Information processing apparatus, method for operating the same, and program for operating the same

Abstract

To provide an information processing apparatus, a method for operating the information processing apparatus, and a program for operating the information processing apparatus that can contribute to verification of the validity of prediction results regarding characteristics of candidate substances encapsulated in vesicles.SOLUTION: A request reception unit receives a prediction request and accepts structural information of a candidate drug. A derivation unit derives feature quantities of the candidate drug from the structural information. A prediction unit predicts encapsulation capability of the candidate drug into a liposome based on contribution feature quantities of the candidate drug. A screen distribution control unit presents, in a comparable manner, the prediction result of the encapsulation capability of the candidate drug into the liposome and known characteristics of a reference drug.SELECTED DRAWING: Figure 15

Description

[Technical Field]

[0001] The technology of the present disclosure relates to an information processing device, an operating method for an information processing device, and an operating program for an information processing device.

[0002] Recently, with the main objectives of enhancing drug efficacy and reducing side effects, active research has been conducted on drug delivery system (DDS) technology, which efficiently delivers drugs to affected areas using pharmaceutical preparations in which drugs are encapsulated in vesicles such as liposomes. For example, Non-Patent Document 1 describes a technology that predicts the encapsulation ability of drugs into vesicles (whether a drug is likely to be encapsulated in vesicles) using machine learning models based on algorithms such as decision trees, k-Nearest Neighbor (k-NN) algorithms, and Support Vector Regression (SVR). [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Ahuva Cern. et al. “Quantitative structure - property relationship modeling of remote liposome loading of drugs” Journal of Controlled Release June 2012, Volume160(Issue2) p.147-157 Summary of the Invention [Problem to be solved by the invention]

[0004] In DDS technology, it is important to know the drug characteristics, such as the drug encapsulation ability of the vesicles or the drug release ability from the vesicles (whether the drug is easily released from the vesicles), before starting the research, development, and / or manufacturing of pharmaceutical formulations (hereinafter collectively referred to as "practical use"). This is because it can discourage researchers and / or manufacturers (hereinafter collectively referred to as "operators") from using drugs that are unlikely to be effective with DDS, such as drugs with relatively low encapsulation ability or relatively high release ability, thereby reducing unnecessary costs. According to the technology described in Non-Patent Document 1, the predicted drug encapsulation ability of the vesicles can be known before the practical use of pharmaceutical formulations.

[0005] However, Non-Patent Document 1 does not describe how to display the prediction results of drug encapsulation in vesicles. Therefore, it is difficult for operators to verify the validity of the prediction results using the technology described in Non-Patent Document 1. In particular, operators involved in the practical work of pharmaceutical formulations have practical knowledge of general formulations such as tablets, but have less practical knowledge of formulations using vesicles than vesicle experts. In this respect, they are unable to empirically predict the properties of drugs entangled in vesicles, making it even more difficult to verify the validity of the prediction results.

[0006] One embodiment of the technology disclosed herein provides an information processing device, an operating method of the information processing device, and an operating program of the information processing device that can contribute to verifying the validity of prediction results of the properties of candidate substances encapsulated in vesicles. [Means for solving the problem]

[0007] The information processing device of the present disclosure includes a processor that receives structural information of a candidate substance to be encapsulated in a vesicle, derives features of the candidate substance from the structural information, predicts the properties of the candidate substance from the features of the candidate substance, and presents the predicted properties of the candidate substance in a manner that allows comparison with the known properties of a reference substance.

[0008] The processor preferably displays, in the feature space, a plot corresponding to the feature of the candidate substance, the plot having a display form according to the prediction result.

[0009] The processor preferably displays, in the feature space, regions corresponding to a plurality of feature quantities of a plurality of reference substances and having a display form according to known properties.

[0010] The processor preferably displays, in the feature space, a plurality of plots corresponding to a plurality of feature quantities of a plurality of reference substances, the plots having a display form according to known properties.

[0011] It is preferable that, upon receiving structural information of a new candidate substance and deriving features of the new candidate substance, the processor additionally displays plots corresponding to the features of the new candidate substance in the feature space in addition to the plots corresponding to the features of the candidate substances for which structural information has been received up to that point.

[0012] Preferably, the processor displays a plurality of plots corresponding to a plurality of candidate substances in a manner that distinguishes each of the plurality of candidate substances.

[0013] The processor preferably switches between displaying and hiding the plots corresponding to the candidate substances in response to an instruction from the operator.

[0014] The display form is preferably a difference in color and / or pattern.

[0015] The processor preferably presents coordinates of the plot corresponding to the feature quantities of the candidate substances in response to an instruction from the operator.

[0016] The feature space is preferably a three-dimensional space.

[0017] The processor preferably scales and / or rotates the feature space in response to an operation instruction from an operator.

[0018] The processor preferably presents chemical structural formulas of candidate substances in response to an instruction from an operator.

[0019] The feature amount is preferably made up of a plurality of types of elements, and the processor preferably predicts the properties of the candidate substance from contributing feature amounts obtained by selecting elements that contribute to the prediction of the properties from among the plurality of types of elements.

[0020] Preferably, the processor inputs the feature quantities of the candidate substances into a machine learning model and causes the machine learning model to output a prediction result.

[0021] It is preferable that multiple types of machine learning models are prepared according to the type of vesicle, and the processor selects and uses a machine learning model according to the type of vesicle.

[0022] The feature preferably includes molecular descriptors as elements.

[0023] Preferably, the properties include at least one of the ability to encapsulate the candidate substance or reference substance in the vesicle and the ability to release the candidate substance or reference substance from the vesicle.

[0024] The vesicles are preferably any of liposomes, lipid nanoparticles, and micelles.

[0025] The method of operating the information processing device disclosed herein includes receiving structural information of a candidate substance to be encapsulated in a vesicle, deriving features of the candidate substance from the structural information, predicting the properties of the candidate substance from the features of the candidate substance, and presenting the predicted properties of the candidate substance in a manner that allows comparison with the known properties of a reference substance.

[0026] The operating program of the information processing device disclosed herein causes a computer to perform processes including receiving structural information of a candidate substance to be encapsulated in a vesicle, deriving features of the candidate substance from the structural information, predicting the properties of the candidate substance from the features of the candidate substance, and presenting the predicted properties of the candidate substance in a manner that allows comparison with the known properties of a reference substance. [Effects of the Invention]

[0027] According to the technology disclosed herein, it is possible to provide an information processing device, an operating method for an information processing device, and an operating program for an information processing device that can contribute to verifying the validity of prediction results for the properties of candidate substances encapsulated in vesicles. [Brief explanation of the drawings]

[0028] [Figure 1] FIG. 1 illustrates an information processing system. [Figure 2] FIG. 1 shows pharmaceutical formulations and liposomes. [Figure 3] FIG. 10 is a diagram showing structural information. [Figure 4] FIG. 2 is a block diagram showing a computer that constitutes an information processing device and an operator terminal. [Figure 5] FIG. 2 is a block diagram showing a processing unit of a CPU of the information processing device. [Figure 6] FIG. 10 is a diagram illustrating feature amounts. [Figure 7] FIG. 10 is a diagram illustrating the processing of a prediction unit. [Figure 8] FIG. 10 is a diagram illustrating processing in the learning phase of a predictive model. [Figure 9] FIG. 10 is a diagram illustrating reference information. [Figure 10] FIG. 10 is a diagram illustrating processing by a screen distribution control unit. [Figure 11] FIG. 10 is a diagram illustrating processing by a screen distribution control unit. [Figure 12] FIG. 2 is a block diagram showing a processing unit of a CPU of an operator terminal. [Figure 13] FIG. 10 is a diagram showing a structural information input screen. [Figure 14] FIG. 10 is a diagram showing a first prediction result display screen. [Figure 15] FIG. 10 is a diagram showing a second prediction result display screen. [Figure 16] FIG. 10 is a diagram showing how new structural information is input on the structural information input screen. [Figure 17]FIG. 10 is a diagram showing a first prediction result display screen when input of new structural information is accepted. [Figure 18] FIG. 10 is a diagram showing a second prediction result display screen when input of new structural information is accepted. [Figure 19] FIG. 10 is a diagram showing a second prediction result display screen in which the plot corresponding to the candidate drug with identification number 2 is not displayed. [Figure 20] FIG. 10 is a diagram showing a second prediction result display screen on which plot coordinates are displayed. [Figure 21] FIG. 10 is a diagram showing a second prediction result display screen displaying the chemical structural formula of the candidate drug with identification number 1. [Figure 22] FIG. 10 is a diagram showing a second prediction result display screen in which the display feature space is enlarged. [Figure 23] FIG. 10 is a diagram showing a second prediction result display screen in which the display feature space is rotated. [Figure 24] 10 is a flowchart showing a processing procedure of the information processing device. [Figure 25] FIG. 10 is a diagram showing an example of a plurality of plots corresponding to a plurality of feature quantities of a plurality of reference drugs, the plots having display forms according to known properties, displayed in a display feature quantity space. [Figure 26] 10A and 10B are diagrams showing examples in which the display form is different in pattern; [Figure 27] FIG. 10 is a diagram showing a plurality of types of prediction models according to the type of liposome. [Figure 28] FIG. 10 is a diagram showing how a prediction model is selected and used depending on the type of liposome. [Figure 29] FIG. 1 shows an example of predicting the release of a candidate drug from a liposome. [Figure 30] FIG. 1 shows pharmaceutical formulations and lipid nanoparticles. [Figure 31] FIG. 1 illustrates pharmaceutical formulations and micelles. DETAILED DESCRIPTION OF THE INVENTION

[0029] As shown in FIG. 1 as an example, an information processing system 10 processes information about a candidate drug 11C and includes an information processing device 12 and an operator terminal 13. The information processing device 12 and the operator terminal 13 are connected via a network 14. The operator terminal 13 is installed in a pharmaceutical company that performs research, development, and / or manufacturing of a pharmaceutical formulation 20 (see FIG. 2) or in an organization that is contracted by a pharmaceutical company to perform the pharmaceutical formulation 20, i.e., a contract research organization (CRO). The operator terminal 13 is operated by an operator OP who performs the pharmaceutical formulation 20 at the pharmaceutical company or the contract research organization (hereinafter collectively referred to as a pharmaceutical facility). The network 14 is, for example, a wide area network (WAN) such as the Internet or a public communication network. Note that, although only one operator terminal 13 is connected to the information processing device 12 in FIG. 1, multiple operator terminals 13 at multiple pharmaceutical facilities are actually connected to the information processing device 12.

[0030] As an example, as shown in FIG. 2, a pharmaceutical preparation 20 includes liposomes 21 encapsulating a drug 11. That is, the pharmaceutical preparation 20 is a so-called liposome preparation. The drug 11 is specifically an anticancer drug, an antifungal agent, an analgesic, or the like. The liposomes 21 are nano-sized particles composed of at least one lipid bilayer. The candidate drug 11C is a drug 11 prepared by an operator OP as a candidate to be encapsulated in the liposomes 21. Therefore, the candidate drug 11C is a drug 11 before being exposed to the actual pharmaceutical preparation 20, and is a drug 11 whose properties associated with the liposomes 21 are unknown. The candidate drug 11C is an example of a "candidate substance" according to the technology of the present disclosure. Furthermore, the liposomes 21 are an example of a "vesicle" according to the technology of the present disclosure.

[0031] Returning to FIG. 1 , the operator terminal 13 transmits a prediction request 15 to the information processing device 12. The prediction request 15 is a request for the information processing device 12 to predict the properties of the candidate drug 11C. The prediction request 15 includes structural information 16. The structural information 16 is information representing the structure of the candidate drug 11C. Although not shown in the figure, the prediction request 15 also includes a terminal ID (Identification Data) for uniquely identifying the operator terminal 13 that transmitted the prediction request 15.

[0032] 3, the structure information 16 includes a candidate drug ID for uniquely identifying the candidate drug 11C, and also includes a SMILES (Simplified Molecular Input Line Entry System) character string of the candidate drug 11C.

[0033] 1 again, when the information processing device 12 receives the prediction request 15, it predicts the properties of the candidate drug 11C. Then, it delivers a prediction result 17 of the properties of the candidate drug 11C to the operator terminal 13 that sent the prediction request 15. When the prediction result 17 is received, the operator terminal 13 makes the prediction result 17 available for viewing by the operator OP.

[0034] 4, the computers constituting the information processing device 12 and the operator terminal 13 basically have the same configuration, and include a storage 25, a memory 26, a CPU (Central Processing Unit) 27, a communication unit 28, a display 29, and an input device 30. These are interconnected via a bus line 31.

[0035] The storage 25 is a hard disk drive built into the computer that constitutes the information processing device 12 and the operator terminal 13, or connected via a cable or network. Alternatively, the storage 25 is a disk array in which multiple hard disk drives are connected in series. The storage 25 stores control programs such as an operating system, various application programs (hereinafter referred to as APs (Application Programs)), and various data associated with these programs. Note that a solid state drive may be used instead of a hard disk drive.

[0036] The memory 26 is a work memory for the CPU 27 to execute processing. The CPU 27 loads a program stored in the storage 25 into the memory 26 and executes processing in accordance with the program. In this way, the CPU 27 comprehensively controls each part of the computer. The CPU 27 is an example of a "processor" according to the technology of the present disclosure. The memory 26 may be built into the CPU 27.

[0037] The communication unit 28 is a network interface that controls the transmission of various information via the network 14, etc. The display 29 displays various screens. The various screens are provided with operation functions using a GUI (Graphical User Interface). The information processing device 12 and the computer that constitutes the operator terminal 13 accept input of operation instructions from an input device 30 via the various screens. The input device 30 is a keyboard, a mouse, a touch panel, a microphone for voice input, etc.

[0038] In the following explanation, the parts of the computer that make up the information processing device 12 (storage 25 and CPU 27) are distinguished by adding the suffix "A" to their symbols, and the parts of the computer that make up the operator terminal 13 (storage 25, CPU 27, display 29, and input device 30) are distinguished by adding the suffix "B" to their symbols.

[0039] 5, an operating program 35 is stored in the storage 25A of the information processing device 12. The operating program 35 is an AP for causing a computer to function as the information processing device 12. In other words, the operating program 35 is an example of an "operating program of an information processing device" according to the technology of the present disclosure. The storage 25A also stores a prediction model 36, reference information 37, display form information 38, and the like.

[0040] When the operating program 35 is started, the CPU 27A of the computer constituting the information processing device 12 works in cooperation with the memory 26, etc. to function as a request receiving unit 40, a read / write (hereinafter abbreviated as RW (Read Write)) control unit 41, a derivation unit 42, a prediction unit 43, and a screen distribution control unit 44.

[0041] The request receiving unit 40 receives various requests from the operator terminal 13. In particular, the request receiving unit 40 receives a prediction request 15 from the operator terminal 13. The prediction request 15 includes the structure information 16, as described above. Therefore, by receiving the prediction request 15, the request receiving unit 40 receives the structure information 16. When receiving the prediction request 15, the request receiving unit 40 outputs the structure information 16 included in the prediction request 15 to the RW control unit 41. Furthermore, the request receiving unit 40 outputs the terminal ID of the operator terminal 13 included in the prediction request 15 to the screen distribution control unit 44.

[0042] The RW control unit 41 controls the storage of various data in the storage 25A and the reading of various data from the storage 25A. For example, the RW control unit 41 stores the structure information 16 from the request receiving unit 40 in the storage 25A. The RW control unit 41 also reads the structure information 16 from the storage 25A and outputs the read structure information 16 to the derivation unit 42.

[0043] Furthermore, the RW control unit 41 reads the prediction model 36 from the storage 25A and outputs the read prediction model 36 to the prediction unit 43. The RW control unit 41 also reads the reference information 37 and the display form information 38 from the storage 25A and outputs the read reference information 37 and the display form information 38 to the screen distribution control unit 44.

[0044] The derivation unit 42 derives the feature quantity 50 of the candidate drug 11C from the structural information 16. The derivation unit 42 outputs the feature quantity 50 to the prediction unit 43.

[0045] As an example, as shown in FIG. 6, the feature 50 includes a candidate drug ID. The feature 50 also includes molecular descriptors as elements. Specifically, the molecular descriptors are theoretical molecular descriptors. More specifically, the molecular descriptors include zero-dimensional descriptors such as configuration descriptors and count descriptors, one-dimensional descriptors such as fingerprints, and two-dimensional descriptors such as graph invariants. The molecular descriptors also include three-dimensional descriptors such as WHIM (Weighted Holistic Invariant Molecular) descriptors and quantum chemistry descriptors, and four-dimensional descriptors such as Volsurf descriptors. In other words, the feature 50 can be considered a multidimensional feature vector having multiple types of molecular descriptors as elements. The molecular descriptors are an example of an "element" according to the technology of the present disclosure.

[0046] Returning to FIG. 5 , the prediction unit 43 uses the prediction model 36 to output a prediction result 17 corresponding to the feature amount 50. The prediction unit 43 outputs the prediction result 17 to the screen delivery control unit 44. The prediction model 36 is a machine learning model based on an algorithm such as support vector regression, boosting, a neural network, or a random forest method. In other words, the prediction model 36 is an example of a "machine learning model" according to the technology of the present disclosure.

[0047] The screen distribution control unit 44 controls the distribution of various screens to the operator terminal 13. Specifically, the screen distribution control unit 44 distributes and outputs various screens to the operator terminal 13 that has sent the various requests in the form of screen data for web distribution created using a markup language such as XML (Extensible Markup Language). At this time, the screen distribution control unit 44 identifies the operator terminal 13 that has sent the various requests based on the terminal ID from the request receiving unit 40. The various screens include a structure information input screen 70 (see FIG. 13) for inputting structure information 16, and a first prediction result display screen 75A (see FIG. 14) and a second prediction result display screen 75B (see FIG. 15) for displaying the prediction result 17. Note that instead of XML, other data description languages ​​such as JSON (Javascript (registered trademark) Object Notation) may be used.

[0048] As an example, as shown in FIG. 7, the prediction unit 43 selects molecular descriptors that contribute to predicting the properties of the candidate drug 11C from among multiple types of molecular descriptors that make up the feature 50. As a result, the prediction unit 43 sets the feature 50 as a contributing feature 50CO. The molecular descriptors that contribute to the prediction are derived in advance, for example, as follows. That is, various combinations of input molecular descriptors are changed and the prediction model 36 is caused to output prediction results 17, and combinations of molecular descriptors that have a relatively large influence on the prediction results 17 are derived as molecular descriptors that contribute to the prediction.

[0049] The prediction unit 43 inputs the contributing feature 50CO into the prediction model 36 and causes the prediction model 36 to output a prediction result 17. The prediction result 17 includes a candidate drug ID. In this example, the characteristic of the candidate drug 11C is the encapsulation ability of the candidate drug 11C into the liposome 21. Therefore, the prediction result 17 registers either "high" or "low" as the encapsulation ability into the liposome 21.

[0050] As an example, as shown in FIG. 8, learning data (also referred to as teacher data or training data) 55 is used in learning the prediction model 36. The learning data 55 is a set of a learning contribution feature 50COL and a correct inclusion 17CA. The learning contribution feature 50COL is a contribution feature 50CO of a drug 11 whose inclusion in a liposome 21 is known. The correct inclusion 17CA is the inclusion in a liposome 21 of the original drug 11 of the learning contribution feature 50COL. The correct inclusion 17CA is, so to speak, data for checking the answer.

[0051] In training the prediction model 36, the training contribution feature 50COL is input to the prediction model 36, which then outputs a training prediction result 17L. This training prediction result 17L is compared with the correct answer entailment 17CA, and a loss calculation is performed for the prediction model 36 using a loss function based on the comparison result. Then, various coefficients of the prediction model 36 are updated according to the result of the loss calculation, and the prediction model 36 is updated according to the update setting.

[0052] In training the prediction model 36, the series of processes described above, including inputting the training contribution features 50COL to the prediction model 36, outputting the training prediction results 17L from the prediction model 36, calculating the loss, updating the settings, and updating the prediction model 36, are repeatedly performed while exchanging the training data 55. The repetition of the series of processes described above is terminated when the prediction accuracy of the training prediction results 17L reaches a predetermined set level. The prediction model 36 whose prediction accuracy has thus reached the set level is stored in the storage 25A. Note that training may be terminated when the series of processes described above has been repeated a set number of times, regardless of the prediction accuracy of the training prediction results 17L. Training of the prediction model 36 may be performed in the information processing device 12 or in a device separate from the information processing device 12. Training of the prediction model 36 may also be continued after the prediction model 36 is stored in the storage 25A.

[0053] As an example, as shown in FIG. 9, the reference information 37 is a registered information of the contributing feature 50CO and the inclusion property in the liposome 21 for each of a plurality of reference drugs 11R. The reference drug 11R is a drug 11 different from the candidate drug 11C, and is a drug 11 whose inclusion property in the liposome 21 is known. There are, for example, several tens to several thousands of reference drugs 11R. The inclusion property of the reference drug 11R in the liposome 21 is obtained by inputting the contributing feature 50CO into a prediction model 36 and outputting a prediction result 17 from the prediction model 36.

[0054] 10, the screen delivery control unit 44 reduces the dimensions of the contributing features 50CO of the candidate drug 11C and the reference drug 11R to three-dimensional display features 50D. Dimensionality reduction techniques include principal component analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE), and uniform manifold approximation and projection (UMAP). If the contributing features 50CO are three-dimensional or less, this dimension reduction process is omitted.

[0055] As an example, as shown in FIG. 11 , the display form information 38 defines a display form according to the degree of encapsulation in the liposome 21. In this example, the display form is represented by different colors. More specifically, the display form of the plot 60 corresponding to the display feature 50D of the candidate drug 11C is green when the degree of encapsulation in the liposome 21 is high and orange when the degree of encapsulation is low. Furthermore, the display form of the regions corresponding to the display feature 50D of the reference drugs 11R is green (region 61H) when the degree of encapsulation in the liposome 21 is high and orange (region 61L) when the degree of encapsulation in the liposome 21 is low. Here, region 61H is a region that includes all of the display feature 50D of the reference drug 11R that has a high degree of encapsulation in the liposome 21 among the reference drugs 11R. Similarly, region 61L is a region that includes all of the display feature 50D of the reference drug 11R that has a low degree of encapsulation in the liposome 21 among the reference drugs 11R.

[0056] The screen distribution control unit 44 generates a display feature space 62 having a plot 60 and regions 61H and 61L based on the display feature 50D of the candidate drug 11C and the reference drug 11R and the display format information 38. FIG. 11 illustrates an example in which the candidate drug 11C has a high degree of encapsulation in the liposome 21. The display feature space 62 is an example of a "feature space" according to the technology of the present disclosure.

[0057] As an example, as shown in FIG. 12 , a prediction AP 65 is stored in the storage 25B of the operator terminal 13. The prediction AP 65 is installed in the operator terminal 13 by the operator OP. The prediction AP 65 is an AP for receiving a service of predicting the properties of a candidate drug 11C by the information processing device 12. When the prediction AP 65 is activated, the CPU 27B of the operator terminal 13 functions as a browser control unit 67 in cooperation with the memory 26 and the like. The browser control unit 67 controls the operation of a web browser dedicated to the prediction AP 65.

[0058] The browser control unit 67 reproduces various screens based on various screen data from the information processing device 12 and displays the reproduced various screens on the display 29B. The browser control unit 67 also accepts various operation instructions input by the operator OP from the input device 30B via the various screens. The browser control unit 67 transmits various requests, including a prediction request 15, to the information processing device 12 in response to the operation instructions.

[0059] When the prediction AP 65 is started and logged in, a structural information input screen 70, as shown in Fig. 13 as an example, is displayed on the display 29B under the control of the browser control unit 67. The structural information input screen 70 is provided with an input box 71 for the structural information 16. In the input box 71, a SMILES string can be written as the structural information 16, or a file containing the SMILES string can be dropped.

[0060] The operator OP inputs the desired structural information 16 into the input box 71, and then selects the prediction button 72 with the cursor 73. When the prediction button 72 is selected, the browser control unit 67 generates a prediction request 15 including the structural information 16 input into the input box 71, and transmits the generated prediction request 15 to the information processing device 12.

[0061] Furthermore, when the information processing device 12 predicts the encapsulation ability of the candidate drug 11C in the liposome 21, a first prediction result display screen 75A shown in FIG. 14 is displayed on the display 29B under the control of the browser control unit 67. The first prediction result display screen 75A displays the SMILES string, which is the structural information 16 of the candidate drug 11C, and the prediction result 17. FIG. 14 illustrates an example in which the encapsulation ability of the candidate drug 11C in the liposome 21 is high, and the word "High" is written in the block representing the prediction result 17. The block representing the prediction result 17 has a display form corresponding to the prediction result 17, similar to the plot 60 and the like. In the example of FIG. 14, the block is colored green.

[0062] A save button 76 and an OK button 77 are provided at the bottom of the first prediction result display screen 75A. When the save button 76 is selected with the cursor 73, the structural information 16 and the prediction result 17 are associated with each other by the candidate drug ID and stored in the storage 25B in the form of, for example, a CSV (Comma-Separated Values) file. When the OK button 77 is selected with the cursor 73, the display on the first prediction result display screen 75A is cleared.

[0063] A display switching tab 78 is provided at the top of the first prediction result display screen 75A. When "3D Graph" on the display switching tab 78 is selected with the cursor 73, a second prediction result display screen 75B, as shown in FIG. 15 as an example, is displayed on the display 29B under the control of the browser control unit 67. The second prediction result display screen 75B displays a display feature space 62 including the plot 60 and regions 61H and 61L, along with a legend 80. By displaying the display feature space 62 including the plot 60 and regions 61H and 61L in this manner, the screen delivery control unit 44 presents a prediction result 17 of the encapsulation ability of the candidate drug 11C in the liposome 21 and the known properties of the reference drug 11R in a manner that allows them to be compared.

[0064] A check box 81, an identification number 82, and an information display button 83 are provided at the top of the second prediction result display screen 75B. Like the plot 60 and the like, the check box 81 has a display form corresponding to the prediction result 17. An identification number 82 is also assigned to the plot 60 in the display feature space 62. A capture button 84, a zoom button 85, a translation button 86, a rotation button 87, and a home button 88 are provided at the bottom of the second prediction result display screen 75B. The functions of these GUIs will be described later. Note that when "Prediction Results" in the display switching tab 78 is selected with the cursor 73 on the second prediction result display screen 75B, the display returns to the first prediction result display screen 75A.

[0065] A re-input button 89 is further provided at the bottom of the second prediction result display screen 75B. When the re-input button 89 is selected with the cursor 73, the display returns to the structural information input screen 70, as shown in FIG. 16 as an example, and it becomes possible to input structural information 16 of a new candidate drug 11C. FIG. 16 illustrates a state in which, in addition to the structural information 16 shown in FIG. 13, the structural information 16 of a new candidate drug 11C, surrounded by a dashed line, has been input into the input box 71. The re-input button 89 may also be provided on the first prediction result display screen 75A.

[0066] When the prediction button 72 is selected with the cursor 73 in the state shown in Figure 16, the browser control unit 67 generates a prediction request 15 including the structural information 16 of the new candidate drug 11C entered in the input box 71, and sends the generated prediction request 15 to the information processing device 12.

[0067] When the information processing device 12 predicts the encapsulation ability of a new candidate drug 11C into a liposome 21, as shown in Figure 17 as an example, the first prediction result display screen 75A displays the structural information 16 and prediction result 17 of the new candidate drug 11C, as shown surrounded by a dotted line, in addition to the structural information 16 and prediction result 17 of the candidate drug 11C for which structural information 16 has been received up to that point.

[0068] 18, for example, in the display feature space 62 of the second prediction result display screen 75B, in addition to the plot 60 with the identification number 82 of "1" that corresponds to the display feature 50D of the candidate drug 11C for which structural information 16 has been received up to that point, a plot 60 with the identification number 82 of "2" that corresponds to the display feature 50D of a new candidate drug 11C is additionally displayed. As can be seen from this explanation, the identification number 82 makes it possible to display multiple plots 60 corresponding to multiple candidate drugs 11C in a manner that allows each of the multiple candidate drugs 11C to be distinguished.

[0069] 16 to 18 show an example in which the encapsulation ability of a certain drug candidate 11C in a liposome 21 is predicted, and then the encapsulation ability of a new drug candidate 11C in a liposome 21 is predicted. However, the encapsulation ability of multiple drug candidates 11C in a liposome 21 is not limited to this example. Alternatively, the structural information 16 of multiple drug candidates 11C may be input from the beginning on the structural information input screen 70, and the encapsulation ability of multiple drug candidates 11C in a liposome 21 may be predicted collectively. In this case, a CSV file in which multiple SMILES character strings are registered may be received in the input box 71 as the structural information 16. Considering fairness among operators OP and the processing load of the CPU 27A, it is preferable to limit the number of drug candidates 11C that can be collectively predicted, such as to 10.

[0070] 19, when a check box 81 is selected with the cursor 73 and unchecked, the plot 60 corresponding to the check box 81 is hidden in the display feature space 62. When the check box 81 is again selected with the cursor 73 and checked again, the plot 60 corresponding to the check box 81 is redisplayed in the display feature space 62. That is, the screen delivery control unit 44 switches between displaying and hiding the plot 60 in response to an operation instruction from the operator OP to select the check box 81. FIG. 19 illustrates an example in which the plot 60 with the identification number 82 of "2" is hidden.

[0071] 20, when a plot 60 is selected with a cursor 73, coordinates 92 of the plot 60 in the display feature space 62 are displayed. That is, the screen distribution control unit 44 presents the coordinates 92 of the plot 60 in response to an operation instruction from an operator OP to select the plot 60.

[0072] 21, when an information display button 83 is selected with the cursor 73, a chemical structural formula 95 of the candidate drug 11C corresponding to the information display button 83 is displayed. That is, the screen delivery control unit 44 presents the chemical structural formula 95 of the candidate drug 11C in response to an operation instruction from the operator OP to select the information display button 83. The chemical structural formula 95 is generated from a SMILES character string as the structural information 16.

[0073] As an example, as shown in Fig. 22, when a predetermined operation such as rotating the mouse wheel is performed after the zoom button 85 is selected with the cursor 73, the display feature space 62 is zoomed in or out. That is, the screen delivery control unit 44 zooms in or out the display feature space 62 in response to an operation instruction from the operator OP. Fig. 22 illustrates a case in which the display feature space 62 is enlarged compared to the default display state shown in Fig. 15 etc.

[0074] 23, when a predetermined operation is performed, such as selecting the rotation button 87 with the cursor 73 and then moving the mouse while right-clicking, the display feature space 62 is rotated. That is, the screen delivery control unit 44 rotates the display feature space 62 in response to an operation instruction from the operator OP.

[0075] The functions of the other buttons are as follows. First, the capture button 84 is a button for storing the display feature space 62 including the plot 60 and the regions 61H and 61L in the storage 25B, for example, in the format of a PNG (Portable Network Graphics) file. The translation button 86 is a button for translating the display feature space 62 in any of the up, down, left, and right directions. The home button 88 is a button for returning the display feature space 62 to the default display. Note that a button for rotating the display feature space 62 while fixing the Z axis may also be provided.

[0076] Next, the operation of the above configuration will be described with reference to the flowchart shown in Fig. 24 as an example. When the operating program 35 is started in the information processing device 12, the CPU 27A functions as the request receiving unit 40, the RW control unit 41, the derivation unit 42, the prediction unit 43, and the screen delivery control unit 44, as shown in Fig. 5. When the prediction AP 65 is started in the operator terminal 13, the CPU 27B functions as the browser control unit 67, as shown in Fig. 12.

[0077] 13 is displayed on the display 29B of the operator terminal 13 under the control of the browser control unit 67. On the structural information input screen 70, the operator OP inputs desired structural information 16 into an input box 71 and selects a prediction button 72 with a cursor 73. This causes a prediction request 15 to be transmitted from the browser control unit 67 to the information processing device 12. As shown in FIG. 1, the prediction request 15 includes the structural information 16.

[0078] In the information processing device 12, the request receiving unit 40 receives the prediction request 15, thereby receiving the structure information 16 included in the prediction request 15 (YES in step ST100). The structure information 16 included in the prediction request 15 is output from the request receiving unit 40 to the RW control unit 41, and stored in the storage 25A under the control of the RW control unit 41 (step ST110). In addition, the terminal ID of the operator terminal 13 included in the prediction request 15 is output from the request receiving unit 40 to the screen delivery control unit 44.

[0079] The structure information 16 is read from the storage 25A by the RW control unit 41 (step ST120). The structure information 16 is output from the RW control unit 41 to the derivation unit .

[0080] 6, the derivation unit 42 derives feature quantities 50 of the candidate drug 11C, including molecular descriptors, from the structural information 16 (step ST130). The feature quantities 50 are output from the derivation unit 42 to the prediction unit 43. In addition, the RW control unit 41 reads out the prediction model 36 from the storage 25A, and outputs the read prediction model 36 to the prediction unit 43.

[0081] 7, in the prediction unit 43, from among the multiple types of molecular descriptors constituting the feature 50, molecular descriptors that contribute to predicting the encapsulation property of the candidate drug 11C in the liposome 21 are selected, and the feature 50 is set as a contributing feature 50CO (step ST140). Then, the contributing feature 50CO is input to the prediction model 36. As a result, the prediction model 36 outputs a prediction result 17 of the encapsulation property of the candidate drug 11C in the liposome 21 (step ST150). The prediction result 17 is output from the prediction unit 43 to the screen distribution control unit 44. In addition, the RW control unit 41 reads out the reference information 37 and the display form information 38 from the storage 25A, and outputs the read out reference information 37 and the display form information 38 to the screen distribution control unit 44.

[0082] In the screen distribution control unit 44, as shown in FIG. 10, the contribution feature 50CO of the candidate drug 11C and the reference drug 11R is reduced in dimension to generate a three-dimensional display feature 50D. Furthermore, as shown in FIG. 11, a display feature space 62 is generated based on the display feature 50D and the display format information 38. The display feature space 62 includes a plot 60 and regions 61H and 61L. The plot 60 corresponds to the display feature 50D of the candidate drug 11C and has a display format corresponding to the prediction result 17. The regions 61H and 61L correspond to the multiple display feature 50D of the multiple reference drugs 11R and have a display format corresponding to the known encapsulation of the reference drug 11R in the liposome 21.

[0083] In this way, the screen distribution control unit 44 generates screen data for the first and second prediction result display screens 75A and 75B (step ST160). The screen data for the first and second prediction result display screens 75A and 75B are distributed to the operator terminal 13 that is the sender of the prediction request 15 under the control of the screen distribution control unit 44 (step ST170).

[0084] 14 and 15, in the operator terminal 13, under the control of the browser control unit 67, the screen data of the first and second prediction result display screens 75A and 75B are reproduced, and the reproduced first and second prediction result display screens 75A and 75B are displayed on the display 29B. As a result, the prediction result 17 of the encapsulation ability of the candidate drug 11C in the liposome 21 and the known properties of the reference drug 11R are presented to the operator OP so as to be comparable.

[0085] The operator OP views the first and second prediction result display screens 75A and 75B, and then verifies the validity of the prediction result 17 by referring to the display feature space 62, and finally decides whether or not to carry out the production of the pharmaceutical preparation 20 using the candidate drug 11C.

[0086] As described above, the CPU 27A of the information processing device 12 includes a request receiving unit 40, a derivation unit 42, a prediction unit 43, and a screen distribution control unit 44. The request receiving unit 40 receives the structural information 16 of the candidate drug 11C by receiving a prediction request 15. The derivation unit 42 derives the feature value 50 of the candidate drug 11C from the structural information 16. The prediction unit 43 predicts the encapsulation of the candidate drug 11C in the liposome 21 based on the contribution feature value 50CO of the candidate drug 11C. The screen distribution control unit 44 outputs and distributes screen data of the second prediction result display screen 75B to the operator terminal 13, thereby presenting a prediction result 17 of the encapsulation of the candidate drug 11C in the liposome 21 and the known characteristics of the reference drug 11R in a comparable manner. This allows the operator OP to easily compare the prediction result 17 with the known characteristics of the reference drug 11R. This contributes to verifying the validity of the prediction result 17 and increases the sense of acceptability of the prediction result 17. These effects are extremely beneficial for the operator OP, who has little practical knowledge of liposome formulations and is unable to empirically predict the properties of the drug 11 entangled in the liposome 21.

[0087] 15 and other figures, the screen delivery control unit 44 displays the plot 60, which corresponds to the display feature 50D of the candidate drug 11C and has a display form corresponding to the prediction result 17, in the display feature space 62. This allows the operator OP to recognize at a glance the position of the candidate drug 11C in the display feature space 62 and the prediction result 17 of the encapsulation of the candidate drug 11C in the liposome 21. This further facilitates verification of the validity of the prediction result 17.

[0088] 15 and other figures, the screen distribution control unit 44 displays in the display feature space 62 the regions 61H and 61L, which correspond to the plurality of display features 50D of the plurality of reference drugs 11R and have display forms according to the known encapsulation properties of the reference drugs 11R in the liposomes 21. This allows the operator OP to recognize at a glance the region 61H with a high encapsulation property in the liposomes 21 and the region 61L with a low encapsulation property. This further facilitates verification of the validity of the prediction result 17.

[0089] As shown in FIGS. 16 and 18, upon receiving the structural information 16 of a new drug candidate 11C and deriving the feature 50 of the new drug candidate 11C, the screen distribution control unit 44 additionally displays the plot 60 corresponding to the display feature 50D of the new drug candidate 11C in the display feature space 62, in addition to the plot 60 corresponding to the display feature 50D of the drug candidate 11C for which the structural information 16 has been received up to that point. This allows the operator OP to simultaneously view the prediction results 17 of multiple drug candidates 11C. Furthermore, the operator OP can simultaneously verify the validity of the prediction results 17 of multiple drug candidates 11C. This further facilitates the verification of the validity of the prediction results 17.

[0090] 18, the screen distribution control unit 44 displays multiple plots 60 corresponding to multiple candidate drugs 11C in a manner that allows each of the multiple candidate drugs 11C to be identified using the identification numbers 82. This allows the operator OP to accurately recognize which plot 60 corresponds to which candidate drug 11C. This reduces the possibility that the operator OP will make a mistake, such as mistaking the prediction results 17.

[0091] 19, the screen delivery control unit 44 switches between displaying and hiding the plot 60 corresponding to the candidate drug 11C in response to an operation instruction from the operator OP. This makes it possible to meet the operator OP's request to verify the validity of the prediction result 17 for each candidate drug 11C.

[0092] 11, the display format is different colors. This improves the visibility of the operator OP in the display feature space 62. This further facilitates verification of the validity of the prediction result 17.

[0093] 20, the screen delivery control unit 44 presents coordinates 92 of the plot 60 corresponding to the display feature 50D of the candidate drug 11C in response to an operation instruction from the operator OP. This allows the operator OP to accurately grasp the position of the candidate drug 11C in the display feature space 62.

[0094] 11 and other figures, the display feature space 62 is a three-dimensional space, which allows the validity of the prediction result 17 to be verified more precisely than in the case of a one-dimensional or two-dimensional space.

[0095] 22 and 23, the screen distribution control unit 44 enlarges, reduces, and / or rotates the display feature space 62 in response to an operation instruction from the operator OP. This allows the operator OP to check in detail the positional relationship between the plot 60 and the areas 61H and 61L, which is difficult to understand in the default display. This further facilitates verification of the validity of the prediction result 17.

[0096] 21, the screen delivery control unit 44 presents the chemical structural formula 95 of the candidate drug 11C in response to an operation instruction from the operator OP. This allows the operator OP to verify the validity of the prediction result 17 while checking the chemical structural formula 95 of the candidate drug 11C.

[0097] As shown in Fig. 6, the feature 50 is composed of multiple types of molecular descriptors. As shown in Fig. 7, the prediction unit 43 predicts the encapsulation of the candidate drug 11C in the liposome 21 from the contributing feature 50CO, which is a selection of molecular descriptors that contribute to the prediction of the encapsulation of the candidate drug 11C in the liposome 21 from the multiple types of molecular descriptors. Therefore, the prediction accuracy of the encapsulation of the candidate drug 11C in the liposome 21 can be improved compared to when the encapsulation of the candidate drug 11C in the liposome 21 is predicted from the feature 50 including all of the multiple types of molecular descriptors.

[0098] 7, the prediction unit 43 inputs the contribution feature quantity 50CO of the candidate drug 11C into the prediction model 36, and causes the prediction model 36 to output a prediction result 17. Therefore, the prediction result 17 with high prediction accuracy can be easily obtained.

[0099] As shown in Figure 6, the feature 50 includes molecular descriptors as elements. Molecular descriptors are commonly used, so the feature 50 can be easily derived.

[0100] As shown in Figures 7 and 9, the characteristic is the encapsulation property of the candidate drug 11C or the reference drug 11R into the liposome 21. This can discourage the operator OP from preparing the pharmaceutical formulation 20 using the candidate drug 11C with a relatively low encapsulation property, thereby reducing unnecessary costs. This can be very useful for the operator OP in selecting the drug 11. This can also be very useful for the operator OP working for a pharmaceutical company in selecting the drug 11 when outsourcing the pharmaceutical formulation 20 to a contract research organization.

[0101] Although regions 61H and 61L corresponding to the plurality of display features 50D of the plurality of reference drugs 11R are displayed in the display feature space 62, this is not limiting. As an example, as shown in FIG. 25 , a plurality of plots 101H and 101L corresponding to the plurality of display features 50D of the plurality of reference drugs 11R may be displayed in the display feature space 62. The plot 101H corresponds to the display feature 50D of the reference drug 11R having a high degree of encapsulation in the liposome 21. In contrast, the plot 101L corresponds to the display feature 50D of the reference drug 11R having a low degree of encapsulation in the liposome 21. As with the regions 61H and 61L in the first embodiment, the plot 101H is displayed in green, and the plot 101L is displayed in orange. In this case, the plot 60 is slightly larger than the plots 101H and 101L.

[0102] In this way, the screen delivery control unit 44 displays, in the display feature space 62, the plots 101H and 101L, which are the multiple plots 101H and 101L corresponding to the multiple display features 50D of the multiple reference drugs 11R and have display forms according to the known encapsulation properties in the liposomes 21. This aspect can also contribute to verifying the validity of the prediction result 17.

[0103] Furthermore, while different colors are used as examples of display formats, this is not limiting. For example, the display format may be different patterns, as shown in the display format information 105 in FIG. 26 . More specifically, the display format of the plot 60 corresponding to the display feature 50D of the candidate drug 11C is a pattern of diagonal lines arranged at equal intervals from the upper right to the lower left when the encapsulation degree in the liposome 21 is high, and a pattern of diagonal lines arranged at equal intervals from the upper left to the lower right when the encapsulation degree is low. Furthermore, the display format of the regions corresponding to the multiple display feature 50D of the multiple reference drugs 11R is a pattern of diagonal lines arranged at equal intervals from the upper right to the lower left (region 61H) when the encapsulation degree in the liposome 21 is high, and a pattern of diagonal lines arranged at equal intervals from the upper left to the lower right (region 61L) when the encapsulation degree in the liposome 21 is low. This embodiment also improves the visibility of the operator OP in the display feature space 62. Although not shown, the display format may be different colors and patterns. Furthermore, the display format may be different color intensities.

[0104] [Second embodiment] As shown in FIG. 27 as an example, in the second embodiment, prediction models 36 corresponding to multiple types of liposomes 21 (liposomes A, B, and C), i.e., a liposome A prediction model 36A, a liposome B prediction model 36B, and a liposome C prediction model 36C, are prepared. As shown in FIG. 28 as an example, a prediction request 110 in the second embodiment includes type designation information 111 in addition to the structural information 16. The type designation information 111 registers the type of liposome 21 designated by the operator OP when inputting the structural information 16. The prediction unit 43 selects and uses a prediction model 36 corresponding to the type of liposome 21 in this type designation information 111. FIG. 28 illustrates an example in which the designated type of liposome 21 is liposome B, and the liposome B prediction model 36B is selected and used.

[0105] As described above, in the second embodiment, a plurality of types of prediction models 36 are prepared according to the types of liposomes 21. The prediction unit 43 selects and uses a prediction model 36 according to the type of liposomes 21. Therefore, prediction according to the type of liposomes 21 can be performed.

[0106] The first prediction result display screen 75A may display the chemical structure formula 95 together with the structural information 16 and the prediction result 17. The first prediction result display screen 75A may be configured to allow unnecessary structural information 16 and prediction result 17 to be deleted.

[0107] The encapsulation property of the reference drug 11R in the liposome 21 registered in the reference information 37 is the predicted result 17 from the prediction model 36, but this is not limitative. The encapsulation property of the reference drug 11R in the liposome 21 obtained by actually conducting an experiment may also be registered.

[0108] Although the output format of the prediction result 17 has been exemplified as two levels, "high" and "low," for the encapsulation ability of the candidate drug 11C in the liposome 21, the output format is not limited to this. For example, the encapsulation ability of the candidate drug 11C in the liposome 21 may be five levels, "very high," "slightly high," "moderate," "slightly low," and "very low." In this case, five regions are displayed in the display feature space 62.

[0109] Although the encapsulation ability into the liposome 21 has been exemplified as an example of a characteristic, this is not limiting. As an example, the release ability of the candidate drug 11C from the liposome 21 may be predicted as a characteristic, as shown in the prediction model 115 and prediction result 116 shown in Figure 29. This embodiment also makes it possible to discourage the operator OP from preparing the pharmaceutical preparation 20 using the candidate drug 11C with a relatively high release ability, thereby reducing unnecessary costs.

[0110] The characteristics may also include the amount of candidate drug 11C encapsulated in liposome 21, the behavior of liposome 21 encapsulating candidate drug 11C when pharmaceutical preparation 20 is applied to a living body (e.g., whether it is easily absorbed by the liver), the safety of pharmaceutical preparation 20 when applied to a living body, and the stability of candidate drug 11C encapsulated in liposome 21 (e.g., whether it takes a short time from encapsulation in liposome 21 to release). Multiple types of characteristics may be predicted at once. In this case, a display feature space 62 is also generated for each of the multiple types of characteristics.

[0111] Instead of the exemplified SMILES character string, a chemical structural formula 95 may be input as the structural information 16. Furthermore, a MOL (Molecular Design Limited) file or an SDF (Structure Data Format) file may be input.

[0112] Although liposomes 21 have been shown as an example of vesicles, the present invention is not limited to this. For example, lipid nanoparticles 121 may be used, as in the pharmaceutical preparation 120 shown in FIG. 30. The lipid nanoparticles 121 are, as the name suggests, particles having a nano-sized diameter and made of lipids. Alternatively, for example, micelles 126 may be used, as in the pharmaceutical preparation 125 shown in FIG. 31. The micelles 126 are particles having a nano-sized diameter and made of surfactants.

[0113] The candidate substance and the reference substance are not limited to the exemplary candidate drug 11C and reference drug 11R, but may also be active ingredients of cosmetics, active ingredients of dietary supplements, or the like.

[0114] The feature quantity 50 may include, in addition to or instead of the exemplified molecular descriptors, molecular weight, the number of carbon-carbon bonds, the number of carbon-oxygen bonds, the number of carbon-nitrogen bonds, the number of six-membered rings, etc. Various parameters obtained by experiments and / or simulations, such as solubility, acid dissociation constant, and charge number, may also be included in the feature quantity 50. Furthermore, the display feature quantity space 62 may be a one-dimensional space or a two-dimensional space.

[0115] The information processing device 12 may be installed in a pharmaceutical facility, or may be installed in a data center independent of the pharmaceutical facility.

[0116] Instead of delivering screen data of the first and second prediction result display screens 75A and 75B to the operator terminal 13, data that is the basis of the first and second prediction result display screens 75A and 75B, such as the prediction result 17, reference information 37, and display format information 38, may be delivered to the operator terminal 13. In this case, under the control of the browser control unit 67, the operator terminal 13 generates the first and second prediction result display screens 75A and 75B based on the prediction result 17, reference information 37, and display format information 38, and displays the first and second prediction result display screens 75A and 75B on the display 29B.

[0117] The method of presenting the prediction result 17 of the property of the candidate drug 11C and the known property of the reference drug 11R to the operator OP is not limited to the delivery of screen data as shown in the example. The prediction result 17 of the property of the candidate drug 11C and the known property of the reference drug 11R, i.e., the display feature space 62 including the plot 60 and the regions 61H and 61L, may be presented to the operator OP by printing them on a paper medium, or the display feature space 62 including the plot 60 and the regions 61H and 61L may be presented to the operator OP by attaching it to an email and sending it to the operator terminal 13.

[0118] The hardware configuration of the computer constituting the information processing device 12 according to the technology of the present disclosure can be modified in various ways. For example, the information processing device 12 can be configured with multiple computers separated as hardware in order to improve processing power and reliability. For example, the functions of the request receiving unit 40 and the RW control unit 41 and the functions of the derivation unit 42, the prediction unit 43, and the screen delivery control unit 44 can be distributed and performed by two computers. In this case, the information processing device 12 is configured with two computers. Furthermore, some or all of the functions of the information processing device 12 may be performed by the operator terminal 13.

[0119] In this way, the hardware configuration of the computer of the information processing device 12 can be changed as appropriate depending on the required performance such as processing power, safety, reliability, etc. Furthermore, not only the hardware but also APs such as the operating program 35 can be duplicated or stored in a distributed manner in multiple storage devices in order to ensure safety and reliability.

[0120] In each of the above embodiments, the following various processors may be used as the hardware structure of processing units that perform various processes, such as the request receiving unit 40, the RW control unit 41, the derivation unit 42, the prediction unit 43, the screen delivery control unit 44, and the browser control unit 67. As described above, the various processors include the CPUs 27A and 27B, which are general-purpose processors that execute software (the operating program 35 and the prediction AP 65) and function as various processing units, as well as dedicated electrical circuits that are processors having a circuit configuration specifically designed to perform specific processes, such as a programmable logic device (PLD) that is a processor whose circuit configuration can be changed after manufacture, such as an FPGA (Field Programmable Gate Array), and an ASIC (Application Specific Integrated Circuit).

[0121] A single processing unit 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 (e.g., a combination of multiple FPGAs and / or a combination of a CPU and an FPGA). Also, multiple processing units may be configured with a single processor.

[0122] Examples of configuring multiple processing units with a single processor include, first, a form in which one processor is configured with a combination of one or more CPUs and software, and this processor functions as multiple processing units, as typified by client and server computers. Second, a form in which a processor is used to realize the functions of an entire system including multiple processing units with a single IC (Integrated Circuit) chip, as typified by System on Chip (SoC). In this way, various processing units are configured using one or more of the above-mentioned various processors as a hardware structure.

[0123] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements.

[0124] From the above description, the technology described in the following supplementary paragraphs can be understood.

[0125] [Additional note 1] a processor; The processor: We accept structural information about candidate substances to be encapsulated in vesicles, deriving feature quantities of the candidate substance from the structural information; predicting properties of the candidate substance from the feature quantities of the candidate substance; Predicting the properties of the candidate substance and the known properties of a reference substance in a comparable manner. Information processing device. [Additional note 2] The processor: The information processing device according to appended item 1, wherein the plot corresponds to the feature quantities of the candidate substances and has a display form according to the prediction result in a feature quantity space. [Additional note 3] The processor: The information processing device according to appended item 2, wherein the information processing device displays in the feature space an area corresponding to a plurality of feature amounts of a plurality of the reference substances and having a display form according to the known characteristics. [Additional note 4] The processor: 3. The information processing device according to claim 2, wherein the plots are a plurality of plots corresponding to a plurality of feature quantities of a plurality of the reference substances, and the plots have a display form according to the known characteristics, and are displayed in the feature quantity space. [Additional note 5] The processor: Upon receiving the structural information of the new candidate substance and deriving the feature amount of the new candidate substance, An information processing device according to any one of appendix 2 to appendix 4, wherein a plot corresponding to a feature quantity of a new candidate substance is additionally displayed in the feature space in addition to a plot corresponding to a feature quantity of the candidate substance for which the structural information has been received up to that point. [Additional note 6] The processor: 6. The information processing device according to claim 5, wherein a plurality of plots corresponding to a plurality of the candidate substances are displayed in a manner that allows each of the plurality of candidate substances to be distinguished. [Additional note 7] The processor: 7. The information processing device according to any one of claims 2 to 6, wherein the display and non-display of plots corresponding to the candidate substances are switched in response to an operation instruction from an operator. [Additional note 8] 8. The information processing device according to any one of claims 2 to 7, wherein the display form is a difference in color and / or pattern. [Additional note 9] The processor: 9. The information processing device according to any one of supplementary items 2 to 8, wherein coordinates of plots corresponding to the feature quantities of the candidate substances are presented in response to an instruction from an operator. [Additional Note 10] 10. The information processing device according to any one of Supplementary Items 2 to 9, wherein the feature space is a three-dimensional space. [Additional Note 11] The processor: 11. The information processing device according to claim 10, wherein the feature space is expanded / contracted and / or rotated in response to an operation instruction from an operator. [Additional Note 12] The processor: 12. The information processing device according to any one of claims 1 to 11, wherein the information processing device presents chemical structural formulas of the candidate substances in response to an instruction from an operator. [Additional Note 13] the feature amount is composed of a plurality of types of elements, The processor: 13. The information processing device according to any one of claims 1 to 12, wherein the information processing device predicts the properties of the candidate substance from contributing feature quantities selected from among the plurality of types of elements that contribute to the prediction of the properties. [Additional Note 14] The processor: 14. The information processing device according to any one of claims 1 to 13, wherein the feature amount of the candidate substance is input to a machine learning model, and the prediction result is output from the machine learning model. [Additional Note 15] A plurality of types of the machine learning model are prepared according to the type of the vesicle, The processor: The information processing device according to claim 14, wherein the machine learning model is selected and used according to the type of vesicle. [Additional Note 16] 16. The information processing device according to any one of claims 1 to 15, wherein the feature amount includes a molecular descriptor as an element. [Additional Note 17] An information processing device described in any one of appendix 1 to appendix 16, wherein the characteristics include at least one of the ability to encapsulate the candidate substance or the reference substance into the vesicle and the ability to release the candidate substance or the reference substance from the vesicle. [Additional Note 18] 18. The information processing device according to any one of claims 1 to 17, wherein the vesicle is any one of a liposome, a lipid nanoparticle, and a micelle.

[0126] The technology of the present disclosure can be appropriately combined with the various embodiments and / or various modified examples described above. Furthermore, it is not limited to the above embodiments, and various configurations can be adopted without departing from the spirit of the present disclosure. Furthermore, the technology of the present disclosure extends not only to programs but also to storage media that non-temporarily store programs.

[0127] 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.

[0128] In this specification, "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."

[0129] 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. [Explanation of symbols]

[0130] 10 Information Processing Systems 11 Drugs 11C candidate drug 11R Reference Drug 12 Information processing equipment 13 Operator terminal 14 Network 15, 110 Forecast Request 16 Structural information 17, 116 Prediction results 17CA Correct Answer Intense 17L Learning prediction results 20, 120, 125 Pharmaceutical preparations 21 Liposomes 25, 25A, 25B Storage 26 memory 27, 27A, 27B CPUs 28 Communications Department 29, 29B Display 30, 30B input devices 31 Bus Line 35 Operating Program 36, 115 Prediction models 36A Predictive model for Liposome A 36B Predictive model for liposome B 36C Prediction model for liposomal C 37 References 38, 105 Display format information 40 Request Reception Unit 41 Read / write control unit (RW control unit) 42 Derivation part 43 Prediction Department 44 Screen distribution control unit 50 features 50CO contributing features 50COL Contributing features for training 50D Display Features 55 training data 60 plots 61H, 61L area 62 Display feature space 65 Predictive Application Program (Predictive AP) 67 Browser control section 70 Structural information input screen 71 Input box 72 Prediction button 73 Cursor 75A First prediction result display screen 75B, 100B Second prediction result display screen 76 Save button 77 OK button 78 Display Switching Tab 80 Legend 81 Checkboxes 82 Identification Number 83 Information display button 84 Capture button 85 Zoom button 86 Parallel movement button 87 Rotation button 88 Home button 89 Re-enter button 92 coordinates 95 Chemical Structural Formula 101H, 101L plots 111 Type Designation Information 121 Lipid Nanoparticles 126 Micelle OP Operator ST100, ST110, ST120, ST130, ST140, ST150, ST160, ST170 Step

Claims

1. a processor; The processor: We accept structural information about candidate substances to be encapsulated in vesicles, deriving feature quantities of the candidate substance from the structural information; predicting properties of the candidate substance from the feature quantities of the candidate substance; Predicting the properties of the candidate substance and the known properties of a reference substance in a comparable manner. Information processing device.

2. The processor: The information processing apparatus according to claim 1 , wherein a plot corresponding to the feature quantities of the candidate substances and having a display form according to the prediction result is displayed in the feature quantity space.

3. The processor: The information processing apparatus according to claim 2 , wherein regions corresponding to a plurality of feature quantities of a plurality of the reference substances and having a display form according to the known properties are displayed in the feature quantity space.

4. The processor: The information processing apparatus according to claim 2 , wherein a plurality of plots corresponding to a plurality of feature quantities of a plurality of the reference substances and having a display form according to the known characteristics are displayed in the feature quantity space.

5. The processor: Upon receiving the structural information of the new candidate substance and deriving the feature amount of the new candidate substance, 3. The information processing apparatus according to claim 2, wherein plots corresponding to feature quantities of new candidate substances are additionally displayed in the feature quantity space in addition to plots corresponding to feature quantities of candidate substances for which the structural information has been received up to that point.

6. The processor: The information processing apparatus according to claim 5 , wherein a plurality of plots corresponding to a plurality of said candidate substances are displayed in a manner that allows each of said plurality of candidate substances to be distinguished.

7. The processor:

3. The information processing apparatus according to claim 2, wherein the plots corresponding to the candidate substances are switched between display and non-display in response to an instruction from an operator.

8. The information processing device according to claim 2 , wherein the display mode is a difference in color and / or pattern.

9. The processor: The information processing apparatus according to claim 2 , wherein the coordinates of a plot corresponding to the feature amount of the candidate substance are presented in response to an instruction from an operator.

10. The information processing apparatus according to claim 2 , wherein the feature space is a three-dimensional space.

11. The processor: The information processing apparatus according to claim 10 , wherein the feature space is expanded / contracted and / or rotated in response to an operation instruction from an operator.

12. The processor:

2. The information processing apparatus according to claim 1, wherein the chemical structural formula of the candidate substance is presented in response to an instruction from an operator.

13. the feature amount is composed of a plurality of types of elements, The processor: The information processing apparatus according to claim 1 , wherein the properties of the candidate substance are predicted from contributing feature quantities obtained by selecting elements that contribute to the prediction of properties from among the plurality of types of elements.

14. The processor: The information processing device according to claim 1 , wherein the feature quantities of the candidate substances are input to a machine learning model, and the prediction results are output from the machine learning model.

15. A plurality of types of the machine learning model are prepared according to the type of the vesicle, The processor: The information processing device according to claim 14 , wherein the machine learning model is selected and used according to the type of the vesicle.

16. The information processing apparatus according to claim 1 , wherein the feature quantity includes a molecular descriptor as an element.

17. The information processing device according to claim 1 , wherein the characteristics include at least one of the ability to encapsulate the candidate substance or the reference substance in the vesicle and the ability to release the candidate substance or the reference substance from the vesicle.

18. The information processing device according to claim 1 , wherein the vesicle is any one of a liposome, a lipid nanoparticle, and a micelle.

19. Accepting structural information of candidate substances to be encapsulated in the vesicles; deriving feature quantities of the candidate substance from the structural information; predicting the properties of the candidate substance from the feature quantities of the candidate substance; and Presenting the predicted properties of the candidate substance in a manner that allows comparison with the known properties of a reference substance; A method for operating an information processing device, comprising:

20. Accepting structural information of candidate substances to be encapsulated in the vesicles; deriving feature quantities of the candidate substance from the structural information; predicting the properties of the candidate substance from the feature quantities of the candidate substance; and Presenting the predicted properties of the candidate substance in a manner that allows comparison with the known properties of a reference substance; An operating program for an information processing device that causes a computer to execute processing including the above.

Images