES notation entered by the user and converts it into a molecular object using a library such as RDKit. The converted molecular object is then provided to a generative AI model to generate new molecular compounds. This generation process uses generative models such as the Variational Autoencoder (VAE) and Generative Adversarial Network (GAN).

[0816] The new SMILES-represented molecular compounds are then converted into molecular objects using the RDKit. For each compound, the server predicts its efficacy. This prediction uses algorithms such as the QED (Quantitative Estimation of Drug-likeness) score and molecular docking simulation. This allows the pharmacological effects and other performance of each compound to be evaluated numerically.

[0817] Finally, the server compiles a list of all the generated molecular compounds and their predicted efficacy, and presents it to the user. For example, the results are presented in a list format, with the SMILES representation of the generated compound paired with its efficacy value. As a concrete example, the compound generated based on ethanol and its predicted efficacy are shown below.

[0818] SMILES: CCOC, Efficacy: 0.5839

[0819] SMILES: CCOCC, Efficacy: 0.7123

[0820] ...

[0821] This system allows users to quickly confirm the structure and efficacy of each generated compound, contributing to the efficiency of the drug development process.

[0822] An example prompt sentence would be of the following format:

[0823] "Generate new molecular compounds based on the SMILES code 'CCO' and predict their efficacy."

[0824] This allows the generative AI model to automatically generate new molecular compounds based on the input SMILES notation and predict their effectiveness.

[0825] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0826] Step 1:

[0827] The server loads and initializes the generative AI model from the specified path. The path to the generative model file is given as input. The initialized generative model is obtained as output. Specifically, the model file is loaded using TensorFlow or PyTorch.

[0828] Step 2:

[0829] The user inputs the SMILES notation of a seed compound into the terminal. The SMILES notation of the seed compound (for example, "CCO") is given as input. The input SMILES notation is used as is as output. Specifically, the user inputs the SMILES notation through an interface such as a text box.

[0830] Step 3:

[0831] The terminal converts the input SMILES notation into a molecule object. As input, a SMILES notation is given. As output, a molecule object is obtained. Specifically, a library such as RDKit is used to convert the SMILES notation into a molecule object.

[0832] Step 4:

[0833] The terminal provides a molecular object to the generative AI model to generate new molecular compounds in SMILES notation. As input, a molecular object and a generative model are given. As output, a list of new SMILES notations is obtained. Specifically, the terminal generates new molecular compounds using a generative model (VAE or GAN).

[0834] Step 5:

[0835] The terminal converts the generated SMILES representation back into molecule objects. As input, it is given a new SMILES representation. As output, it obtains a list of molecule objects. Specifically, it uses the RDKit again to convert the SMILES representation into molecule objects.

[0836] Step 6:

[0837] The server predicts the efficacy of each generated molecular compound. As input, it receives a list of generated molecular objects. As output, it obtains an efficacy score corresponding to each molecular compound. Specifically, it calculates efficacy using algorithms such as QED score and molecular docking simulation.

[0838] Step 7:

[0839] The server integrates the generated SMILES representations of compounds with their corresponding validity scores. As input, it receives pairs of the generated SMILES representations of molecular compounds and validity scores. As output, it obtains a list of the integrated results. Specifically, it compiles the results in a list format and presents them to the user.

[0840] Step 8:

[0841] The user checks the results through a terminal. As input, a list of integrated results from the server is given. As output, the results are displayed to the user. Specifically, the generated compounds and their effectiveness are viewed through a web browser or a dedicated app.

[0842] This makes the operation of the entire system more concrete and clear, and allows you to understand the detailed flow and specific operation of each processing step.

[0843] (Application example 1)

[0844] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0845] The traditional drug development process has the problem of taking a lot of time and effort to generate new molecular compounds and evaluate their effectiveness. To streamline this process, a system is needed that automates the generation of new compounds, efficacy evaluation, and integration of the results quickly and accurately. Furthermore, integration with factory automation is also necessary to operate this system effectively in pharmaceutical and chemical factories.

[0846] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0847] In this invention, the server includes a means for generating molecular compounds, a means for predicting the effectiveness of the generated molecular compounds, a means for integrating the results of the generation and prediction, a means for automating evaluation and reporting of the generated molecular compounds, a means for a user to input the molecular structure of the compound, and a means for linking with a factory automation system to enable use in chemical factories and pharmaceutical factories. This enables rapid and accurate evaluation of each generated molecular compound, and realizes efficiency and automation of the entire process.

[0848] A "molecular compound" is a chemical substance formed by bonding between molecules.

[0849] A "generating means" is a method or device for creating new molecular structures based on specific input data.

[0850] "Efficacy predictors" refer to algorithms or models for assessing the biological or pharmacological effects of generated compounds.

[0851] A "means of synthesizing results" is a method for compiling the generated molecular compounds and their predicted efficacy into a single list or report format.

[0852] The "means for automating evaluation and reporting" is a system for collecting data on generated compounds in real time, evaluating the data, and automatically presenting the results to the user.

[0853] The "means for inputting molecular structure" refers to an interface or device that allows a user to input structural information about a compound.

[0854] "Means for linking with factory automation systems to enable use in chemical and pharmaceutical factories" refers to a method or device for automating the product production and evaluation processes in a factory and linking the system with existing factory automation systems.

[0855] The present invention provides a system for producing molecular compounds and predicting their effectiveness in chemical and pharmaceutical factories. As an example, the present invention will be described using the system in a chemical factory.

[0856] System configuration and program description

[0857] 1. Initialize the model

[0858] The server first loads and initializes the generative AI model from the specified path. This generative AI model is used to generate new molecular compounds. The server manages the generative model and the efficacy prediction model.

[0859] 2. The user inputs the molecular structure

[0860] Users input the molecular structure of a compound in SMILES notation through a dedicated interface that is built into computer terminals used on-site at chemical and pharmaceutical plants.

[0861] 3. Generation of new compounds

[0862] The server passes the input SMILES notation to the generative AI model to generate a new molecular compound. Various data for the generated compound is then converted back into a molecular object.

[0863] 4. Prediction of efficacy of the resulting compounds

[0864] The server predicts the efficacy of each compound generated using a specific algorithm, and evaluates the efficacy of the compound based on its chemical structure.

[0865] 5. Synthesis and reporting of results

[0866] The server integrates the SMILES representations of all generated compounds and their predicted effectiveness. It compiles them into a list and provides immediate feedback to the user on the evaluation results. The evaluation results are also automatically compiled into a report and saved in collaboration with other systems within the factory.

[0867] Hardware and Software

[0868] Hardware used

[0869] Server (High-Performance Computing Server)

[0870] Factory terminals (user interface computers)

[0871] Factory automation systems (production equipment for chemical and pharmaceutical factories)

[0872] Software used

[0873] Generative AI model (model for generating molecular compounds)

[0874] Chemical information processing library (e.g., chemoinformatics)

[0875] Efficacy prediction algorithms (specific algorithms for predicting drug effects)

[0876] Specific examples of explanation

[0877] For example, if a user inputs the SMILES representation of ethanol ("CCO"), new molecular compounds are generated based on this SMILES representation. For each generated compound, the server predicts its effectiveness and aggregates the results in a list. The user receives a list like this:

[0878] SMILES: CCOC, Efficacy: 0.5839

[0879] SMILES: CCOCC, Efficacy: 0.7123

[0880] This allows users to quickly confirm the structure and efficacy of each compound generated. The system will significantly improve the efficiency of research and development processes in chemical and pharmaceutical factories.

[0881] Prompt Sentence Examples

[0882] "Generate new compounds based on ethanol (SMILES: "CCO") and predict their effectiveness."

[0883] In this way, this invention automates the creation of new molecular compounds and the prediction of their efficacy, greatly streamlining the research and development process. By linking with factory automation systems, it can be easily used in actual production sites, and is expected to speed up and improve the accuracy of drug efficacy evaluation.

[0884] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0885] Step 1:

[0886] The server initializes the generative model. The server loads the generative AI model from the specified path and initializes it. Specifically, the server loads the model file and instantiates it. This process makes the generative AI model ready to run. The input is the path to the model file, and the output is the initialized generative AI model.

[0887] Step 2:

[0888] A user uses a terminal to input the molecular structure of a compound in SMILES notation. This input is done through a dedicated interface. The terminal receives the SMILES notation entered by the user and sends it to the server. The input is the SMILES notation from the user, and the output is the SMILES notation received by the server.

[0889] Step 3:

[0890] The server passes the received SMILES notation to the generative AI model to generate new molecular compounds. Specifically, the server passes the input SMILES notation to the model, which generates multiple new SMILES notations. These new SMILES notations are converted back into molecular objects. The input is the SMILES notation received from the user, and the output is a list of the generated new molecular compounds.

[0891] Step 4:

[0892] The server predicts the efficacy of each generated molecular compound. This prediction is performed using a specific algorithm, which evaluates drug efficacy based on the input molecular structure information. Specifically, the server inputs the structural information of each molecular compound into the prediction algorithm and outputs a numerical value of efficacy. The input is a list of newly generated molecular compounds, and the output is a numerical value representing the efficacy of each compound.

[0893] Step 5:

[0894] The server integrates the generated SMILES representations of all compounds and their predicted efficacy. Specifically, the server pairs the SMILES representation of each compound with its efficacy value and compiles it into a list. This can be displayed or saved through an interface with the user or other systems. The input is the SMILES representation of each compound and its efficacy value, and the output is an integrated list.

[0895] Step 6:

[0896] The server generates a report from the integrated results and connects it to other automation systems in the chemical or pharmaceutical plant. The server outputs the generated report in an appropriate format and connects it to the plant's quality control system or database. The input is the integrated list, and the output is the report, along with its storage and connection.

[0897] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0898] The system of the present invention combines the basic functions of generating molecular compounds using generative models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system allows users to efficiently develop new drug compounds and consider emotional feedback on the results.

[0899] System configuration

[0900] Model initialization

[0901] The server initializes a generative AI model from a given model path, which is used to generate new molecular compounds based on the SMILES representation of the input seed compound.

[0902] Entering seed compounds

[0903] The user inputs the SMILES notation of the seed compound. For example, the SMILES notation of ethanol, "CCO," is input. A molecular object is generated based on this notation.

[0904] Compound generation

[0905] The terminal generates new compounds using a generative model. The input SMILES notation is passed to the generative model, and a specified number of new molecular compounds with SMILES notation are generated. The generated compounds are converted into molecular objects.

[0906] Prediction of efficacy

[0907] The server predicts the efficacy of each compound generated, and evaluates the efficacy of each compound numerically using a prediction algorithm.

[0908] Activating the Emotion Engine

[0909] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's emotional state based on their input and operation history. The emotion engine provides feedback based on the user's emotions and dynamically adjusts the generation and prediction process.

[0910] Results synthesis and feedback

[0911] The server integrates the generated compounds, their efficacy, and the user's emotional state, allowing the user to check the structure and efficacy of each generated compound while receiving feedback based on their own emotions.

[0912] Natural language processing explanation

[0913] The processing of the system will be specifically explained in natural language below.

[0914] 1. Initialize the model

[0915] The server loads and initializes a generative model from the specified model path. This generative model is used to generate new molecular structures based on seed compounds.

[0916] 2. Entering the species

[0917] The user inputs the SMILES notation of a seed compound (e.g., "CCO") into the terminal and generates a molecular object that will be the basis of the compound.

[0918] 3. Compound generation

[0919] The terminal passes the input SMILES notation to the generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[0920] 4. Prediction of efficacy

[0921] The server uses an algorithm to predict the efficacy of each compound, which quantifies the efficacy of each compound.

[0922] 5. Activating the Emotional Engine

[0923] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[0924] 6. Integration of results and feedback

[0925] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds, the efficacy of each compound, and emotional feedback.

[0926] Specific examples

[0927] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each compound, the server will predict its effectiveness and create a list like this:

[0928] SMILES: CCOC, Efficacy: 0.5839

[0929] SMILES: CCOCC, Efficacy: 0.7123

[0930] ...

[0931] Additionally, the emotion engine monitors the user's reactions and provides emotional feedback if the user is surprised by the generated results:

[0932] User Emotion: Surprise

[0933] Feedback: The compounds produced appear to be highly effective.

[0934] This process allows users to efficiently generate new compounds and verify their effectiveness, while receiving feedback that takes into account the user's emotions. The entire system dynamically adjusts to improve the user experience.

[0935] The processing flow will be explained below.

[0936] Step 1:

[0937] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[0938] Step 2:

[0939] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[0940] Step 3:

[0941] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[0942] Step 4:

[0943] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[0944] Step 5:

[0945] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[0946] Step 6:

[0947] The server activates an emotion engine based on the user's input history and operation patterns to analyze the user's emotional state. The emotion engine evaluates the user's emotions in real time and prepares emotion-based feedback.

[0948] Step 7:

[0949] The server integrates the molecular compounds predicted to be effective with the user's emotional state. Specifically, it compiles the SMILES representation of each generated compound, its effectiveness value, and the user's emotional state into a single list.

[0950] Step 8:

[0951] The user receives the list provided by the server, checks the generated compounds and their effectiveness, and receives feedback from the emotion engine. This allows the user to quickly evaluate the structure and effectiveness of new compounds, and also receive feedback based on their own emotions.

[0952] Example 2

[0953] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0954] Conventional molecular compound generation systems can generate new drug compounds and predict their efficacy, but they lack the ability to provide dynamic feedback based on the user's emotions. This limits the user experience and makes it difficult to obtain feedback based on the user's emotions. Furthermore, the generation and prediction processes are not fully integrated, resulting in incomplete information provided to the user.

[0955] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0956] In this invention, the server includes: a means for generating a molecular compound using a generative model; a means for predicting the effectiveness of the generated molecular compound; a means for accepting a notation of a seed compound input by a user; a prediction means including an algorithm for quantifying the effectiveness of the generated molecular compound; an emotion engine means for recognizing the emotional state of the user and providing feedback; and a means for integrating the results of the generation and prediction and generating feedback based on the user's emotions. This allows the user to receive feedback based on their own emotions while confirming the generation and effectiveness of a new molecular compound.

[0957] A "generative model" refers to an algorithm or program that generates new data based on input data.

[0958] A "molecular compound" refers to a compound made up of multiple atoms chemically bonded together.

[0959] "Efficacy" is a concept that indicates the effectiveness of a drug or compound for a specific purpose or effect.

[0960] "SMILES notation" is an abbreviation for Simplified Molecular Input Line Entry System, and refers to a method of expressing molecular structures as characters.

[0961] An "emotion engine" refers to a system or algorithm that recognizes and analyzes a user's emotional state based on their input and operation history, and provides appropriate feedback.

[0962] "Feedback" refers to information or guidance provided to a user by a system or algorithm, which changes depending on the user's behavior and emotions.

[0963] A "molecule object" is a data object that holds structural information about a chemical molecule and is used for analysis and simulation.

[0964] A "predictive algorithm" is a set of computational methods or procedures for analyzing given data and predicting future outcomes or effectiveness.

[0965] The system of the present invention combines the basic functions of generating molecular compounds using generative AI models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system is designed to enable users to efficiently develop new drug compounds and take into account emotional feedback on the results.

[0966] System configuration

[0967] The system mainly consists of three components: the server, the terminal, and the user. The operation of each component is described in detail below.

[0968] Model initialization

[0969] The server loads and initializes the generative AI model from the specified path. This generative model is used to generate new molecular compounds based on the SMILES representation of the input seed compound. For example, if the file path of the generative AI model is " / models / generative_model.pth", the server loads the model from this path and initializes the necessary parameters.

[0970] Entering seed compounds

[0971] The user uses the terminal's input form to input the SMILES representation of a seed compound. For example, the SMILES representation of ethanol is "CCO." Once this input is made, the terminal validates the input to ensure that it is in the correct format. The server then creates a molecule object based on this SMILES representation.

[0972] Compound generation

[0973] The terminal passes the user-entered SMILES notation to the generative AI model, which generates the specified number of new compounds. The generated new SMILES notation is converted into molecular objects by the server. These modified molecular objects are then analyzed for their shape and properties.

[0974] Prediction of efficacy

[0975] The server uses a specific algorithm to predict the efficacy of each compound generated, for example, by using a drug efficacy evaluation model to quantify the efficacy of the compound generated, and this quantified data is then provided to the user.

[0976] Activating the Emotion Engine

[0977] The server uses an emotion engine to analyze the user's emotional state. It recognizes the user's emotional state based on their input history and operation patterns, and generates the results as feedback. For example, it uses the "EmotionRecognitionAPI" to identify the user's emotions.

[0978] Results synthesis and feedback

[0979] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds and information about the efficacy of each compound. Feedback is also provided based on the user's emotions.

[0980] Specific examples

[0981] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. The server will generate the following list:

[0982] SMILES: CCOC, Efficacy: 0.5839

[0983] SMILES: CCOCC, Efficacy: 0.7123

[0984] ...

[0985] The emotion engine monitors the user's reactions and, for example, if the user is surprised by the generated results, provides that emotion as feedback:

[0986] User Emotion: Surprise

[0987] Feedback: The compounds produced appear to be highly effective.

[0988] This process allows users to generate new compounds and check their effectiveness while receiving feedback based on their own emotions. The entire system is dynamically adjusted to improve the user experience.

[0989] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0990] Step 1: Initialize the model

[0991] The server loads and initializes the generative AI model from the specified model path.

[0992] Input: Model file path (e.g. " / models / generative_model.pth")

[0993] Output: An initialized generative AI model

[0994] Specific operation: The server loads the model parameters based on the file path and initializes the generated AI model.

[0995] Step 2: Enter the seed compound

[0996] The user inputs the SMILES notation of the seed compound using an input form on the terminal.

[0997] Input: SMILES notation entered by the user (e.g. "CCO")

[0998] Output: Validation result of the input SMILES notation

[0999] Specific operation: The terminal validates the input SMILES notation and displays a warning if it is invalid. If it is valid, the server creates a molecule object based on the SMILES notation.

[1000] Step 3: Compound generation

[1001] The device passes the SMILES notation entered by the user to the generative AI model to generate new compounds.

[1002] Input: SMILES notation and generative AI model

[1003] Output: SMILES representation of the new compound generated

[1004] Specific operation: Based on the SMILES notation input to the generative AI model, the terminal generates a specified number of new SMILES notations. The server converts these generated SMILES notations into molecular objects.

[1005] Step 4: Predicting efficacy

[1006] The server uses a specific algorithm to predict the efficacy of each compound generated.

[1007] Input: Generated molecule object

[1008] Output: Potency score for each compound

[1009] Specific operation: The server uses a drug efficacy evaluation model to quantify and predict the efficacy of each compound generated.

[1010] Step 5: Activating the Emotion Engine

[1011] The server uses an emotion engine to analyze the user's emotional state.

[1012] Input: User input history and operation patterns

[1013] Output: User's emotional state and feedback

[1014] Specific operation: The emotion engine analyzes the user's input and operation history to recognize the current emotional state. Based on the recognized emotion, the server generates feedback.

[1015] Step 6: Consolidate and feedback results

[1016] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[1017] Input: Generated list of compounds, efficacy score for each compound, and user's emotional state

[1018] Output: Final result list and feedback to the user

[1019] Specific operation: The integrated result list and feedback information are provided to the user through the terminal, and the user decides on the next action based on this.

[1020] (Application example 2)

[1021] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1022] Conventional compound generation systems generate molecular compounds and predict their efficacy, but lack a dynamic adjustment function that takes into account the user's emotional feedback, resulting in issues with user experience and efficiency. Therefore, it is necessary to provide feedback that reflects the user's emotions about the generated compounds and adjust the generation and prediction process based on that feedback.

[1023] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating molecular compounds using a generative model, means for predicting the effectiveness of the generated molecular compounds, means for integrating the results of the generation and prediction, and means for dynamically adjusting the generation and prediction processes using an emotion engine that recognizes the user's emotional state. This makes it possible to dynamically adjust the compound generation process while providing feedback that reflects the user's emotions to the generated compounds.

[1024] A "generative model" is an algorithm or program for generating new molecular compounds based on the molecular structure of chemical substances.

[1025] A "molecular compound" is a chemical substance made up of multiple atoms bonded together.

[1026] "Efficacy" is an index used to evaluate the degree to which the produced molecular compound has the desired medicinal effect or function.

[1027] An "emotion engine" is a system or software for recognizing and analyzing a user's emotional state.

[1028] "Dynamic adjustment" means changing and optimizing processes in real time or incrementally based on user sentiment and other real-time data.

[1029] "SMILES notation" is a format for expressing molecular structures as strings of characters.

[1030] The present invention is a system that generates molecular compounds, predicts their effectiveness, and dynamically adjusts the process by taking into account the user's emotions. This system integrates a generative model, an emotion engine, and molecular generation and prediction functions. Specific embodiments for implementing the invention are described below.

[1031] System Program

[1032] The server uses a generative model to generate molecular compounds and predict their efficacy, and an emotion engine to recognize user emotions and dynamically adjust the process based on them.

[1033] Processing Description

[1034] Initializing the generative model

[1035] The server loads and initializes a generative model from a specified model path. This generative model is an algorithm that generates new molecular compounds based on the molecular structures of chemical substances.

[1036] Entering seed compounds

[1037] The user inputs the SMILES notation of a compound into the terminal and generates a molecular object that will be the basis for the compound. Specifically, the SMILES notation of ethanol, "CCO," is input.

[1038] Compound generation

[1039] The server passes the input SMILES representation to a generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[1040] Prediction of efficacy

[1041] The server predicts the efficacy of each compound generated. Using a prediction algorithm, the efficacy of each compound is evaluated as a numerical value.

[1042] Activating the Emotion Engine

[1043] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[1044] Results synthesis and feedback

[1045] The server integrates the SMILES representation of the generated compounds, the efficacy values, and the user's emotional state to finally provide an evaluation of the product and feedback to the user.

[1046] Hardware and software used

[1047] Software used:

[1048] Generative model (molecular generation algorithm)

[1049] Emotion engine (emotion recognition software)

[1050] Efficacy prediction algorithm

[1051] Hardware used:

[1052] Server (performs the generation and prediction process)

[1053] Terminal (where the user enters input)

[1054] Specific examples

[1055] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal generates a new compound. The server predicts the validity of this compound and creates a list like the following:

[1056] Compounds produced and their effectiveness:

[1057] SMILES: CCOC, Efficacy: 0.5839

[1058] SMILES: CCOCC, Efficacy: 0.7123

[1059] At the same time, the emotion engine monitors the user's reactions and provides emotional feedback if, for example, the user is surprised by the generated results.

[1060] User Emotion: Surprise

[1061] Feedback: The compound produced appears to be highly effective.

[1062] Prompt Sentence Examples

[1063] The following is a specific example of a prompt when entering information:

[1064] Generate new compounds based on the SMILES representation of medicinal compounds entered by the user. Then, predict the medicinal effects of the compounds and provide feedback based on the user's sentiment. Here is the user's input data:

[1065] SMILES: CCO (ethanol)

[1066] User operation history: ["input", "confirm"]

[1067] By following the above steps, the system can generate new molecular compounds based on the compounds entered by the user, predict their efficacy, and provide feedback that reflects the user's emotions, thereby improving the efficiency of the generation process and the user experience.

[1068] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1069] Step 1:

[1070] The server reads and initializes the generative model from a specified model path.

[1071] Specifically, the algorithm of the generative model is loaded and its environment is set up. At this stage, the generative model is ready to generate molecular structures.

[1072] The input data is a model path, and the output is an initialized generative model.

[1073] Step 2:

[1074] The user inputs the SMILES notation of the compound into the terminal.

[1075] Specifically, the user enters the SMILES representation of the compound (eg, "CCO") into a designated text field using a keyboard or other input device.

[1076] The input data is a SMILES representation, and the output is a SMILES representation of the input compound.

[1077] Step 3:

[1078] The server passes the input SMILES representation to the generative model and generates the specified number of new compounds.

[1079] Specifically, the generative model generates multiple new molecular compounds based on the input SMILES notation and outputs the SMILES notation of each compound.

[1080] The input data is the SMILES representation entered by the user, and the output is a list of new SMILES representations that are generated.

[1081] Step 4:

[1082] The server predicts the efficacy of each compound generated.

[1083] Specifically, the SMILES representation of each compound is input into a prediction algorithm, and its effectiveness is evaluated as a numerical value, which indicates the compound's efficacy.

[1084] The input data is a list of newly generated SMILES representations, and the output is a numerical value indicating the potency of each compound.

[1085] Step 5:

[1086] The server uses an emotion engine to analyze the user's emotional state.

[1087] Specifically, the emotion engine recognizes and analyzes the user's emotions based on the user's operation history and input data. The analysis results indicate the user's emotional state.

[1088] The input data is the user's operation history and input data, and the output is the analysis result of the user's emotional state.

[1089] Step 6:

[1090] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[1091] Specifically, the generated compound list, efficacy values, and the user's emotional state are provided to the user as feedback.

[1092] The input data are the SMILES representation of the generated compound, its efficacy value, and the user's emotional state, and the output is integrated feedback.

[1093] Step 7:

[1094] The user receives the feedback provided by the server.

[1095] Specifically, the user checks the feedback displayed on the device's display and makes a decision on the next step.

[1096] The input data is the consolidated feedback from the server and the output is the user's decision.

[1097] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1098] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1099] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1100] [Fourth embodiment]

[1101] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1102] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1103] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1104] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1105] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1106] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1107] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1108] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1109] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1110] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1111] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1112] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1113] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1114] The system of the present invention uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[1115] System configuration

[1116] Model initialization

[1117] The server initializes the generative AI model. The path to the model is specified, and the generative model is instantiated using this path.

[1118] Entering seed compounds

[1119] The user inputs the seed compound in SMILES notation, which is then converted into a molecular object.

[1120] Compound generation

[1121] The terminal generates new compounds using a generative AI model. First, the SMILES representation of the input seed compound is provided to the generative model, which then generates a specified number of new molecular compounds with the SMILES representation. The generated compounds are then converted back into molecular objects.

[1122] Prediction of efficacy

[1123] The server predicts the efficacy of the new molecular compounds generated. The efficacy of each compound is calculated using a specific algorithm, which basically evaluates the drug's efficacy based on the compound's structural information.

[1124] Integration of results

[1125] The server aggregates all generated molecular compounds and their predicted efficacy values, specifically, by creating a list of pairs of each compound's SMILES representation and its efficacy value.

[1126] Natural language processing explanation

[1127] The processing of the system will be explained in natural language below.

[1128] 1. Initialize the model

[1129] The server loads and initializes the generative model from the specified path. This generative model is used to generate new compounds based on the given seed compounds.

[1130] 2. Entering the species

[1131] The user inputs the SMILES notation of a seed compound, which becomes the basis for generating compounds. For example, suppose the SMILES notation of ethanol, "CCO", is input.

[1132] 3. Compound generation

[1133] The terminal passes this input SMILES notation to the generative model to generate a new molecular compound represented by SMILES notation, which is then converted back into a molecular object.

[1134] 4. Prediction of efficacy

[1135] The server has the function of predicting the efficacy of each compound generated. The efficacy of each compound is evaluated using a series of algorithms. For example, the prediction algorithm takes the compound's structure as input and outputs its efficacy as a numerical value.

[1136] 5. Synthesis of results

[1137] The server compiles the generated compounds and their predicted efficacies in a list format and presents it to the user. For example, pairs of the generated compounds' SMILES representations and their efficacies are displayed.

[1138] Specific examples

[1139] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each generated compound, the server will predict its effectiveness, resulting in a list like this:

[1140] SMILES: CCOC, Efficacy: 0.5839

[1141] SMILES: CCOCC, Efficacy: 0.7123

[1142] ...

[1143] This allows users to quickly confirm the structure and efficacy of each generated compound, making this system a powerful tool for streamlining the drug development process and accelerating the delivery of treatments.

[1144] The processing flow will be explained below.

[1145] Step 1:

[1146] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[1147] Step 2:

[1148] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[1149] Step 3:

[1150] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[1151] Step 4:

[1152] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[1153] Step 5:

[1154] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[1155] Step 6:

[1156] The server integrates the results with the molecular compounds whose efficacy is predicted, specifically by creating a list of pairs of the SMILES representation of each compound and its efficacy value.

[1157] Step 7:

[1158] The user receives the list provided by the server and checks the generated compounds and their effectiveness. For example, each entry in the list (the compound's SMILES representation and effectiveness value) is displayed on the screen. This allows the user to quickly evaluate the structure of a new compound and its effectiveness.

[1159] Example 1

[1160] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1161] In the modern drug development process, the generation and evaluation of novel active ingredients is a time-consuming and costly task. In particular, there is a need for methods to rapidly and efficiently generate a large number of candidate molecules and predict their efficacy at an early stage. Traditional methods require enormous resources, both experimentally and computationally, and new approaches are needed to solve this challenge.

[1162] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1163] In this invention, the server includes a means for loading and initializing a generative model from a specified path, a means for inputting the molecular structure of a seed compound in SMILES notation and converting it into a molecular object, a means for generating a new molecular compound in SMILES notation using the generative AI model, a means for predicting the effectiveness of the newly generated molecular compound, and a means for integrating the results of the generation and prediction, thereby enabling the rapid and efficient generation of a large number of candidate molecules and the prediction of their effectiveness.

[1164] The "specified path" is a string that indicates the directory where the generative model file is saved or the location of the file.

[1165] A "generative model" is a machine learning model for generating new molecular compounds from given input.

[1166] The "means for initialization" is a function that loads a generative model and executes a series of operations to make it usable.

[1167] A "seed compound" is an existing chemical compound from which new compounds can be made.

[1168] "SMILES notation" is a standard notation for expressing chemical structural formulas in string format.

[1169] A "molecular object" is a molecular representation generated from SMILES notation in a data format used for computational processing and structural analysis.

[1170] A "generative AI model" is a model for generating new molecular compounds using artificial intelligence technology.

[1171] The "means for predicting efficacy" is an algorithm or calculation method for calculating the pharmacological effect or other performance of the generated molecular compound.

[1172] The "means for integrating results" is a function that brings together the information on the generated compounds and the results of their effectiveness evaluation into a list or the like.

[1173] This invention relates to a system that uses generative models to generate molecular compounds and predict their efficacy, thereby streamlining the drug development process. This system is primarily composed of a server, terminals, and users.

[1174] The server first loads and initializes the generative AI model from the specified path. This generative model is often implemented using machine learning libraries such as TensorFlow or PyTorch. For example, the model path is specified as "models / generative_model.pth", and the model is instantiated based on this.

[1175] The user inputs the SMILES notation of the seed compound through the terminal. SMILES notation is a standard notation for expressing chemical structural formulas in string format, and the user starts generating a compound by inputting the SMILES notation of the compound. For example, the SMILES notation of ethanol, "CCO", is input.

[1176] The device receives the SMILES notation entered by the user and converts it into a molecular object using a library such as RDKit. The converted molecular object is then provided to a generative AI model to generate new molecular compounds. This generation process uses generative models such as the Variational Autoencoder (VAE) and Generative Adversarial Network (GAN).

[1177] The new SMILES-represented molecular compounds are then converted into molecular objects using the RDKit. For each compound, the server predicts its efficacy. This prediction uses algorithms such as the QED (Quantitative Estimation of Drug-likeness) score and molecular docking simulation. This allows the pharmacological effects and other performance of each compound to be evaluated numerically.

[1178] Finally, the server compiles a list of all the generated molecular compounds and their predicted efficacy, and presents it to the user. For example, the results are presented in a list format, with the SMILES representation of the generated compound paired with its efficacy value. As a concrete example, the compound generated based on ethanol and its predicted efficacy are shown below.

[1179] SMILES: CCOC, Efficacy: 0.5839

[1180] SMILES: CCOCC, Efficacy: 0.7123

[1181] ...

[1182] This system allows users to quickly confirm the structure and efficacy of each generated compound, contributing to the efficiency of the drug development process.

[1183] An example prompt sentence would be of the following format:

[1184] "Generate new molecular compounds based on the SMILES code 'CCO' and predict their efficacy."

[1185] This allows the generative AI model to automatically generate new molecular compounds based on the input SMILES notation and predict their effectiveness.

[1186] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1187] Step 1:

[1188] The server loads and initializes the generative AI model from the specified path. The path to the generative model file is given as input. The initialized generative model is obtained as output. Specifically, the model file is loaded using TensorFlow or PyTorch.

[1189] Step 2:

[1190] The user inputs the SMILES notation of a seed compound into the terminal. The SMILES notation of the seed compound (for example, "CCO") is given as input. The input SMILES notation is used as is as output. Specifically, the user inputs the SMILES notation through an interface such as a text box.

[1191] Step 3:

[1192] The terminal converts the input SMILES notation into a molecule object. As input, a SMILES notation is given. As output, a molecule object is obtained. Specifically, a library such as RDKit is used to convert the SMILES notation into a molecule object.

[1193] Step 4:

[1194] The terminal provides a molecular object to the generative AI model to generate new molecular compounds in SMILES notation. As input, a molecular object and a generative model are given. As output, a list of new SMILES notations is obtained. Specifically, the terminal generates new molecular compounds using a generative model (VAE or GAN).

[1195] Step 5:

[1196] The terminal converts the generated SMILES representation back into molecule objects. As input, it is given a new SMILES representation. As output, it obtains a list of molecule objects. Specifically, it uses the RDKit again to convert the SMILES representation into molecule objects.

[1197] Step 6:

[1198] The server predicts the efficacy of each generated molecular compound. As input, it receives a list of generated molecular objects. As output, it obtains an efficacy score corresponding to each molecular compound. Specifically, it calculates efficacy using algorithms such as QED score and molecular docking simulation.

[1199] Step 7:

[1200] The server integrates the generated SMILES representations of compounds with their corresponding validity scores. As input, it receives pairs of the generated SMILES representations of molecular compounds and validity scores. As output, it obtains a list of the integrated results. Specifically, it compiles the results in a list format and presents them to the user.

[1201] Step 8:

[1202] The user checks the results through a terminal. As input, a list of integrated results from the server is given. As output, the results are displayed to the user. Specifically, the generated compounds and their effectiveness are viewed through a web browser or a dedicated app.

[1203] This makes the operation of the entire system more concrete and clear, and allows you to understand the detailed flow and specific operation of each processing step.

[1204] (Application example 1)

[1205] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1206] The traditional drug development process has the problem of taking a lot of time and effort to generate new molecular compounds and evaluate their effectiveness. To streamline this process, a system is needed that automates the generation of new compounds, efficacy evaluation, and integration of the results quickly and accurately. Furthermore, integration with factory automation is also necessary to operate this system effectively in pharmaceutical and chemical factories.

[1207] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1208] In this invention, the server includes a means for generating molecular compounds, a means for predicting the effectiveness of the generated molecular compounds, a means for integrating the results of the generation and prediction, a means for automating evaluation and reporting of the generated molecular compounds, a means for a user to input the molecular structure of the compound, and a means for linking with a factory automation system to enable use in chemical factories and pharmaceutical factories. This enables rapid and accurate evaluation of each generated molecular compound, and realizes efficiency and automation of the entire process.

[1209] A "molecular compound" is a chemical substance formed by bonding between molecules.

[1210] A "generating means" is a method or device for creating new molecular structures based on specific input data.

[1211] "Efficacy predictors" refer to algorithms or models for assessing the biological or pharmacological effects of generated compounds.

[1212] A "means of synthesizing results" is a method for compiling the generated molecular compounds and their predicted efficacy into a single list or report format.

[1213] The "means for automating evaluation and reporting" is a system for collecting data on generated compounds in real time, evaluating the data, and automatically presenting the results to the user.

[1214] The "means for inputting molecular structure" refers to an interface or device that allows a user to input structural information about a compound.

[1215] "Means for linking with factory automation systems to enable use in chemical and pharmaceutical factories" refers to a method or device for automating the product production and evaluation processes in a factory and linking the system with existing factory automation systems.

[1216] The present invention provides a system for producing molecular compounds and predicting their effectiveness in chemical and pharmaceutical factories. As an example, the present invention will be described using the system in a chemical factory.

[1217] System configuration and program description

[1218] 1. Initialize the model

[1219] The server first loads and initializes the generative AI model from the specified path. This generative AI model is used to generate new molecular compounds. The server manages the generative model and the efficacy prediction model.

[1220] 2. The user inputs the molecular structure

[1221] Users input the molecular structure of a compound in SMILES notation through a dedicated interface that is built into computer terminals used on-site at chemical and pharmaceutical plants.

[1222] 3. Generation of new compounds

[1223] The server passes the input SMILES notation to the generative AI model to generate a new molecular compound. Various data for the generated compound is then converted back into a molecular object.

[1224] 4. Prediction of efficacy of the resulting compounds

[1225] The server predicts the efficacy of each compound generated using a specific algorithm, and evaluates the efficacy of the compound based on its chemical structure.

[1226] 5. Synthesis and reporting of results

[1227] The server integrates the SMILES representations of all generated compounds and their predicted effectiveness. It compiles them into a list and provides immediate feedback to the user on the evaluation results. The evaluation results are also automatically compiled into a report and saved in collaboration with other systems within the factory.

[1228] Hardware and Software

[1229] Hardware used

[1230] Server (High-Performance Computing Server)

[1231] Factory terminals (user interface computers)

[1232] Factory automation systems (production equipment for chemical and pharmaceutical factories)

[1233] Software used

[1234] Generative AI model (model for generating molecular compounds)

[1235] Chemical information processing library (e.g., chemoinformatics)

[1236] Efficacy prediction algorithms (specific algorithms for predicting drug effects)

[1237] Specific examples of explanation

[1238] For example, if a user inputs the SMILES representation of ethanol ("CCO"), new molecular compounds are generated based on this SMILES representation. For each generated compound, the server predicts its effectiveness and aggregates the results in a list. The user receives a list like this:

[1239] SMILES: CCOC, Efficacy: 0.5839

[1240] SMILES: CCOCC, Efficacy: 0.7123

[1241] This allows users to quickly confirm the structure and efficacy of each compound generated. The system will significantly improve the efficiency of research and development processes in chemical and pharmaceutical factories.

[1242] Prompt Sentence Examples

[1243] "Generate new compounds based on ethanol (SMILES: "CCO") and predict their effectiveness."

[1244] In this way, this invention automates the creation of new molecular compounds and the prediction of their efficacy, greatly streamlining the research and development process. By linking with factory automation systems, it can be easily used in actual production sites, and is expected to speed up and improve the accuracy of drug efficacy evaluation.

[1245] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1246] Step 1:

[1247] The server initializes the generative model. The server loads the generative AI model from the specified path and initializes it. Specifically, the server loads the model file and instantiates it. This process makes the generative AI model ready to run. The input is the path to the model file, and the output is the initialized generative AI model.

[1248] Step 2:

[1249] A user uses a terminal to input the molecular structure of a compound in SMILES notation. This input is done through a dedicated interface. The terminal receives the SMILES notation entered by the user and sends it to the server. The input is the SMILES notation from the user, and the output is the SMILES notation received by the server.

[1250] Step 3:

[1251] The server passes the received SMILES notation to the generative AI model to generate new molecular compounds. Specifically, the server passes the input SMILES notation to the model, which generates multiple new SMILES notations. These new SMILES notations are converted back into molecular objects. The input is the SMILES notation received from the user, and the output is a list of the generated new molecular compounds.

[1252] Step 4:

[1253] The server predicts the efficacy of each generated molecular compound. This prediction is performed using a specific algorithm, which evaluates drug efficacy based on the input molecular structure information. Specifically, the server inputs the structural information of each molecular compound into the prediction algorithm and outputs a numerical value of efficacy. The input is a list of newly generated molecular compounds, and the output is a numerical value representing the efficacy of each compound.

[1254] Step 5:

[1255] The server integrates the generated SMILES representations of all compounds and their predicted efficacy. Specifically, the server pairs the SMILES representation of each compound with its efficacy value and compiles it into a list. This can be displayed or saved through an interface with the user or other systems. The input is the SMILES representation of each compound and its efficacy value, and the output is an integrated list.

[1256] Step 6:

[1257] The server generates a report from the integrated results and connects it to other automation systems in the chemical or pharmaceutical plant. The server outputs the generated report in an appropriate format and connects it to the plant's quality control system or database. The input is the integrated list, and the output is the report, along with its storage and connection.

[1258] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1259] The system of the present invention combines the basic functions of generating molecular compounds using generative models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system allows users to efficiently develop new drug compounds and consider emotional feedback on the results.

[1260] System configuration

[1261] Model initialization

[1262] The server initializes a generative AI model from a given model path, which is used to generate new molecular compounds based on the SMILES representation of the input seed compound.

[1263] Entering seed compounds

[1264] The user inputs the SMILES notation of the seed compound. For example, the SMILES notation of ethanol, "CCO," is input. A molecular object is generated based on this notation.

[1265] Compound generation

[1266] The terminal generates new compounds using a generative model. The input SMILES notation is passed to the generative model, and a specified number of new molecular compounds with SMILES notation are generated. The generated compounds are converted into molecular objects.

[1267] Prediction of efficacy

[1268] The server predicts the efficacy of each compound generated, and evaluates the efficacy of each compound numerically using a prediction algorithm.

[1269] Activating the Emotion Engine

[1270] The server uses an emotion engine to recognize the user's emotions. It analyzes the user's emotional state based on their input and operation history. The emotion engine provides feedback based on the user's emotions and dynamically adjusts the generation and prediction process.

[1271] Results synthesis and feedback

[1272] The server integrates the generated compounds, their efficacy, and the user's emotional state, allowing the user to check the structure and efficacy of each generated compound while receiving feedback based on their own emotions.

[1273] Natural language processing explanation

[1274] The processing of the system will be specifically explained in natural language below.

[1275] 1. Initialize the model

[1276] The server loads and initializes a generative model from the specified model path. This generative model is used to generate new molecular structures based on seed compounds.

[1277] 2. Entering the species

[1278] The user inputs the SMILES notation of a seed compound (e.g., "CCO") into the terminal and generates a molecular object that will be the basis of the compound.

[1279] 3. Compound generation

[1280] The terminal passes the input SMILES notation to the generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[1281] 4. Prediction of efficacy

[1282] The server uses an algorithm to predict the efficacy of each compound, which quantifies the efficacy of each compound.

[1283] 5. Activating the Emotional Engine

[1284] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[1285] 6. Integration of results and feedback

[1286] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds, the efficacy of each compound, and emotional feedback.

[1287] Specific examples

[1288] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. For each compound, the server will predict its effectiveness and create a list like this:

[1289] SMILES: CCOC, Efficacy: 0.5839

[1290] SMILES: CCOCC, Efficacy: 0.7123

[1291] ...

[1292] Additionally, the emotion engine monitors the user's reactions and provides emotional feedback if the user is surprised by the generated results:

[1293] User Emotion: Surprise

[1294] Feedback: The compounds produced appear to be highly effective.

[1295] This process allows users to efficiently generate new compounds and verify their effectiveness, while receiving feedback that takes into account the user's emotions. The entire system dynamically adjusts to improve the user experience.

[1296] The processing flow will be explained below.

[1297] Step 1:

[1298] The server receives the given model path and initializes the generative AI model. Specifically, it creates an instance of the generative model and prepares it for use in subsequent processing.

[1299] Step 2:

[1300] The user inputs a SMILES representation of a seed compound, which is a textual representation of the chemical structure, e.g., "CCO." The input SMILES representation is converted into a molecular object of the compound.

[1301] Step 3:

[1302] The terminal inputs the SMILES representation of the seed compound input in step 2 into the generative AI model. It then invokes the generate method of the generative model to generate a specified number (e.g., 100) of new molecular compounds with SMILES representation.

[1303] Step 4:

[1304] The terminal converts the generated new SMILES notations into molecule objects by using the Chem.MolFromSmiles method and storing them as a list.

[1305] Step 5:

[1306] The server predicts the efficacy of each generated molecular compound. Here, the generated molecular object is passed to the predict_efficacy method to calculate the efficacy value. The algorithm in this step quantifies the efficacy of the molecular compound based on its structural information.

[1307] Step 6:

[1308] The server activates an emotion engine based on the user's input history and operation patterns to analyze the user's emotional state. The emotion engine evaluates the user's emotions in real time and prepares emotion-based feedback.

[1309] Step 7:

[1310] The server integrates the molecular compounds predicted to be effective with the user's emotional state. Specifically, it compiles the SMILES representation of each generated compound, its effectiveness value, and the user's emotional state into a single list.

[1311] Step 8:

[1312] The user receives the list provided by the server, checks the generated compounds and their effectiveness, and receives feedback from the emotion engine. This allows the user to quickly evaluate the structure and effectiveness of new compounds, and also receive feedback based on their own emotions.

[1313] Example 2

[1314] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1315] Conventional molecular compound generation systems can generate new drug compounds and predict their efficacy, but they lack the ability to provide dynamic feedback based on the user's emotions. This limits the user experience and makes it difficult to obtain feedback based on the user's emotions. Furthermore, the generation and prediction processes are not fully integrated, resulting in incomplete information provided to the user.

[1316] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1317] In this invention, the server includes: a means for generating a molecular compound using a generative model; a means for predicting the effectiveness of the generated molecular compound; a means for accepting a notation of a seed compound input by a user; a prediction means including an algorithm for quantifying the effectiveness of the generated molecular compound; an emotion engine means for recognizing the emotional state of the user and providing feedback; and a means for integrating the results of the generation and prediction and generating feedback based on the user's emotions. This allows the user to receive feedback based on their own emotions while confirming the generation and effectiveness of a new molecular compound.

[1318] A "generative model" refers to an algorithm or program that generates new data based on input data.

[1319] A "molecular compound" refers to a compound made up of multiple atoms chemically bonded together.

[1320] "Efficacy" is a concept that indicates the effectiveness of a drug or compound for a specific purpose or effect.

[1321] "SMILES notation" is an abbreviation for Simplified Molecular Input Line Entry System, and refers to a method of expressing molecular structures as characters.

[1322] An "emotion engine" refers to a system or algorithm that recognizes and analyzes a user's emotional state based on their input and operation history, and provides appropriate feedback.

[1323] "Feedback" refers to information or guidance provided to a user by a system or algorithm, which changes depending on the user's behavior and emotions.

[1324] A "molecule object" is a data object that holds structural information about a chemical molecule and is used for analysis and simulation.

[1325] A "predictive algorithm" is a set of computational methods or procedures for analyzing given data and predicting future outcomes or effectiveness.

[1326] The system of the present invention combines the basic functions of generating molecular compounds using generative AI models and predicting their efficacy with an emotion engine that recognizes the user's emotions. This system is designed to enable users to efficiently develop new drug compounds and take into account emotional feedback on the results.

[1327] System configuration

[1328] The system mainly consists of three components: the server, the terminal, and the user. The operation of each component is described in detail below.

[1329] Model initialization

[1330] The server loads and initializes the generative AI model from the specified path. This generative model is used to generate new molecular compounds based on the SMILES representation of the input seed compound. For example, if the file path of the generative AI model is " / models / generative_model.pth", the server loads the model from this path and initializes the necessary parameters.

[1331] Entering seed compounds

[1332] The user uses the terminal's input form to input the SMILES representation of a seed compound. For example, the SMILES representation of ethanol is "CCO." Once this input is made, the terminal validates the input to ensure that it is in the correct format. The server then creates a molecule object based on this SMILES representation.

[1333] Compound generation

[1334] The terminal passes the user-entered SMILES notation to the generative AI model, which generates the specified number of new compounds. The generated new SMILES notation is converted into molecular objects by the server. These modified molecular objects are then analyzed for their shape and properties.

[1335] Prediction of efficacy

[1336] The server uses a specific algorithm to predict the efficacy of each compound generated, for example, by using a drug efficacy evaluation model to quantify the efficacy of the compound generated, and this quantified data is then provided to the user.

[1337] Activating the Emotion Engine

[1338] The server uses an emotion engine to analyze the user's emotional state. It recognizes the user's emotional state based on their input history and operation patterns, and generates the results as feedback. For example, it uses the "EmotionRecognitionAPI" to identify the user's emotions.

[1339] Results synthesis and feedback

[1340] The server integrates the generated compounds' SMILES representations, efficacy values, and the user's emotional state. Finally, the user receives a list of generated compounds and information about the efficacy of each compound. Feedback is also provided based on the user's emotions.

[1341] Specific examples

[1342] For example, if the user inputs ethanol (SMILES: "CCO"), the terminal will generate new compounds based on the ethanol molecule. The server will generate the following list:

[1343] SMILES: CCOC, Efficacy: 0.5839

[1344] SMILES: CCOCC, Efficacy: 0.7123

[1345] ...

[1346] The emotion engine monitors the user's reactions and, for example, if the user is surprised by the generated results, provides that emotion as feedback:

[1347] User Emotion: Surprise

[1348] Feedback: The compounds produced appear to be highly effective.

[1349] This process allows users to generate new compounds and check their effectiveness while receiving feedback based on their own emotions. The entire system is dynamically adjusted to improve the user experience.

[1350] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1351] Step 1: Initialize the model

[1352] The server loads and initializes the generative AI model from the specified model path.

[1353] Input: Model file path (e.g. " / models / generative_model.pth")

[1354] Output: An initialized generative AI model

[1355] Specific operation: The server loads the model parameters based on the file path and initializes the generated AI model.

[1356] Step 2: Enter the seed compound

[1357] The user inputs the SMILES notation of the seed compound using an input form on the terminal.

[1358] Input: SMILES notation entered by the user (e.g. "CCO")

[1359] Output: Validation result of the input SMILES notation

[1360] Specific operation: The terminal validates the input SMILES notation and displays a warning if it is invalid. If it is valid, the server creates a molecule object based on the SMILES notation.

[1361] Step 3: Compound generation

[1362] The device passes the SMILES notation entered by the user to the generative AI model to generate new compounds.

[1363] Input: SMILES notation and generative AI model

[1364] Output: SMILES representation of the new compound generated

[1365] Specific operation: Based on the SMILES notation input to the generative AI model, the terminal generates a specified number of new SMILES notations. The server converts these generated SMILES notations into molecular objects.

[1366] Step 4: Predicting efficacy

[1367] The server uses a specific algorithm to predict the efficacy of each compound generated.

[1368] Input: Generated molecule object

[1369] Output: Potency score for each compound

[1370] Specific operation: The server uses a drug efficacy evaluation model to quantify and predict the efficacy of each compound generated.

[1371] Step 5: Activating the Emotion Engine

[1372] The server uses an emotion engine to analyze the user's emotional state.

[1373] Input: User input history and operation patterns

[1374] Output: User's emotional state and feedback

[1375] Specific operation: The emotion engine analyzes the user's input and operation history to recognize the current emotional state. Based on the recognized emotion, the server generates feedback.

[1376] Step 6: Consolidate and feedback results

[1377] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[1378] Input: Generated list of compounds, efficacy score for each compound, and user's emotional state

[1379] Output: Final result list and feedback to the user

[1380] Specific operation: The integrated result list and feedback information are provided to the user through the terminal, and the user decides on the next action based on this.

[1381] (Application example 2)

[1382] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1383] Conventional compound generation systems generate molecular compounds and predict their efficacy, but lack a dynamic adjustment function that takes into account the user's emotional feedback, resulting in issues with user experience and efficiency. Therefore, it is necessary to provide feedback that reflects the user's emotions about the generated compounds and adjust the generation and prediction process based on that feedback.

[1384] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for generating molecular compounds using a generative model, means for predicting the effectiveness of the generated molecular compounds, means for integrating the results of the generation and prediction, and means for dynamically adjusting the generation and prediction processes using an emotion engine that recognizes the user's emotional state. This makes it possible to dynamically adjust the compound generation process while providing feedback that reflects the user's emotions to the generated compounds.

[1385] A "generative model" is an algorithm or program for generating new molecular compounds based on the molecular structure of chemical substances.

[1386] A "molecular compound" is a chemical substance made up of multiple atoms bonded together.

[1387] "Efficacy" is an index used to evaluate the degree to which the produced molecular compound has the desired medicinal effect or function.

[1388] An "emotion engine" is a system or software for recognizing and analyzing a user's emotional state.

[1389] "Dynamic adjustment" means changing and optimizing processes in real time or incrementally based on user sentiment and other real-time data.

[1390] "SMILES notation" is a format for expressing molecular structures as strings of characters.

[1391] The present invention is a system that generates molecular compounds, predicts their effectiveness, and dynamically adjusts the process by taking into account the user's emotions. This system integrates a generative model, an emotion engine, and molecular generation and prediction functions. Specific embodiments for implementing the invention are described below.

[1392] System Program

[1393] The server uses a generative model to generate molecular compounds and predict their efficacy, and an emotion engine to recognize user emotions and dynamically adjust the process based on them.

[1394] Processing Description

[1395] Initializing the generative model

[1396] The server loads and initializes a generative model from a specified model path. This generative model is an algorithm that generates new molecular compounds based on the molecular structures of chemical substances.

[1397] Entering seed compounds

[1398] The user inputs the SMILES notation of a compound into the terminal and generates a molecular object that will be the basis for the compound. Specifically, the SMILES notation of ethanol, "CCO," is input.

[1399] Compound generation

[1400] The server passes the input SMILES representation to a generative model to generate the specified number of new compounds, which are then converted back into molecular objects.

[1401] Prediction of efficacy

[1402] The server predicts the efficacy of each compound generated. Using a prediction algorithm, the efficacy of each compound is evaluated as a numerical value.

[1403] Activating the Emotion Engine

[1404] The server uses an emotion engine to analyze the user's emotional state, recognize emotions based on the user's input history and operation patterns, and provide feedback based on the generated and predicted emotions.

[1405] Results synthesis and feedback

[1406] The server integrates the SMILES representation of the generated compounds, the efficacy values, and the user's emotional state to finally provide an evaluation of the product and feedback to the user.

[1407] Hardware and software used

[1408] Software used:

[1409] Generative model (molecular generation algorithm)

[1410] Emotion engine (emotion recognition software)

[1411] Efficacy prediction algorithm

[1412] Hardware used:

[1413] Server (performs the generation and prediction process)

[1414] Terminal (where the user enters input)

[1415] Specific examples

[1416] For example, if a user inputs ethanol (SMILES: "CCO"), the terminal generates a new compound. The server predicts the validity of this compound and creates a list like the following:

[1417] Compounds produced and their effectiveness:

[1418] SMILES: CCOC, Efficacy: 0.5839

[1419] SMILES: CCOCC, Efficacy: 0.7123

[1420] At the same time, the emotion engine monitors the user's reactions and provides emotional feedback if, for example, the user is surprised by the generated results.

[1421] User Emotion: Surprise

[1422] Feedback: The compound produced appears to be highly effective.

[1423] Prompt Sentence Examples

[1424] The following is a specific example of a prompt when entering information:

[1425] Generate new compounds based on the SMILES representation of medicinal compounds entered by the user. Then, predict the medicinal effects of the compounds and provide feedback based on the user's sentiment. Here is the user's input data:

[1426] SMILES: CCO (ethanol)

[1427] User operation history: ["input", "confirm"]

[1428] By following the above steps, the system can generate new molecular compounds based on the compounds entered by the user, predict their efficacy, and provide feedback that reflects the user's emotions, thereby improving the efficiency of the generation process and the user experience.

[1429] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1430] Step 1:

[1431] The server reads and initializes the generative model from a specified model path.

[1432] Specifically, the algorithm of the generative model is loaded and its environment is set up. At this stage, the generative model is ready to generate molecular structures.

[1433] The input data is a model path, and the output is an initialized generative model.

[1434] Step 2:

[1435] The user inputs the SMILES notation of the compound into the terminal.

[1436] Specifically, the user enters the SMILES representation of the compound (eg, "CCO") into a designated text field using a keyboard or other input device.

[1437] The input data is a SMILES representation, and the output is a SMILES representation of the input compound.

[1438] Step 3:

[1439] The server passes the input SMILES representation to the generative model and generates the specified number of new compounds.

[1440] Specifically, the generative model generates multiple new molecular compounds based on the input SMILES notation and outputs the SMILES notation of each compound.

[1441] The input data is the SMILES representation entered by the user, and the output is a list of new SMILES representations that are generated.

[1442] Step 4:

[1443] The server predicts the efficacy of each compound generated.

[1444] Specifically, the SMILES representation of each compound is input into a prediction algorithm, and its effectiveness is evaluated as a numerical value, which indicates the compound's efficacy.

[1445] The input data is a list of newly generated SMILES representations, and the output is a numerical value indicating the potency of each compound.

[1446] Step 5:

[1447] The server uses an emotion engine to analyze the user's emotional state.

[1448] Specifically, the emotion engine recognizes and analyzes the user's emotions based on the user's operation history and input data. The analysis results indicate the user's emotional state.

[1449] The input data is the user's operation history and input data, and the output is the analysis result of the user's emotional state.

[1450] Step 6:

[1451] The server integrates the generated SMILES representation of the compound, the efficacy value, and the user's emotional state.

[1452] Specifically, the generated compound list, efficacy values, and the user's emotional state are provided to the user as feedback.

[1453] The input data are the SMILES representation of the generated compound, its efficacy value, and the user's emotional state, and the output is integrated feedback.

[1454] Step 7:

[1455] The user receives the feedback provided by the server.

[1456] Specifically, the user checks the feedback displayed on the device's display and makes a decision on the next step.

[1457] The input data is the consolidated feedback from the server and the output is the user's decision.

[1458] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1459] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1460] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1461] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1462] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1463] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1464] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1465] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1466] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1467] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1468] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1469] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1470] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1471] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1472] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1473] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1474] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1475] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1476] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1477] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1478] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1479] The following is further disclosed regarding the above embodiment.

[1480] (Claim 1)

[1481] A means for generating molecular compounds using a generative model;

[1482] a means for predicting the efficacy of the molecular compounds produced;

[1483] means for integrating the results of said generation and prediction;

[1484] A system including:

[1485] (Claim 2)

[1486] 2. The system according to claim 1, wherein the generative model converts the molecular structure of an input compound into a SMILES representation and generates a new molecular compound based on the SMILES representation.

[1487] (Claim 3)

[1488] 2. The system of claim 1, wherein the means for predicting efficacy comprises an algorithm for calculating drug efficacy of the generated molecular compound.

[1489] "Example 1"

[1490] (Claim 1)

[1491] A means to load and initialize a generative model from a specified path;

[1492] A means to input the molecular structure of a seed compound in SMILES notation and convert it into a molecular object,

[1493] a means for generating new SMILES molecular compounds using a generative AI model;

[1494] a means of predicting the efficacy of newly generated molecular compounds;

[1495] means for integrating the results of said generation and prediction;

[1496] A system including:

[1497] (Claim 2)

[1498] 2. The system according to claim 1, wherein the generative model converts the molecular structure of an input compound into a SMILES representation and generates a new molecular compound based on the SMILES representation.

[1499] (Claim 3)

[1500] 2. The system of claim 1, wherein the means for predicting efficacy comprises an algorithm for calculating drug efficacy of the generated molecular compound.

[1501] "Application Example 1"

[1502] (Claim 1)

[1503] a means for producing a molecular compound;

[1504] a means for predicting the efficacy of the molecular compounds produced;

[1505] means for integrating the results of said generation and prediction;

[1506] a means for automating the evaluation and reporting of the generated molecular compounds;

[1507] A means for a user to input a molecular structure of a compound;

[1508] A means of linking with factory automation systems to enable use in chemical and pharmaceutical factories;

[1509] A system including:

[1510] (Claim 2)

[1511] 2. The system according to claim 1, wherein the generative model converts the molecular structure of an input compound into a molecular representation and generates a new molecular compound based on the molecular representation.

[1512] (Claim 3)

[1513] 2. The system of claim 1, wherein the means for predicting efficacy comprises an algorithm for calculating drug efficacy of the generated molecular compound.

[1514] "Example 2: Combining Emotion Engines"

[1515] (Claim 1)

[1516] A means for generating molecular compounds using a generative model;

[1517] a means for predicting the efficacy of the molecular compounds produced;

[1518] means for accepting a notation of a seed compound input by a user;

[1519] a predictive means including an algorithm for quantifying the effectiveness of the generated molecular compound;

[1520] emotion engine means for recognizing the user's emotional state and providing feedback;

[1521] means for integrating the results of the generation and prediction to generate feedback based on the user's emotions;

[1522] A system including:

[1523] (Claim 2)

[1524] 2. The system according to claim 1, wherein the generative model converts the molecular notation of an input compound into a specific notation format and generates a new molecular compound based on the notation format.

[1525] (Claim 3)

[1526] The system of claim 1, wherein the means for predicting efficacy includes an algorithm for evaluating the drug efficacy of the generated molecular compound.

[1527] "Application example 2 when combining emotion engines"

[1528] (Claim 1)

[1529] A means for generating molecular compounds using a generative model;

[1530] a means for predicting the efficacy of the molecular compounds produced;

[1531] means for integrating the results of said generation and prediction;

[1532] means for dynamically adjusting said generation and prediction process using an emotion engine that recognizes the emotional state of a user;

[1533] A system including:

[1534] (Claim 2)

[1535] 2. The system according to claim 1, wherein the generative model converts the molecular structure of an input compound into a SMILES representation and generates a new molecular compound based on the SMILES representation.

[1536] (Claim 3)

[1537] 2. The system of claim 1, wherein the means for predicting efficacy comprises an algorithm for calculating drug efficacy of the generated molecular compound. [Explanation of symbols]

[1538] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means for generating molecular compounds using a generative model; a means for predicting the efficacy of the molecular compounds produced; means for integrating the results of said generation and prediction; A system including:

2. 2. The system according to claim 1, wherein the generative model converts the molecular structure of an input compound into a SMILES representation and generates a new molecular compound based on the SMILES representation.

3. The system of claim 1 , wherein the means for predicting efficacy comprises an algorithm for calculating drug efficacy of the generated molecular compound.

Images