WO2025089029A1 - Determination method, determination device, determination system, determination program, and recording medium - Google Patents

Abstract

This determination method is for determining the addition amount of a PEG reagent used in ligation with respect to a plurality of small RNA molecules in a biological sample collected from a subject by employing small RNA data that indicate the results of the measurement of expression levels of the plurality of small RNA molecules. The determination method includes an addition amount determination step for determining the above-mentioned addition amount from the above-mentioned small RNA data by using determination criteria based on the relationship between measurement data that indicate the results of the measurement of expression levels of a plurality of small RNA molecules in biological samples collected from a plurality of subjects and addition amount data that indicate addition amounts of a PEG reagent used in ligation with respect to the plurality of small RNA molecules in the biological samples.

Description

Determination method, determination device, determination system, determination program, and recording medium

This disclosure relates to a determination method, a determination device, a determination system, a determination program, and a recording medium.

　Japanese Patent No. 7021097 discloses a disease prevalence assessment device that includes a sample data acquisition unit and a prevalence assessment unit. The sample data acquisition unit acquires sample data including the expression levels of multiple types of miRNA in a sample derived from a living body. The prevalence assessment unit uses a trained model to output prevalence assessment results for the multiple diseases in the multiple body parts for the acquired sample data. The trained model is a trained model that can assess the prevalence of each of the multiple diseases, including multiple malignant diseases or multiple benign diseases, including cases where the patient is suffering from multiple diseases, obtained in advance by machine learning using training data including multiple sample data having items for identifying the presence or absence of multiple diseases in multiple body parts.

　WO 2021/132547 discloses a testing method for testing for a disease using a disease marker, the testing method including a specimen data acquisition step and a discrimination step. The specimen data acquisition step acquires marker data indicating the results of measuring a disease marker in a body fluid sample collected from a subject, and preparation data indicating the preparation conditions of the body fluid sample. The discrimination step determines the presence or absence of a disease in the subject by inputting the marker data and preparation data acquired in the specimen data acquisition step into a trained model that has been machine-learned to learn the correlation between a set of marker data indicating the results of measuring a disease marker in a body fluid sample and preparation data indicating the preparation conditions of the body fluid sample, and the presence or absence of a disease in the subject from which the body fluid sample was collected.

Non-Patent Document 1 discloses that, using 20 types of fluorescently labeled synthetic miRNAs, the reaction efficiency of each miRNA in a 3' ligation enzymatic reaction in which a nucleic acid sequence is added to the 3' end of the miRNA was investigated under multiple conditions in which the type of ligation enzyme, the amount of PEG reagent added, the reaction time, etc. were varied, and the results showed that the efficiency of the miRNA ligation reaction may vary depending on the PEG concentration.
Non-patent document 1: Song Y, Liu KJ, Wang TH. Elimination of ligation dependent artifacts in T4 RNA ligase to achieve high efficiency and low bias microRNA capture. PLoS One. 2014 Apr 10

When diagnosing a disease based on the expression levels of multiple small RNAs in a biological sample from a subject, if ligation is performed at the 5' or 3' ends of multiple small RNAs using an adapter and a PEG reagent, the expression levels of the small RNAs that are ultimately measured may change depending on the amount of PEG reagent added. Therefore, the amount of PEG reagent added affects the diagnosis of the disease.

In particular, PEG reagents have a relatively high viscosity, and there is a tendency for the differences in technique between people who add the PEG reagent to be large. For this reason, there is a tendency for the amount of PEG reagent added to vary.

The present disclosure aims to provide a determination method, determination device, determination system, determination program, and recording medium that can determine with high accuracy the amount of PEG reagent added when ligating an adapter and a PEG reagent to the 5' or 3' end of multiple small RNAs.

The determination method according to one embodiment of the present disclosure is a method for determining the amount of PEG reagent added used when performing ligation on multiple small RNAs in a biological sample collected from a subject, using small RNA data showing the results of measuring the expression levels of the multiple small RNAs, and includes an addition amount determination step for determining the amount added from the small RNA data using a determination criterion based on the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added used when performing ligation on the multiple small RNAs in the biological sample.

According to the present disclosure, it is possible to determine with high accuracy the amount of PEG reagent to be added when ligating an adapter and a PEG reagent to the 5' or 3' end of multiple small RNAs.

FIG. 2 is a flow chart showing each step of a determination method according to the present embodiment. FIG. 2 is a schematic block diagram of an example of a computer that functions as a determination device according to the present embodiment. 1 is a block diagram showing a functional configuration of a determination device according to an embodiment of the present invention; A diagram showing the criteria for binary judgment in addition amount learned models 1 to 4 of this embodiment. 13 is a table showing the judgment results in additive amount learned models 1 to 4 of the present embodiment. FIG. 2 is a conceptual diagram illustrating a continuous value regression model according to the present embodiment. 11 is a graph showing the determination results in the continuous value regression model of the present embodiment.

Below, an example of an embodiment of the technology of the present disclosure will be described with reference to the drawings. Note that components and processes that perform the same operation, action, and function will be given the same reference numerals throughout the drawings, and duplicated explanations may be omitted as appropriate. Each drawing is merely a schematic illustration to allow a sufficient understanding of the technology of the present disclosure. Therefore, the technology of the present disclosure is not limited to the illustrated examples. Furthermore, in this embodiment, explanations of configurations that are not directly related to the present disclosure or well-known configurations may be omitted.

<Determination Method 10>
First, a description will be given of a determination method 10 according to the present embodiment. Fig. 1 is a schematic diagram showing each step of the determination method 10 according to the present embodiment.

Determination method 10 is a method for determining the amount of PEG reagent added (hereinafter, sometimes referred to as the amount of PEG added) used when performing ligation on multiple small RNAs in a biological sample collected from a subject, using small RNA data that shows the results of measuring the expression levels of the multiple small RNAs.

Examples of biological samples collected from subjects include body fluids such as blood, serum, urine, tears, saliva, sweat, semen, lymph, tissue fluid, body cavity fluid (e.g., pleural fluid, ascites, etc.), cerebrospinal fluid, amniotic fluid, vaginal fluid, and nasal mucus. The biological sample may be any sample that can be collected from a living body and from which the expression levels of multiple small RNAs can be measured. The term "subject" refers to the subject from which the biological sample is collected. The subject from which the biological sample is collected may be either a human or a non-human animal. Examples of non-human animals include non-human mammals (monkeys, dogs, cats, mice, rats, rabbits, cows, horses, pigs, and sheep), and birds (chickens, quails, etc.).

An example of a small RNA is microRNA. The small RNA may be a small RNA other than microRNA (e.g., piRNA and tsRNA). Ligation and PEG reagents will be described later.

Specifically, as shown in FIG. 1, the determination method 10 includes a collection step 11, a ligation step 13, a measurement step 12, an addition amount determination step 14, and a disease determination step 16. In this embodiment, the collection step 11, the ligation step 13, the measurement step 12, the addition amount determination step 14, and the disease determination step 16 are performed in this order, for example.

Since the determination method 10 is a method having a sampling step 11, a measurement step 12, and a disease determination step 16, it can also be called a sampling method, a measurement method, and a disease determination method. Furthermore, if analysis, testing, etc. is performed based on the determination in the disease determination step 16, it can also be called an analysis method, a testing method, etc. Each step of the determination method 10 will be explained below.

<Collecting step 11>
The collecting step 11 is a step of collecting a biological sample from a subject. The collecting step 11 is performed before the measurement step 12. In the case where serum is collected as the biological sample in the collecting step 11, for example, a collection process is performed as follows.

First, blood is collected using a vacuum blood collection tube containing a serum separating agent (blood collection process).
Next, immediately after the blood collection process, the blood is mixed by inversion.
The blood is then allowed to coagulate at room temperature for at least 30 minutes.
Next, the mixture is centrifuged and the serum is separated (centrifugation treatment).
Then, the serum is separated and stored at −80° C. (storage treatment).

In the collection process 11, the process from collection of a biological sample from a subject to completion of the centrifugation operation (specifically, the process from the blood collection process to the preservation process described above) is performed within a predetermined time (e.g., 2 hours). The centrifugation operation is, for example, the process from the centrifugation process to the preservation process described above.

<Ligation step 13>
Ligation is an enzymatic reaction in which a base sequence called an adapter is bound to the 5' or 3' end of a nucleic acid such as DNA or RNA including a small RNA. The ligation step 13 is a step in which a base sequence called an adapter is bound to the 5' or 3' end of a small RNA. In the ligation step 13, a PEG reagent is used when performing the ligation. Specifically, the ligation step 13 is performed, for example, in the preparation of a library for NGS.

The PEG reagent is a reagent that contains PEG (PolyEthylene Glycol). Any PEG reagent that contains at least PEG can be used, and it is possible to use one that contains other components. PEG is composed of a polymer of the monomer Ethylene Glycol, and there are several types of PEG due to differences in the degree of polymerization and average molecular weight (for example, PEG 200, PEG 300, PEG 400, PEG 600, PEG 1000, PEG 2000, PEG 4000, PEG 6000, PEG 8000, PEG 10000, PEG 20000, PEG 500000, PEG 2000000, PEG 4000000, etc.). However, since all of them have a common chemical structure (i.e., ethylene glycol, a monomer), their chemical properties are similar, so any type of PEG can be used. PEG functions as a catalyst to promote the enzymatic reaction in ligation, and reduces the bias in reactivity of each small RNA.

As an example of the PEG reagent, a reagent containing the following components can be used.
Nuclease-free water
T4 RNA Ligase Reaction Buffer(NEB/M0242)
100% DMSO (13445-74/Nacalai Tesque)
50% PEG (NEB/B1004)
Total RNA

<Measurement step 12>
The measurement step 12 is a step of measuring the expression levels of multiple small RNAs in the subject's biological sample. Specifically, in the measurement step 12, a next-generation sequencer (NGS) is used as a measurement device to measure multiple small RNAs contained in the subject's biological sample (for example, in serum) and identify the base sequence of each small RNA. Next, the number of identified small RNAs is counted for each base sequence to obtain the number of reads of the small RNA in the NGS. This number of reads of the small RNA corresponds to the expression level (specifically, the absolute expression level) of the small RNA. In other words, this number of reads of the small RNA means small RNA data showing the result of measuring the expression levels of multiple small RNAs in the subject's biological sample. However, although the expression levels of multiple small RNAs (for example, microRNAs) in a biological sample (for example, serum) are absolute values derived from the living body, it is difficult to quantify the expression levels of small RNAs in a biological sample with an absolute value because it is necessary to quantify the expression levels through a process of going through a measurement device or reagent treatment. Therefore, the expression levels of small RNAs may be expressed as relative values by processing the data from the NGS measurement results, and such relative values also refer to small RNA data.

Furthermore, in order to measure the absolute expression level, the number of reads of small RNA output by NGS may be normalized to obtain a relative expression level (i.e., relative expression level). As a means of normalization, for example, RPM (Read Per Million) normalization or normalization using a small RNA as an internal standard may be used. In this way, the small RNA data may indicate a normalized relative expression level. Note that the small RNA data may be an absolute value quantified as an absolute value.

In addition, NGS makes it possible to simultaneously measure the expression levels of multiple small RNAs in multiple biological samples (e.g., serum samples collected from multiple subjects). That is, in measurement step 12, the expression levels of multiple small RNAs are measured for multiple biological samples in the same process.

In addition to next-generation sequencers, other measuring devices such as DNA chips, quantitative PCR, and flow cytometers can also be used as long as the expression levels of multiple small RNAs can be measured. As described above, if the expression levels of multiple small RNAs can be measured and a small RNA showing the measurement results can be obtained, various methods, including publicly known methods, can be used to measure the expression levels of multiple small RNAs.

<Additional amount determination step 14>
The amount of PEG reagent added is determined using small RNA data indicating the results of measuring the expression levels of the plurality of small RNAs in the amount determination step 14. The determination of the amount of PEG reagent added includes a concept of determining whether the amount is appropriate and a concept of determining the amount itself.

In this embodiment, in the addition amount determination step 14, the small RNA data obtained in the measurement step 12 is input into the addition amount trained model to determine whether the amount of PEG reagent to be added is appropriate.

The criterion for determining whether the amount of PEG reagent to be added is appropriate is set, for example, based on whether a correct judgment can be made in the disease assessment process 16. If a correct judgment can be made in the disease assessment process 16 even if the amount of PEG reagent added is low, the criterion becomes one that allows a low amount to be added.

The allowable amount of addition varies depending on the disease to be judged in the disease judgment process 16, and the judgment criteria in the addition amount judgment process 14 also vary correspondingly. Furthermore, the judgment process is not limited to disease judgment, and various judgments can be applied, and the allowable amount of addition varies depending on the disease to be judged, and the judgment criteria in the addition amount judgment process 14 also vary correspondingly.

<How to generate additive amount trained model>
Here, the added amount trained model used in the added amount determination step 14 is generated, for example, as follows: That is, the added amount trained model is generated by machine learning the correlation between measurement data indicating the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects, and added amount data indicating the added amount of PEG reagent used when performing ligation on the multiple small RNAs in the biological samples collected from the multiple subjects.

Specifically, the measurement data is the result of measuring multiple small RNAs in biological samples collected from multiple subjects. The multiple subjects may include subjects with a disease and subjects without a disease (i.e., healthy subjects), or may include only subjects with a disease and subjects without a disease (i.e., healthy subjects).

The measurement data is obtained by the same measurement method as the small RNA data described above. Specifically, a next-generation sequencer (NGS) is used as a measurement device to measure multiple small RNAs contained in a biological sample from a subject, and to identify the base sequence of each small RNA. Next, the number of identified small RNAs is counted for each base sequence to determine the number of small RNA reads in the NGS. This number of small RNA reads corresponds to the expression level (specifically, absolute expression level) of the small RNA. In other words, this number of small RNA reads becomes the measurement data.

The amount of PEG added data is specifically data that indicates whether the amount of PEG added related to the measurement data is appropriate. As an example, the amount of PEG added data can be data that indicates that an amount of PEG added that allows a correct judgment to be performed in the disease assessment process 16 is appropriate (good), and that an amount of PEG added that does not allow a correct judgment to be performed in the disease assessment process 16 is inappropriate (bad).

Specifically, as the measurement data to be subjected to machine learning (hereinafter sometimes referred to as training data), a trained model for the amount of PEG added can be generated using measurement data obtained from biological samples collected from multiple subjects under conditions in which the amount of PEG added is appropriate, and measurement data obtained from biological samples collected from multiple subjects under conditions in which the amount of PEG added is inappropriate.

The disease of the subject that is machine-learned by the additive amount trained model used in the additive amount determination step 14 may include the disease determined in the disease determination step 16. That is, if the disease determined in the disease determination step 16 is a cancer disease, the measurement data of a sample from a subject with a cancer disease is included. Note that the type of cancer disease determined in the disease determination step 16 and the cancer disease of the measurement data may be the same or different. Therefore, for example, if the disease determined in the disease determination step 16 is pancreatic cancer, the measurement data of a sample from a subject with lung cancer may be used as training data.

In the addition amount determination step 14, measurement data from the top 100 small RNAs with the highest expression levels in biological samples from multiple subjects is used as training data.

In this way, in this embodiment, measurement data is used that excludes small RNAs with relatively low expression levels. Small RNAs with relatively low expression levels have the advantage that they can easily function as explanatory variables, since even small fluctuations in expression level result in large fold changes. However, they also have the disadvantage of being easily affected by fluctuations due to factors other than the element of interest (specifically, the amount of PEG added) and fluctuations due to measurement errors, making them less robust. Therefore, by using only small RNAs with relatively high expression levels as training data, the effect of strengthening robustness can be achieved.

In the addition amount determination step 14, the measurement data excluding small RNAs whose expression level was zero in the biological samples of any of the subjects is used as training data. Since the variation from zero is an infinite fold change, which is difficult to quantify and is particularly susceptible to variations due to factors other than the amount of PEG added and variations due to measurement errors, small RNAs whose expression level is zero in any of the samples used as training data are excluded from the training data.

As described above, in this embodiment, the learning model is made to qualitatively learn whether the amount of PEG to be added is appropriate, thereby generating an added amount trained model. The added amount trained model then judges whether the amount of PEG to be added is appropriate, and outputs this suitability, thereby judging whether the amount of PEG to be added is appropriate in the added amount judgment process 14. Therefore, the added amount judgment process 14 displays a judgment result that the amount of PEG to be added is appropriate, or a judgment result that the amount of PEG to be added is inappropriate.

In this embodiment, the learning model is trained to learn whether the amount of PEG added is appropriate, and an additive amount trained model that performs a binary judgment is generated, but this is not limited to this. For example, an additive amount trained model may be generated so that the amount of PEG added is judged as one of three values: appropriate (good), warning, and inappropriate (bad).

As described above, in this embodiment, the learning model is trained on the appropriateness of the amount of PEG added qualitatively to generate the learned model of the amount of PEG added, but this is not limited to the above. For example, the learning model may be trained on the amount of PEG added quantitatively to generate the learned model of the amount of PEG added. In this case, the amount of PEG added is judged by the learned model of the amount of PEG added, and the judgment result of the amount of PEG added is output, for example, as a judgment value. The judgment value here is not a binary value of 0 or 1, but a value having a predetermined numerical range. Then, in the addition amount judgment process 14, the appropriateness of the amount of PEG added is judged based on the judgment result of the judgment unit that judges the appropriateness of the amount of PEG added based on a comparison between the output judgment value and a threshold value. That is, in the addition amount judgment process 14, if the judgment unit judges that the output judgment value is equal to or greater than the threshold value, the judgment result that the amount of PEG added is inappropriate is displayed, and if the judgment unit judges that the output judgment value is less than the threshold value, the judgment result that the amount of PEG added is appropriate is displayed.

<Modification of the addition amount determination step 14>
As described above, in the addition amount determination step 14, the amount of PEG added is determined using an addition amount trained model that has been machine-learned to determine the correlation between the measurement data and the addition amount data, but this is not limited to this.

For example, the amount of PEG added may be determined using a regression model that determines the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on multiple small RNAs in the biological samples collected from the multiple subjects.

In this modified example, the amount of PEG to be added is determined by inputting small RNA data into a regression model in the amount-to-add determination step 14. Specifically, in the amount-to-add determination step 14, for example, the small RNA data is input into a regression model to output a value for the amount of PEG to be added.

In this case, the amount of PEG to be added is determined by a regression model, and the determination result of the amount of PEG to be added is output, for example, as a determination value. The determination value here is not a binary value of 0 or 1, but a value having a predetermined numerical range. Then, in the amount-of-addition determination step 14, the appropriateness of the amount of PEG to be added may be determined based on the determination result of a determination unit that determines whether the amount of PEG to be added is appropriate based on a comparison between the output determination value and a threshold value. That is, in the amount-of-addition determination step 14, if the determination unit determines that the output determination value is equal to or greater than the threshold value, the determination result that the amount of PEG to be added is inappropriate is displayed, and if the determination unit determines that the output determination value is less than the threshold value, the determination result that the amount of PEG to be added is appropriate is displayed.

The addition amount determination step 14 is not limited to the use of the above-mentioned addition amount learned model and the above-mentioned regression model, and it is possible to use a determination criterion based on various algorithms. The determination criterion used in the addition amount determination step 14 is a determination criterion based on the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when ligating multiple small RNAs in the biological samples, and it is sufficient that the addition amount can be determined from the small RNA data.

<Medition determination step 16>
The disease determination step 16 is a step of determining the presence or absence of a disease. Specifically, in the disease determination step 16, the small RNA data obtained in the measurement step 12 is input to a disease-trained model that has been machine-learned to determine the correlation between measurement data showing the results of measuring a plurality of small RNAs in a biological sample and disease data showing the presence or absence of a disease in a subject from whom the biological sample was collected, thereby determining the presence or absence of a disease.

In the disease assessment step 16, if the result of the assessment in the addition amount assessment step 14 is that the amount of PEG added is appropriate, the presence or absence of a disease is assessed. In other words, if the result of the assessment in the addition amount assessment step 14 is that the amount of PEG added is inappropriate, the disease assessment step 16 is not executed. In this case, for example, the result of the assessment that the amount of PEG added is inappropriate may be presented to the user (i.e., the person who executes the assessment method).

It is desirable that the disease determined by the disease discrimination trained model is the same as the disease used for training in the additive amount trained model. For example, if the disease used for training in the additive amount trained model is cancer, the disease determined by the disease discrimination trained model is also cancer. Specifically, it is desirable that the disease type of the measurement data used as training data in the disease discrimination trained model is the same as the disease type of the measurement data used as training data in the additive amount trained model.

As described above, if the addition amount determination step 14 determined that the amount of PEG added was inappropriate, the disease determination step 16 was not executed, but this is not limited to the above. For example, even if the addition amount determination step 14 determined that the amount of PEG added was inappropriate, the disease determination step 16 may be executed if the determination result is to be obtained as reference data, for example. In this case, for example, after the determination result that the amount of PEG added was inappropriate is obtained, the fact that the determination was made may be presented to the user. Also, in this embodiment, the disease determination step 16 was executed after the addition amount determination step 14, but it may be executed before the addition amount determination step 14. In this case, if the addition amount determination step 14 determined that the amount of PEG added was inappropriate, the determination result of the disease determination step 16 is treated as reference data, for example.

<Determination system 20>
Next, a description will be given of a determination system 20 that executes the above-mentioned determination method. The determination system 20 includes a measurement device 21 and a determination device 30, as shown in FIG.

<Measuring device 21>
The measurement device 21 is an example of a measurement unit, and is a device that executes the above-mentioned measurement step 12. That is, the measurement device 21 measures the expression levels of multiple small RNAs contained in each of multiple samples. As the measurement device 21, for example, an NGS is used.

<Determination device 30>
The determination device 30 is an example of a determination unit. The determination device 30 is a device that executes the above-mentioned addition amount determination step 14. That is, the determination device 30 acquires small RNA data indicating the results of measuring a plurality of small RNAs in a biological sample of a subject, and inputs the acquired small RNA data into an addition amount trained model to determine the amount of PEG added to the biological sample collected from the subject.

Furthermore, the determination device 30 executes the disease determination step 16 described above. That is, the determination device 30 determines the presence or absence of a disease by inputting the small RNA data obtained in the measurement step 12 into a disease-trained model that has been machine-learned to determine the correlation between measurement data showing the results of measuring multiple small RNAs in a biological sample and disease presence data showing the presence or absence of a disease in the subject from whom the biological sample was collected.

The determination device 30 functions as a computer, and as shown in FIG. 2, has a CPU (Central Processing Unit) 31, a ROM (Read Only Memory) 32, a RAM (Random Access Memory) 33, a storage 34, an input unit 35, a display unit 36, and a communication interface (I/F) 37. Each component is connected to each other via a bus 39 so that they can communicate with each other.

CPU 31 (an example of a processor) is a central processing unit that executes various programs and controls each part. That is, CPU 31 reads a program from ROM 32 or storage 34, and executes the program using RAM 33 as a working area. CPU 31 controls each of the above components and performs various calculation processes according to the program stored in ROM 32 or storage 34. CPU 31 is an example of a processor.

ROM 32 records various programs and various data. RAM 33 temporarily stores programs or data as a working area. Storage 34 is composed of a HDD (Hard Disk Drive) or SSD (Solid State Drive), and records various programs including the operating system, and various data.

In this embodiment, for example, a judgment program for executing a judgment process that performs the above-mentioned judgment method is recorded in storage 34. The judgment program may be a single program, or may be a group of programs consisting of multiple programs or modules. The judgment program may be recorded in ROM 32. ROM 32 and storage 34 function as an example of a non-transitory recording medium.

An example of a processor is not limited to the aforementioned CPU, which is a general-purpose processor, but may be, for example, a dedicated processor made up of a circuit designed specifically to execute a specific process. Also, an example of a processor is not limited to a single processor, but may be a processor made up of multiple processors working together at physically separate locations.

The input unit 35 includes a pointing device such as a mouse and a keyboard, and is used to perform various inputs. The input unit 35 also receives as input information on the expression levels of multiple small RNAs measured by the measurement device 21.

The display unit 36 is, for example, a liquid crystal display, and displays various information. The determination device 30 can present the determination result of the amount of PEG added and the determination result of the presence or absence of a disease to the user through the display unit 36. The display unit 36 may also function as the input unit 35 by adopting a touch panel system.

The communication interface 37 is an interface for communicating with other devices, and uses standards such as Ethernet (registered trademark), FDDI (Fiber Distributed Data Interface), and Wi-Fi (registered trademark).

As shown in FIG. 3, in the judgment device 30, the CPU 31 executes the judgment program to function as a judgment function unit 160 and a disease judgment unit 170.

The judgment function unit 160 executes the aforementioned addition amount judgment step 14. That is, in the addition amount judgment step 14, the judgment function unit 160 inputs the small RNA data obtained by the measurement device 21 into the addition amount learned model to judge whether the amount of PEG added to the biological sample collected from the subject is appropriate (see the aforementioned addition amount judgment step 14).

The disease determination unit 170 executes the disease determination step 16. That is, the disease determination unit 170 determines the presence or absence of a disease by inputting the small RNA data obtained by the measurement device 21 into a disease-trained model that has been machine-learned to determine the correlation between measurement data showing the results of measuring multiple small RNAs in a biological sample and disease data showing the presence or absence of a disease in the subject from whom the biological sample was collected (see the disease determination step 16 described above).

In the determination device 30, the CPU 31 may perform processing to output the determination result of the amount of PEG reagent added and the determination result of the presence or absence of a disease. This processing is performed, for example, by displaying on the display unit 36 and transmitting to an external device (e.g., the cloud).

In this embodiment, the determination system 20 includes the measurement device 21 and the determination device 30, but the determination system 20 may be configured with a single device. In this case, the single device functions as an example of the measurement unit and the determination unit.

The determination device 30 may also be composed of multiple devices. For example, the determination device 30 may be composed of multiple (e.g., two) devices that share the functions of the addition amount determination step 14 and the disease determination step 16 described above.

<Effects of this embodiment>
In this embodiment, as described above, in the addition amount determination step 14, the amount of PEG reagent to be added is determined by inputting the small RNA data obtained in the measurement step 12 into an addition amount trained model that has machine-learned the correlation between measurement data indicating the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data indicating the amount of PEG reagent added when performing ligation on the multiple small RNAs in the biological samples.

In this way, the amount of addition is determined by inputting small RNA data into an addition amount trained model that has been machine-learned to learn the correlation between measurement data and addition amount data, so the amount of PEG reagent added can be determined with a high degree of accuracy compared to, for example, determination by visual inspection or measurement using a measurement kit.

In this way, in this embodiment, the amount of PEG reagent added can be determined with high accuracy, so highly accurate information on the amount of PEG reagent added can be obtained, and for example, in the disease determination process 16, there is no need to obtain information on the amount of PEG reagent added by inputting it by the user.

<Example>
Next, examples will be described. Note that the examples are merely examples of the technology of the present disclosure, and the technology of the present disclosure is not limited to the contents of the examples.

<Acquisition of training data and evaluation data>
In this example, serum was obtained as a biological sample from 24 healthy subjects by performing the collection step 11 within 2 hours. The serum was subjected to the ligation step 13 using five PEG reagent formulations in which the amount of PEG was increased or decreased based on the basic formulation below, and a total of 120 samples (24 subjects x 5 formulations) were obtained.

For the serum that underwent ligation step 13 (i.e., 120 samples), the expression levels of multiple small RNAs were measured using NGS, and small RNA data showing the measurement results were obtained. Of the obtained small RNA data, small RNAs with zero expression levels were excluded, and small RNA data for the top 100 small RNAs with the highest expression levels was used. Of the small RNA data for the 120 samples, the small RNA data for 100 samples (20 people x 5 prescriptions) was used as training data, and the small RNA data for 20 samples (4 people x 5 prescriptions) was used as evaluation data.

[Basic recipe]
Nuclease-free water 1.0μL
T4 RNA Ligase Reaction Buffer (NEB/M0242) 1.0μL
100% DMSO (13445-74/Nacalai Tesque) 1.0 μL
50% PEG (NEB/B1004) 1.0 μL
Total RNA 4.0μL

[Increase or decrease in PEG content]
Formulation 1: The amount of PEG was reduced by 50% in the basic formulation. Formulation 2: The amount of PEG was reduced by 25% in the basic formulation. Formulation 3: The amount of PEG was unchanged in the basic formulation. Formulation 4: The amount of PEG was increased by 25% in the basic formulation. Formulation 5: The amount of PEG was increased by 50% in the basic formulation.

<Creating additive amount trained model>
We generated four types of binary decision models: additive amount trained models 1 to 4 (see Figure 4).

[Addition amount trained model 1]
The training data obtained from the formula 1 was treated as a bad data group, and the training data obtained from the formulas 2 to 5 were treated as a good data group, and the training data for 100 samples was used to generate the additive amount trained model 1. In other words, the additive amount trained model 1 is a model that performs a binary judgment in which the formula 1 is bad and the formulas 2 to 5 are good.

[Addition amount trained model 2]
The training data obtained from the recipes 1 and 2 were treated as a bad data group, and the training data obtained from the recipes 3 to 5 were treated as a good data group, and the training data for 100 samples was used to generate the additive amount trained model 2. In other words, the additive amount trained model 2 is a model that performs a binary judgment in which the recipes 1 and 2 are treated as bad, and the recipes 3 to 5 are treated as good.

[Addition amount trained model 3]
The training data obtained from the recipes 1 to 3 were treated as a bad data group, and the training data obtained from the recipes 4 and 5 were treated as a good data group, and the training data for 100 samples was used to generate the additive amount trained model 3. In other words, the additive amount trained model 3 is a model that performs a binary judgment in which the recipes 1 to 3 are bad and the recipes 4 and 5 are good.

[Addition amount trained model 4]
The training data obtained from the formulas 1 to 4 were treated as a bad data group, and the training data obtained from the formula 5 were treated as a good data group, and the training data for 100 samples was used to generate the additive amount trained model 4. In other words, the additive amount trained model 4 is a model that performs a binary judgment in which the formulas 1 to 4 are treated as bad and the formula 5 is treated as good.

<Example of learning model>
Various linear and nonlinear algorithms known as machine learning algorithms can be used, or multiple algorithms can be combined. For example, the following algorithms can be used:

Random forest
Gradient boosted trees
Extreme gradient boosted trees
Light-gradient boosted machine
Neural networks
Regularized regression
Elastic-net
K-Nearest neighbors
Support vector machine
Generalized additive model

<Performance verification of additive amount trained model>
The evaluation data for 20 samples was input to each of the additive amount trained models 1 to 4, and a binary judgment of good or bad was performed. In the additive amount trained models 1 to 4, a good judgment with an AUC exceeding 0.96 was possible (see FIG. 5).

In addition, by using a combination of additive amount trained models 1 to 4, it is possible to estimate the amount of PEG added within a range that can be specified for prescriptions 1 to 5. For example, if additive amount trained model 2 judges the product as good and additive amount trained model 3 judges the product as bad, the amount of PEG added is estimated to be in the range of -25% to +25% of the basic prescription (1.0 μL).

<Generating a continuous regression model>
A continuous value regression model was generated to estimate the percentage of the amount of PEG added relative to the basic formula (1.0 μL) using all of the training data obtained from formulations 1 to 5. The continuous value regression model estimates whether the amount is -50%, -25%, 0%, +25%, or +50% relative to the basic formula (1.0 μL) (see FIG. 6).

<Performance check of continuous value regression model>
The evaluation data for the 20 samples were input into a continuous regression model to estimate the amount of PEG added (see FIG. 7). In FIG. 7, the actual amount of PEG added is shown on the horizontal axis, and the estimated amount of PEG added is shown on the vertical axis. As shown in FIG. 7, it was demonstrated that the amount of PEG added can be estimated by the continuous regression model.

The present invention is not limited to the above-described embodiment, and various modifications, changes, and improvements are possible without departing from the spirit of the invention. The above-described modifications may be combined as appropriate.

In the determination method 10, as described above, the small RNA data obtained in the measurement step 12 is input to a disease-trained model that has been machine-learned to learn the correlation between the measurement data showing the results of measuring the expression levels of multiple small RNAs in a biological sample and the disease data showing the presence or absence of a disease in the subject from whom the biological sample was collected, thereby indicating the presence or absence of a disease. Therefore, it can be said that the measured expression level of the small RNA indicates the presence or absence of a disease based on the judgment criteria in the disease-trained model. Therefore, the determination method 10 is a determination method that determines the amount of PEG reagent used when ligating multiple small RNAs in a biological sample collected from a subject using small RNA data showing the results of measuring the expression levels of the multiple small RNAs, and the expression level of the small RNA indicates the presence or absence of the disease based on the judgment criteria in the disease-trained model that has been machine-learned to learn the correlation between the measurement data showing the results of measuring the expression levels of multiple small RNAs in a biological sample and the disease data showing the presence or absence of a disease in the subject from whom the biological sample was collected.

<Additional Notes>
(Aspect 1)
A method for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
The method includes an addition amount determination step of determining the amount of addition from the small RNA data based on a determination criterion based on a correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on the multiple small RNAs in the biological samples.
(Aspect 2)
The addition amount determining step includes:
The method according to aspect 1, further comprising inputting the small RNA data into an added amount trained model that has been machine-learned to determine a correlation between the measurement data and the added amount data, thereby determining the added amount.
(Aspect 3)
In the addition amount determination step,
The method according to aspect 2, further comprising inputting the small RNA data into the additive amount trained model to determine whether the additive amount is appropriate.
(Aspect 4)
In the addition amount determination step,
The method according to aspect 2, further comprising inputting the small RNA data into the additive amount trained model to output the additive amount.
(Aspect 5)
In the addition amount determination step,
The method according to aspect 4, further comprising: determining whether the amount of addition is appropriate based on a determination result of a determination unit that determines whether the amount of addition is appropriate based on a comparison between the output amount of addition and a threshold value.
(Aspect 6)
The addition amount determining step includes:
The method according to aspect 1, further comprising inputting the small RNA data into a regression model that determines a correlation between the measurement data and the data on the amount of addition, thereby determining the amount of addition.
(Aspect 7)
In the addition amount determination step,
The method according to aspect 6, further comprising inputting the small RNA data into the regression model to determine whether the amount of addition is appropriate.
(Aspect 8)
In the addition amount determination step,
The method according to aspect 6, further comprising inputting the small RNA data into the regression model to output the amount to be added.
(Aspect 9)
In the addition amount determination step,
The method according to aspect 8, further comprising: determining whether the amount of addition is appropriate based on a determination result of a determination unit that determines whether the amount of addition is appropriate based on a comparison between the output amount of addition and a threshold value.
(Aspect 10)
a disease determination step of determining the presence or absence of a disease by inputting the small RNA data into a disease trained model that has been machine-learned to determine a correlation between measurement data showing the results of measuring a plurality of small RNAs in a biological sample and disease presence data showing the presence or absence of a disease in a subject from whom the biological sample was collected;
The method according to any one of aspects 1 to 9, further comprising:
(Aspect 10-2)
The expression level of the small RNA indicates the presence or absence of the disease based on a judgment criterion in a disease-trained model that machine-learns the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in a biological sample and disease occurrence data showing the presence or absence of the disease in the subject from whom the biological sample was collected.
The method according to any one of aspects 1 to 9.
(Aspect 11)
In the addition amount determination step, the appropriateness of the addition amount is determined,
The method according to aspect 10, wherein the morbidity determination step determines whether or not the subject is afflicted with the disease when the addition amount determination step determines that the addition amount is appropriate.
(Aspect 12)
A determination device that determines an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data indicating a result of measuring the expression levels of the plurality of small RNAs, comprising:
A processor is provided.
The processor,
A determination device that determines the amount of addition from the small RNA data based on a determination criterion based on the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on multiple small RNAs in the biological samples.
(Aspect 13)
A system for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
A measurement unit that acquires small RNA data indicating the results of measuring a plurality of small RNAs in a biological sample of a subject;
a determination unit that determines the amount of addition from the small RNA data based on a determination criterion based on a correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when ligating the multiple small RNAs in the biological samples;
A determination system having the above configuration.
(Aspect 14)
On the computer,
A process for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
A determination program for executing a determination process that determines the amount of addition from the small RNA data based on a determination criterion based on the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on multiple small RNAs in the biological samples.
(Aspect 15)
On the computer,
A process for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
A non-temporary recording medium having recorded thereon a judgment program for executing a judgment process for determining the amount of addition from the small RNA data based on a judgment criterion based on a correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on multiple small RNAs in the biological samples.

The disclosure of Japanese Patent Application No. 2023-182778, filed on October 24, 2023, is incorporated herein by reference in its entirety. All documents, patent applications, and technical standards described herein are incorporated herein by reference to the same extent as if each individual document, patent application, and technical standard was specifically and individually indicated to be incorporated by reference.

Claims (15)

A method for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
The method includes an addition amount determination step of determining the amount of addition from the small RNA data based on a determination criterion based on a correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on the multiple small RNAs in the biological samples. The addition amount determining step includes:
The method according to claim 1 , further comprising: inputting the small RNA data into an additive amount trained model that has been machine-learned to determine a correlation between the measurement data and the additive amount data, thereby determining the additive amount. In the addition amount determination step,
The method according to claim 2 , further comprising inputting the small RNA data into the additive amount trained model to determine whether the additive amount is appropriate. In the addition amount determination step,
The method according to claim 2 , further comprising: inputting the small RNA data into the additive amount trained model to output the additive amount. In the addition amount determination step,
The method according to claim 4 , further comprising the step of: determining whether the amount of addition is appropriate based on a result of a determination made by a determination unit that determines whether the amount of addition is appropriate based on a comparison between the output amount of addition and a threshold value. The addition amount determining step includes:
The method according to claim 1 , further comprising inputting the small RNA data into a regression model that determines a correlation between the measurement data and the data on the amount of addition, thereby determining the amount of addition. In the addition amount determination step,
The method according to claim 6 , further comprising inputting the small RNA data into the regression model to determine whether the amount of addition is appropriate. In the addition amount determination step,
The method according to claim 6 , further comprising inputting the small RNA data into the regression model to output the amount to be added. In the addition amount determination step,
The method according to claim 8 , further comprising the step of: determining whether the amount of addition is appropriate based on a determination result of a determination unit that determines whether the amount of addition is appropriate based on a comparison between the output amount of addition and a threshold value. a disease determination step of determining the presence or absence of a disease by inputting the small RNA data into a disease trained model that has been machine-learned to determine a correlation between measurement data showing the results of measuring a plurality of small RNAs in a biological sample and disease presence data showing the presence or absence of a disease in a subject from whom the biological sample was collected;
The method of claim 1 further comprising: In the addition amount determination step, the appropriateness of the addition amount is determined,
The method according to claim 10 , wherein in the disease determination step, the presence or absence of the disease is determined when the addition amount determination step determines that the addition amount is appropriate. A determination device that determines an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data indicating a result of measuring the expression levels of the plurality of small RNAs, comprising:
A processor is provided.
The processor,
A determination device that determines the amount of addition from the small RNA data based on a determination criterion based on the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on multiple small RNAs in the biological samples. A system for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
A measurement unit that acquires small RNA data indicating the results of measuring a plurality of small RNAs in a biological sample of a subject;
a determination unit that determines the amount of addition from the small RNA data based on a determination criterion based on a correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when ligating the multiple small RNAs in the biological samples;
A determination system having the above configuration. On the computer,
A process for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
A determination program for executing a determination process that determines the amount of addition from the small RNA data based on a determination criterion based on the correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on multiple small RNAs in the biological samples. On the computer,
A process for determining an amount of a PEG reagent used in ligating a plurality of small RNAs in a biological sample collected from a subject, using small RNA data showing a result of measuring the expression levels of the plurality of small RNAs, comprising:
A non-temporary recording medium having recorded thereon a judgment program for executing a judgment process for determining the amount of addition from the small RNA data based on a judgment criterion based on a correlation between measurement data showing the results of measuring the expression levels of multiple small RNAs in biological samples collected from multiple subjects and addition amount data showing the amount of PEG reagent added when performing ligation on multiple small RNAs in the biological samples.