JP2019124563A - Mass analyzer, method for mass analysis, mass analysis system, and mass analysis analyzing program - Google Patents

Abstract

To provide a mass analyzer for detecting an inspection target in blood, the mass analyzer allowing an easy, rapid, and accurate detection of the inspection target in the blood.SOLUTION: A mass analyzer includes: ionization means for ionizing components of an inspection target in blood of an inspection target body; and analysis means for conducting a mass analysis on ions generated by the ionization means.SELECTED DRAWING: Figure 2A

Description

  The present invention relates to a mass spectrometer, a mass spectrometry method, a mass spectrometry system, and a mass spectrometry analysis program.

In recent years, various methods for detecting biomarkers in circulating blood have been developed to diagnose disease progression in patients.
For example, circulating tumor cells (CTC or CTSs) are tumor cells present in the peripheral blood of patients with primary and metastatic origin and are detected in various types of cancer patients. ing. In recent years, CTC's DNA and RNA profiles are being elucidated, and it has become clear that CTC is involved in cancer recurrence and metastasis and treatment resistance. As a clinical test to predict cancer prognosis and treatment response to chemotherapy, more accurate detection of CTC is expected.
As a method of detecting CTC, a method of microscopically observing cells after staining by processing with an antibody against cell surface antigen of CTC has been proposed (see, for example, Patent Document 1 and Non-patent Documents 1 and 2). ).

Patent No. 6052756

T. Kawada et al .: “Circulating tumor cells in patients with head and neck squamous cell carcinoma: Feasibility of detection and quantitation” Head Neck 39 (11) 2180-2186, 2017 August 17. S. Zheng et al .: “3D microfilter device for viable circulating tumor cells (CTC) enrichment from blood” Biomed Microdevices. 13 (1) 203-13, 2011 February.

However, in the CTC detection method described in Patent Document 1 or Non-patent Documents 1 and 2 described above, a series of pretreatments related to labeling with an antibody must be performed, which takes time and effort. Therefore, it has been desired to provide a method capable of detecting CTC easily and quickly with high accuracy.
In addition, it has been desired to provide a method capable of detecting a desired test target in blood simply, quickly, and with high accuracy, as well as CTC.

  The present invention provides a mass spectrometer, a mass spectrometry method, a mass spectrometry system, and a mass spectrometry analysis for detecting a test target in blood, which can detect the test target in blood simply and quickly with high accuracy. Provide a program.

As means for solving the above-mentioned problems, a mass spectrometer according to the present invention comprises an ionization means for ionizing a component to be examined with respect to a test object in blood of a subject;
And analysis means for mass analyzing ions generated by the ionization means.
Further, a mass spectrometry system of the present invention is a mass spectrometer of the present invention,
Information of at least one of mass spectrum information created based on mass spectrometry results obtained by the analysis means, and mass spectrum analysis result information obtained by learning the mass spectrum information To determine whether the state of presence of the test object in blood exceeds a predetermined tolerance or whether the mass spectrum exhibits a disease-positive pattern, by using
When it is determined that the presence state of the test object exceeds the predetermined allowable range or the mass spectrum exhibits a disease-positive pattern, the subject is derived from the presence state of the test object And an analysis device having determination means for determining the possibility of falling under the disease.

  According to the present invention, a mass spectrometer, a mass spectrometry method, a mass spectrometry system, and a mass spectrometry for detecting a test target in blood, which can detect the test target in blood simply and quickly with high accuracy. An analysis program can be provided.

FIG. 1 is a schematic view showing an example of a mass spectrometer according to the first embodiment. FIG. 2A is a schematic view showing an example of a mass spectrometer according to the second embodiment. FIG. 2B is an enlarged photograph showing an example of the tip of the first tubular member. FIG. 3: is a photograph which shows an example of the schematic block diagram which shows the use condition of the mass spectrometer which concerns on 2nd Embodiment. FIG. 4A is a schematic view showing how collection of an inspection target is performed by automation in the mass spectrometer according to the second embodiment of FIG. 3. FIG. 4B is a schematic view showing how collection of an inspection object is performed by automation in the mass spectrometer according to the second embodiment of FIG. 3. FIG. 5 is a schematic view showing an example of a mass spectrometer according to the third embodiment. FIG. 6 is a schematic view showing another example of the mass spectrometer according to the third embodiment. FIG. 7 is a schematic view illustrating the process of separating blood into mononuclear cells. FIG. 8 is a schematic diagram for explaining information derived from mass spectrum information used by the determination means of the analysis apparatus. FIG. 9 is a block diagram showing an example of the analysis apparatus of the present invention. FIG. 10 is a block diagram showing an example of the data structure. FIG. 11 is a flowchart showing an example of the flow of processing in the analysis apparatus of the present invention. FIG. 12 is a flowchart showing another example of the flow of processing in the analysis apparatus of the present invention. FIG. 13 is a diagram showing an example of measurement data (text format). FIG. 14 is a diagram showing an example in which the data shown in FIG. 13 is expressed in the form of mass spectrum. FIG. 15 is a diagram showing an example of a chromatogram. FIG. 16 is a diagram showing an example of a representative mass spectrum created by data in a designated time zone of the chromatogram of FIG. FIG. 17 is a view showing an example of an input screen (condition setting screen) in automatic (semi-automatic) creation of a representative mass spectrum. FIG. 18 is a view showing an example of a chromatogram and a representative mass spectrum displayed in automatic (semi-automatic) creation of a representative mass spectrum. FIG. 19 is a diagram showing an example of an input screen in registration of the created representative mass spectrum. FIG. 20 is a view showing an example of a label input screen in labeling of the created representative mass spectrum. FIG. 21 is a view showing an example of a setting screen for outputting mass spectrum data to be subjected to statistical analysis. FIG. 22 is a diagram showing an example of a setting screen in internal symbol determination. FIG. 23 is a diagram showing an example of an input / output screen for determination of an internal label and processing of mass spectrum evaluation by the determined internal label. FIG. 24 is a diagram showing an example of a setting screen (input screen) of statistical analysis processing. FIG. 25 is a diagram showing an example of a file output by marker search by a test method. FIG. 26 is a diagram showing an example of a result (score plot) reduced to two principal components by the reduction method. FIG. 27 is a diagram showing an example of the errata indicating the result of the verification method. FIG. 28 is a diagram showing an example of an output screen that summarizes the results of statistical analysis and verification. FIG. 29 is a diagram showing an example of the result of a test for detection of cancer cells in blood using the mass spectrometry system of the present invention. FIG. 30 is a diagram showing an example of the result of a test for detection of cancer cells in blood using the mass spectrometry system of the present invention.

(Mass spectrometry system)
The mass spectrometry system of the present invention has a mass spectrometer and an analyzer, and further has other devices as needed.
The profile (analytical result) of the test object in blood can be obtained simply and quickly with high accuracy for the test object in blood by the mass spectrometer in the mass spectrometry system of the present invention.
In particular, even if the test object is not subjected to the conventional pretreatment for labeling in which an antibody is added, the test object in the blood can be tested by directly ionizing the test object and mass analyzing the generated ions. Profile (analysis results) can be obtained.
Further, by analyzing the mass spectrum obtained as a result of mass analysis by the mass analysis system of the present invention, the existence state in the blood to be inspected can be determined with high accuracy, and as a result, the existence of the inspection object It is possible to determine the possibility of corresponding to the disease derived from the condition.
That is, the mass spectrometry system of the present invention is simple, quick, and easy by acquiring the profile (analysis result) possessed by the test object itself without the pretreatment of adding the antibody (labeling the molecule to be detected). It is possible to detect a test object in blood with high accuracy.
For example, when the test object is CTC, it is possible to ionize CTC directly without labeling with antibody, and to use the profile (analysis result) of the CTC cell itself, for simple, fast and high precision. CTC detection can be performed. In addition to CTC, for example, when the test object is a lipid, the mass spectrometry system of the present invention detects the presence state of the lipid in the blood, whereby the subject is a lipid such as hyperlipidemia. It is possible to determine the possibility of corresponding to a disease whether or not it is a disease derived from Also, for example, when the test object is a carbohydrate, the subject is a disease derived from a carbohydrate such as diabetes by detecting the state of the carbohydrate in the blood using the mass spectrometry system of the present invention. It is possible to determine the possibility of falling under the presence or absence of the disease.

(1. Mass spectrometer and mass spectrometry method)
The mass spectrometer of the present invention comprises an ionization means and an analysis means, and further comprises other means as required.
The mass spectrometric method of the present invention includes an ionization step, an analysis step, and further includes other steps as necessary.

  The mass spectrometric method of the present invention can be suitably carried out by the mass spectroscope of the present invention, the ionization step can be carried out by an ionization means, the analysis step can be carried out by an analysis means, and the other steps are other It can be done by means of

<Ionization means and ionization step>
The ionization step is a step of ionizing a component to be examined with respect to a test target in blood of a subject, and is implemented by an ionization means.
Here, as the subject, all animals including human are considered.

<< Inspection target >>
The subject of mass spectrometry of the present invention is a test subject present in blood.
Here, the blood contains blood cell components (cellular components, blood cells), platelets, plasma components (liquid components) and the like.
The blood cell components include red blood cells, white blood cells, platelets and the like. The white blood cells include neutrophils, eosinophils, basophils, lymphocytes, monocytes and the like. The lymphocytes include NK cells (natural killer cells), B cells (B lymphocytes), T cells (T lymphocytes) and the like.
The plasma components include water, proteins, fats, sugars, inorganic salts, low molecular weight metabolites, foreign molecules and the like.

The test object is not particularly limited as long as it is a biomarker in blood, and can be appropriately selected according to the purpose. For example, cells, amino acids, peptides, proteins, nucleic acids, oligonucleotides (oligo DNA), Oligoribonucleotides (oligo RNA), lipids, carbohydrates, vitamins, electrolytes, low molecular weight metabolites and the like can be mentioned. Here, a biomarker refers to a biological indicator, and refers to a substance to be measured in blood which can recognize the presence or progression of a certain disease.
Examples of the cells to be examined include circulating tumor cells in blood (CTC), red blood cells, white blood cells and the like.
Examples of the protein to be tested include albumin, hemoglobin, γ-globulin, fibrinogen, antithrombin III, transferrin, ceruloplasmin, cytokines including growth factors, chemokines and the like.
Examples of the lipid to be examined include neutral fat, HDL cholesterol, LDL cholesterol and the like.
Examples of the carbohydrate to be tested include glucose and 1.5AG (1.5 anhydroglycitol).
Examples of the nucleic acid to be tested include circulating normal DNA, circulating tumor DNA, non-coding RNA (miRNA, transfer RNA, ribosomal RNA, etc.) and the like.

The test target may be fructosamine, HbA1c (glycohemoglobin), etc. in which a protein and a carbohydrate are combined.
Moreover, an enzyme, a hormone, a tumor marker etc. may be sufficient as said test object. Here, the tumor marker refers to a substance that is an index of a malignant tumor.
Examples of the enzyme include γ-GTP, AST (aspartate aminotransferase), ALT (alanine aminotransferase), amylase and the like.
Examples of the hormone include adrenocorticotropic hormone, luteinizing hormone, follicle stimulating hormone, prolactone, thyroid hormone, parathyroid hormone, aldosterone, insulin, estrogen, progesterone, growth hormone and the like.
Examples of the tumor marker include AFP (α-fetoprotein), CA19-9 (serial Lewis A saccharide), CA125 (glycoprotein), CEA (carcinoembryonic antigen), PSA (prostate specific antigen), SSC (squamous epithelium) Cancer related antigens) and the like.

  It is preferable that the test subject is not added with an antibody (except for an antibody derived from the same species as the subject). By not performing pretreatment for adding an antibody, detection of a test object can be performed simply and rapidly. For example, when detecting CTCs in human blood, it is preferable to apply an ionization step to CTCs to which no antibody (except for human antibodies) has been added.

As a sample for use in mass spectrometry, blood collected from a subject may be used as it is, or the collected blood may be subjected to mononuclear cell separation treatment to separate a mononuclear cell layer, and the mononuclear cell layer may be separated. May be used.
The mononuclear cell separation treatment method is not particularly limited as long as the mononuclear cell layer can be separated, and can be appropriately selected according to the purpose. For example, Ficoll (Ficoll-Paque PLUS, manufactured by GE Healthcare Japan Co., Ltd.) And a general separation method using a medium for separating lymphocytes.
When blood is separated by Ficoll, it is separated into each layer, for example, as shown in FIG. Therefore, it is preferable to use a separated layer as an analyte sample used in the mass spectrometric method of the present invention.
For example, when the test object is CTC, it is preferable to use a mononuclear cell layer among the separated layers.
In FIG. 7, the symbol a indicates a separation solution (for example, a lymphocyte separation solution (specific gravity: 1.119), b indicates blood, c indicates plasma, d indicates mononuclear cells (lymphocytes, monocytes), platelets And e shows erythrocytes and granulocytes.

  When the collected blood is used as a sample sample as it is, the time required from the collection of the blood to the ionization is a short time. However, when mononuclear cell separation processing is performed, the processing time for mononuclear cell separation is added, but even in this case, the time required from the collection of blood to the ionization is approximately 60 minutes or less is there. This is a much shorter time than the processing time conventionally required for pretreatment, for example, addition of antibody for detection of CTC.

It is preferable to use the collected blood as it is as a sample sample because the treatment is simple and the time is short. On the other hand, it is preferable to separate the mononuclear cell layer from blood and use the mononuclear cell layer as a sample sample from the viewpoint of more accurate detection. Therefore, whether or not the mononuclear cell separation process is to be performed may be appropriately determined in consideration of the type of test object and the specific form of the ionization means (the specific form of the ionization means will be described later). .
For example, when the mononuclear cell separation process is performed and the separated mononuclear cell layer is used as a sample, the following <first embodiment> or the <second embodiment> may be used as a mass spectrometer of the present invention. It is possible to use a mass spectrometer having the ionization means described in the section above.
Also, for example, when using the collected blood as it is as a sample sample without performing mononuclear cell separation processing, the mass spectrometer of the present invention is described in the following <Third Embodiment>. A mass spectrometer having an ionization means can be used.

Examples of the ionization means include means using ambient ionization. Ambient ionization is a method of direct gas phase ionization of an analyte sample under normal environment (for example, normal temperature, atmospheric pressure environment).
In the present invention, electrospray ionization, which is one of ambient ionization methods, is preferably mentioned. Moreover, the line type ionization method described below is also preferably mentioned.
The following case can be mentioned as a specific aspect of the said ionization means.
(1) Ionizing means using an aspect having a needle-like member, a power source, and an ion collecting member (hereinafter referred to as “Probe Electrospray Ionization” (hereinafter sometimes referred to as “PESI”)) (2) Ionization means using an embodiment having a first tubular member, a second tubular member, a power supply, and an ion induction member (hereinafter referred to as remote sampling electrospray ionization), 3) There is an ionization means using an embodiment having a linear member and an ionization member (hereinafter, referred to as a line type ionization method).

In all of the ionization means (1) to (3), the time from ionizing the sample to obtaining the judgment result of mass spectrometry may be very short.
With the mass spectrometer of the present invention, for example, when the sample is ionized, the judgment result of mass spectrometry can be obtained within about 5 minutes, and real-time analysis becomes possible.

<< Ionization means using probe electrospray ionization method >>
An ionization means using a probe electrospray ionization (PESI) method has a needle-like member, a power source, and an ion collecting member, and may have other members as necessary.
The ionization means using the PESI method can perform real-time mass analysis because ionization is performed immediately after the sample sample is collected by the needle-like member.

  As an ionization means using such a PESI method, for example, those disclosed in WO 2010/047399 and the like can be used.

The needle-like member is used to contact or pierce the tip of a sample to collect a test sample or a test target contained in the sample.
The power source is used to apply a voltage to the needle-like member in order to generate by electrospray the sample sample taken from the tip of the needle-like member and ions derived from the test object.
The ion collection member is used to collect the generated ions.

In the ionization means using the PESI method, the needle-like member is moved to cause the tip of the needle-like member to contact or pierce the sample sample, and a part of the sample sample is captured on the tip of the needle-like member.
If the tip of the needle-like member is made extremely thin, a very small range of sample sample is captured on the tip of the needle-like member on the surface of the sample specimen.
When a high voltage is applied to the needle-like member having the sample sample attached to the tip, a strong electric field acts on the sample sample at the tip of the needle-like member, and molecules of the sample sample are ionized while being detached by the electrospray method.

  Among the advantages of the ionization means using PESI method are: (i) measurement is possible under atmospheric pressure, (ii) very small amounts of sample are sufficient (several picoliters), (iii) minimally invasive (Iv) Pre-treatment of test object is unnecessary, (v) Insensitive to salts and surfactants, (vi) Real-time analysis is possible.

-Needle-like member-
The needle-like member is a member that brings the sample into contact with or pierces the tip of the sample to collect (capture) the sample.
There is no restriction | limiting in particular about the shape of a needle-like member, a structure, a magnitude | size, a material, etc. According to the objective, it can select suitably.
The shape of the needle-like member can be a tip shape that can cause an electrospray phenomenon.
Examples of the structure of the needle-like member include a solid structure, a single layer structure, and a multiple layer structure.
The size of the needle-like member (including the thickness of the needle and the like) can be appropriately selected according to the amount of the sample to be captured at the tip and the type of the test object.
The material is preferably durable, less susceptible to contamination by bacteria, etc., preferably sterilizable, preferably stainless steel, titanium, tungsten, nickel, aluminum, platinum, gold, silver or those plated with these. Furthermore, as surface treatment of these, hydrophilicity or hydrophobicity may be adjusted by electric field polishing, acid treatment, chemical treatment.
Specifically as a needle-like member, what processed the stainless steel needle | hook etc. can be used.

-Electrode-
The electrode is a member that applies a voltage to the needle-like member in a state in which the sample is attached to the tip of the needle-like member.
The standard of voltage is 2 kV to 3 kV, and electrospray can be generated by applying a high voltage to the needle-like member.

-Ion collection member-
The ion collection member is a member which is provided in the vicinity of the tip of the needle-like member, and collects ions generated from a sample sample or a test object attached to the needle-like member.
As the ion collection member, for example, an ion collection and transport tube can be used. The inner diameter of the ion collection and transport tube is preferably 1 mm to 5 mm.
The opening at the tip of the ion collection and transport tube can be formed in a size and shape that does not interfere with the movement of the needle-like member, and the collection efficiency of the ions should be such that the opening at the tip is flared It is preferable from the point of
There is no restriction | limiting in particular about the shape of an ion collection member, a structure, a magnitude | size, a material, etc. According to a test substance sample, it can select suitably.
At this time, by setting the opening of the ion collecting / transporting tube in a position close to the tip of the needle-like member, the generated ions can be efficiently sucked.

-Other components-
There is no restriction | limiting in particular as another member, According to the objective, it can select suitably, For example, a control member etc. are mentioned.
There is no restriction | limiting in particular as a control member if it is a member which can control movement of each member, According to the objective, it can select suitably, For example, apparatuses, such as a computer, etc. are mentioned.

<< Ionization means using remote sampling electrospray ionization method >>
The ionization means using the remote sampling electrospray ionization method comprises a first tubular member, a second tubular member, a gas flow generating member, a power supply and an ion guiding member, and the other as required Of the

The first tubular member is used to contact an analyte sample at one end and collect an examination object contained in the analyte sample or the analyte sample. Further, it is used for generating the sample derived from the other end and ions derived from the test object by electrospray.
The second tubular member is used to concentrically insert the first tubular member.
The gas flow generating member is used to flow gas to the second tubular member at high speed.
The power supply is applied to the counter electrode such that a potential difference is generated between the other end of the first tubular member and a counter electrode opposed to the other end disposed in the region where the electrospray is generated. Used to apply a voltage.
The ion guide member is used to guide the generated ions to the mass analysis means.

In the ionization means using the remote sampling electrospray ionization method, one end of the first tubular member is brought into contact with the analyte sample, and the one end of the first tubular member captures the analyte sample.
The first tubular member is concentrically inserted into the second tubular member. Gas flows at high speed in the gap between the inside of the second tubular member and the outside of the first tubular member.
The high velocity gas flow is generated by the gas flow generating member.
The pressure around the other end of the first tubular member is reduced due to the venturi effect generated by the high velocity gas flow in the second tubular member. The pressure difference between one end of the first tubular member and the other end of the first tubular member aspirates the sample, and the sample collected from one end of the first tubular member is directed to the other end of the first tubular member. Moving.
The high velocity gas flow in the second tubular member exerts a suction action on the specimen sample and assists in the spraying of the charged droplets, which also contributes to stable ion generation.

A voltage is applied to the counter electrode such that a potential difference is generated between the other end of the first tubular member and the counter electrode disposed at a position opposite to the other end in the region where the electrospray is generated. When a strong electric field acts on the sample at the other end, the molecules of the sample are ionized while being separated by the electrospray method.
For example, the other end of the first tubular member is electrically held at ground or any potential suitable for the analyte sample. Since a potential difference is generated between the other end of the first tubular member and the counter electrode and an electrospray is generated when a high potential is applied to the other end, for example, the first tubular member is set to the potential of the ground, A negative potential of about -3 kV is applied to the counter electrode. Then, since the first tubular member is safe to touch and the high potential is applied to the other end, an electrospray is generated.
The generated ions are transported to the mass spectrometer via the ion guide member.
In a mass spectrometer using remote sampling electrospray ionization, mass spectrometry results are obtained almost immediately when the first tubular member comes in contact with an analyte sample.

  An advantage of the ionization means using the remote sampling electrospray ionization method is that, in addition to the advantage of the ionization means using the PESI method described above, a large number of samples can be measured continuously. Assuming that the sample sample is obtained and the operation of spraying ions by electrospray is performed on only one side, the step of obtaining the sample sample is included between the ionization steps, and these steps are repeated. On the other hand, when acquiring a sample from one side and immediately spraying ions by electrospray from the other side, sample samples may be continuously obtained without interruption by ion spraying. it can. In synchronization with the acquisition of the specimen sample, ions are sprayed on the other side and instantaneously subjected to mass analysis, so measurement of the specimen sample becomes possible continuously. This enables more rapid measurement.

In addition, if a mass spectrometer using the remote sampling electrospray ionization method of the present invention is used, the ionization step can be automated for a plurality of samples.
For example, as shown in FIG. 3, a 3-axis motorized automatic sampling device can be constructed.
The sample stage is driven by two linear actuators in the horizontal (xy) axis. The position of the first tubular member, which is the sampling probe, is controlled by another linear actuator in the vertical (z) axis.
The sample stage has a 96-well sample plate on which a plurality of sample samples are disposed.
As shown in FIG. 4A, when the first tubular member is moved to the lower position and brought into contact with the sample of the sample, mass spectrometry results are obtained almost immediately by the mass spectrometer.
Next, as shown in FIG. 4B, the first tubular member is moved to the upper position and separated from the sample sample. Change the position of the 96 well sample plate.
Further, as shown in FIG. 4A, the first tubular member is moved to the lower position and brought into contact with the next sample to obtain a mass analysis result of the sample.
By controlling this repeated process with a computer, it is possible to construct a continuous sampling automated mass spectrometer.

-First tubular member-
The first tubular member is a member which brings the test object into contact with one end to collect the test object, and generates the sample derived from the other end and ions derived from the test object by electrospray.
The shape, structure, size (length, thickness, etc.), material, etc. of the first tubular member are not particularly limited, and may be appropriately selected in consideration of the amount of sample to be passed and the type of sample to be passed. be able to.
However, the length of the first tubular member should be short to a certain extent so that the sample can be easily aspirated and the ions are easily sprayed by electrostatic action and capillary action. In addition, the thickness of the first tubular member should be thick enough to prevent clogging of the sample.
The length of the tube of the first tubular member is preferably 1 cm to 5 cm. The inner diameter of the first tubular member is preferably about 0.1 mm, and the outer diameter is preferably about 0.2 mm.
The material is preferably durable, less susceptible to contamination by bacteria, etc., and is preferably sterilizable, preferably stainless steel, titanium, tungsten, nickel, aluminum, platinum, gold, silver or those plated with these. Can be mentioned.

-Second tubular member-
The second tubular member is a member for concentrically enclosing the first tubular member.
Gas flows at high speed in the gap between the inside of the second tubular member and the outside of the first tubular member.
The shape, structure, size (length, thickness, etc.), material, and the like of the second tubular member are not particularly limited and may be appropriately selected. However, the inner diameter of the second tubular member is not particularly limited. The outer diameter of the tubular member and the amount and speed of the gas to be passed may be taken into consideration and appropriately selected.
Preferably, the other end of the first tubular member slightly protrudes from the second tubular member so as not to cause backflow of gas. Therefore, for example, the length of the second tubular member may be appropriately selected so that the other end of the first tubular member protrudes from the second tubular member by about 0.5 mm to 1 mm.
The length of the tube of the second tubular member is preferably 1 cm to 3 cm. The inner diameter of the second tubular member is preferably 0.3 mm to 0.5 mm, and the outer diameter is preferably about 1.6 mm.
The material is preferably durable, less susceptible to contamination by bacteria, etc., and is preferably sterilizable, preferably stainless steel, titanium, tungsten, nickel, aluminum, platinum, gold, silver or those plated with these. Can be mentioned.

-Electrode-
The electrode applies a voltage to the opposite electrode so that a potential difference is generated between the other end of the first tubular member and the opposite electrode opposed to the other end disposed in the region where the electrospray is generated. It is a member.
For example, in the case where the first tubular member is set to the ground potential, when a negative potential of about 3 kV is applied to the opposite electrode, a high voltage is applied to the other end of the first tubular member, causing electrospray. It can be done.

-Gas flow generating member-
The gas flow generating member is not particularly limited as long as it is a device for compressing the gas to increase the pressure and continuously feeding it, and can be appropriately selected according to the purpose. The type of gas to be handled, the compression method, the structure of the apparatus, and the like can be appropriately selected.
For example, an air compressor can be used as the gas flow generating member. This allows, for example, high velocity air flow to be generated in the second tubular member. The moving speed of the sample can be adjusted by the thickness of the first tubular member and the flow velocity of the air flowing in the second tubular member. Therefore, the flow rate of air may be adjusted so that the moving speed of the sample is in a desired range, and various conditions of the gas flow generating member may be set so that the flow rate of air is in a desired range.
The air flow velocity in the second tubular member is preferably, for example, 2 L / min to 5 L / min.

-Ion induction member-
The ion guide member is a member for guiding the generated ions to the mass analysis means.
As an ion induction member, an ion transport tube can be used, for example.
There is no restriction | limiting in particular about the shape of an ion transport member, a structure, a magnitude | size, a material, etc. According to a test object, it can select suitably.
As a material of the ion transport member, for example, a flexible polymer tube can be selected.
The length of the ion transport tube is preferably 50 cm to 100 cm. Moreover, it is preferable that an internal diameter is 1 mm-5 mm.

-Other components-
The other members are as described above in the section of << Method Using Probe Electrospray Ionization Method >>.

<< ionization means using line type ionization method >>
The ionization means using the line ionization method has a linear member and an ionization member, and may have other members as necessary.
The linear member is used to transport the droplets of the sample solution using the sample solution prepared using the limiting dilution method. Here, the sample liquid refers to a liquid sample (sample) containing a test sample or a test target contained in the test sample.
The ionization member is used to generate ions derived from the inspection object from the droplets attached to the linear member.

In order to cause droplets of the sample solution to adhere to the linear member, the sample solution supply element is used to drop the sample solution onto the linear member in units of droplets. Preferably, one droplet is dropped onto the linear member. The description of the sample liquid supply element will be described later.
It is preferable that the sample solution be prepared using a limiting dilution method. This makes it possible to drop one cell / droplet and obtain a profile for each cell (here, the result of mass spectrometry).
The droplets of the dropped sample liquid are transported to the ionization member by moving the linear member by a driving member (for example, a roller).
The droplets attached to the linear member by the ionization member are ionized, and the generated ions are immediately subjected to mass analysis.

-Linear member-
There is no restriction | limiting in particular about the material of a linear member, a shape, a structure, a magnitude | size, etc. According to a test object, it can select suitably.
As a material of a linear member, a metal wire, a natural fiber, a chemical fiber etc. are mentioned, for example.

  Examples of the metal wire include stainless steel, titanium, tungsten, piano wire, cobalt-based alloy wire, alloy wire exhibiting pseudoelasticity (including super elastic alloy), steel wire, brass wire, copper wire, aluminum wire, etc. Be

  Natural fibers include, for example, vegetable fibers such as cotton (cotton), hemp, manila hemp, palm, and IGSA, and wool such as wool, silk (silk) down, feather, etc. Natural fiber yarns are used.

  Chemical fibers include regenerated fibers such as rayon, cupra and casein fibers, semi-synthetic fibers such as acetate, triacetate and promix, polyamides, polyvinyl alcohol, polyvinyl chloride, polyvinylidene chloride, polyesters, polyacrylonitrile The synthetic fiber of a system, polyolefin type, polyether ester type, polyurethane type etc., etc. are mentioned, The chemical fiber yarn which spun these is used.

-Ionizing member-
The ionization member is a member that generates ions from droplets of the sample liquid attached to the linear member by the ionization method.
The ionization method is not particularly limited and can be appropriately selected according to the purpose. For example, APCI (Atospheric Pressure Chemical Ionization) method, EI (electron ionization) method, CI (chemical ionization) Method, DEI (desorption electron ionization) method, DCI (desorption chemical ionization) method, FAB (fast atom bombardment) method, FRIT-FAB (FRIT-fast atom bombardment; Frit fast atom bombardment) method, ESI (electrosp Examples include ray ionization (electrospray ionization) and MALDI (matrix-assisted laser desorption ionization). Among these, the APCI method and the ESI method are particularly preferable.

-Other components-
The other members are as described above in the section of << Method Using Probe Electrospray Ionization Method >>.
In addition, the ionization means using the line ionization method may further include a sample liquid supply element, a driving member, a pretreatment member, and a post-treatment member.

-Sample liquid supply element-
The sample liquid supply element is not particularly limited as long as the sample liquid can be dropped drop by drop, and can be appropriately selected according to the purpose. For example, a piezo terminal can be used.

-Drive member-
The driving member is not particularly limited as long as the linear member can be moved in a predetermined direction, and can be appropriately selected according to the purpose. Examples thereof include a conveying roller.
Adjustment of the moving speed of the linear member by the drive member can be appropriately performed.

-Pretreatment member-
The sample liquid can be pretreated by utilizing the time from the attachment of the sample liquid droplets to the linear member by the sample liquid supply element to the ionization. Examples of this pretreatment include deproteinization and salt removal treatment.

-Post-processing member-
The linear member having undergone the ionization step may be subjected to post-treatment while moving from the ionization member to the sample liquid supply element. Examples of the post-treatment include various treatments such as washing, heating, and chemical treatment for washing the linear member.

<Analytical means and analysis process>
The analysis step is a step of mass analyzing ions generated by the ionization step, and is performed by an analysis means.

There is no restriction | limiting in particular as a mass spectrometer used for mass spectrometry, According to the objective, it can select suitably, For example, (orthogonal type) time-of-flight mass spectrometer, (linear) ion trap mass spectrometer, four Examples thereof include a polar mass spectrometer, a Fourier transform mass spectrometer, and a spectrometer combining these mass spectrometers as appropriate.
In mass spectrometry, an analyte sample is ionized and its ions are separated according to the difference in mass-to-charge ratio (m / z). Therefore, from the mass spectrometer, the mass-to-charge ratio is plotted on the horizontal axis, and the data on the basis of the ion intensity plotted on the vertical axis is output as raw data.

Here, an embodiment of the mass spectrometer of the present invention will be described in detail with reference to the drawings.
In the drawings, the same components may be denoted by the same reference symbols and redundant description may be omitted. Further, the number, positions, shapes, and the like of the following constituent members are not limited to those in the present embodiment, and can be set to a preferable number, positions, shapes, and the like in practicing the present invention.

First Embodiment (Embodiment of Ionizing Means Using PESI Method)>
FIG. 1 is a schematic view showing an example of a mass spectrometer having an ionization means using a probe electrospray ionization method as a first embodiment.
The mass spectrometer 100 of FIG. 1 includes an ionization unit 101 and an analysis unit 102.

The ionization means 101 includes a needle member 1, a power source 7, a personal computer (PC), and an ion collection member 8. In FIG. 1, 5 is a linear actuator and 11 is a suction device. Further, in FIG. 1, 2 indicates that the needle-like member 1 moves up and down, 3 indicates a sample (sample), 4 indicates a state in which ions are generated by electrospray, and 9 indicates generated ions. Show.
In the first embodiment, a needle made of stainless steel is used as the needle-like member 1.
The needle-like member 1 is vertically movable with respect to the sample sample by the operation of the linear actuator 5.
When the needle-like member 1 moves to the lower position (lower end), the sample of the sample adheres to the tip of the needle-like member 1 by contacting or piercing the sample.
When the needle-like member 1 moves to the upper position (upper end), a high voltage is applied from the power source 7 and electrospray is generated from the tip of the needle-like member 1.

As the ion collecting member 8, an ion collecting / transporting tube with an inner diameter of 1 mm to 5 mm (made of a flexible resin whose inner surface is subjected to a smoothing process to prevent charging) is used. Aspirate the ions generated by the spray.
The aspirated ions are conveyed to the mass spectrometer 12 via the ion opening 10.
Since a voltage is applied between the tip of the needle-like member 1 and the opening of the ion collection and transport tube 8, generated ions are attracted at the opening of the ion collection and transport tube 8.

  Reference numeral 12 in FIG. 1 denotes a mass spectrometer, which analyzes ions generated by the ionization means 101 in real time. In the first embodiment, any one of a quadrupole mass spectrometer, a Fourier transform mass spectrometer, an ion trap mass spectrometer, and a time-of-flight mass spectrometer is used as a mass spectrometer.

  In PESI, although the linear actuator 5 and the power supply 7 are controlled in synchronization by a controller (not shown), the use of the controller is omitted by adopting a configuration in which a voltage is applied from the power supply only when not sampling. can do. Thus, the device can be simplified, electric shock can be prevented, and safety can be improved.

Next, the operation at the time of ionization will be described.
When the position of the needle-like member 1 is lowered, the tip of the needle-like member 1 contacts or pierces the sample. When the position of the needle-like member 1 is raised, a voltage is applied. As a result, from the tip of the needle-like member 1, ions derived from the sample sample or the test object are generated. The generated ions are immediately aspirated from the opening of the ion collecting and transporting tube as the ion collecting member 8 and carried to the mass spectrometer 12.

Second Embodiment (Embodiment of Ionization Means Using Remote Sampling Electrospray Ionization)>
FIG. 2A is a schematic view showing an example of a mass spectrometer having an ionization means using remote sampling electrospray ionization as a second embodiment.
The mass spectrometer 200 of FIG. 2A includes an ionization unit 201 and an analysis unit 202.

The ionization means 201 includes a first tubular member 21, a second tubular member 22, a gas flow generating member 25, a power source 20, and an ion guiding member 28. In FIG. 2A, 18 is a specimen sample (sample), 19 is a generated positive ion, 23 is a holding member of the tubular member, 24 is the other end of the first tubular member 21 from which ions are generated by electrospray, 26 is Reference numeral 27 denotes an insulating member, and 29 denotes a mass spectrometer.
Moreover, the enlarged photograph of the one end side of the 1st tubular member 21 of FIG. 2A is shown to FIG. 2B.
Stainless steel is used as the first tubular member 21.
The holding member 23 of the tubular member and the counter electrode 26 are insulated by the insulating member 27. In the second embodiment, polyetheretherketone (PEEK) is used as the insulating member 27.
The first tubular member 21 can move up and down with respect to the sample.
The first tubular member 21 is concentrically inserted into the second tubular member 22.
In the gap between the inside of the second tubular member and the outside of the first tubular member, high-speed air flow is generated by the gas flow generating member (here, an air compressor is used).
In the second embodiment, the outer diameter of the first tubular member 21 is 0.2 mm, and the inner diameter of the second tubular member 22 is 0.5 mm. The gap between the first tubular member 21 and the second tubular member 22 is 0.15 mm, and the air flows at high speed through the gap.

When the first tubular member 21 is moved to the lower position (lower end), one end of the first tubular member 21 comes into contact with the analyte sample, whereby the analyte sample is injected into the first tubular member 21.
Due to the venturi effect generated by the high velocity air flow in the second tubular member 22, the aspiration of the sample is smoothly performed, and the sample collected from one end of the first tubular member 21 is the first tubular. Move to the other end 24 of the member.
Power is supplied to the counter electrode so that a potential difference is generated between the other end 24 of the first tubular member 21 and the counter electrode 26 disposed at a position opposite to the other end 24 in a region where electrospray is generated. A voltage is applied from 30. In the second embodiment, the first tubular member 21 is set to the ground potential, and a voltage of −3 kV is applied to the counter electrode 26. Thereby, electrospray is generated from the other end 24 of the first tubular member 21.
The generated ions are supplied to the mass spectrometer 29 via the ion guide member 28 in the flow of the air flow. Here, in the second embodiment, a flexible polymer tube with an inner diameter of 3 mm is used as the ion guide member 28.
When the first tubular member 21 is moved to the upper position (upper end), the collection of the sample is stopped.

  The mass spectrometer 29 analyzes the ions generated by the ionization means 201 in real time. In the second embodiment, any one of a quadrupole mass spectrometer, a Fourier transform mass spectrometer, an ion trap mass spectrometer, and a time-of-flight mass spectrometer is used as a mass spectrometer.

<Third Embodiment (embodiment of ionization means using line ionization method)>
FIG. 5 is a schematic view showing an example of a mass spectrometer having an ionization means using a line ionization method as a third embodiment.
The mass spectrometer 300 of FIG. 5 includes an ionization unit 301 and an analysis unit 302.

The ionization means 301 includes a sample solution supply element 31, a droplet 32 of a sample solution which is a sample, a linear member 33, a drive member 34, and an ionization member 35.
The analysis means 302 comprises a mass spectrometer 36.
The linear member 33 transports the droplet 32 of the sample liquid.
In order to attach the droplets of the sample solution to the linear member, the sample solution supply element 31 is used to drop one droplet at a time onto the linear member. In the third embodiment, a piezo terminal is used as a sample liquid supply element.
The linear member 33 is moved by the drive roller which is the drive member 34, and the droplet of the sample liquid is transported to the ionization member 35.

The ionization member 35 is used to ionize the droplets 32 attached to the linear member. In the third embodiment, as the ionization member 35, ESI or APCI is switched and used depending on the target molecule.
The sample solution generated by the ionization member 35 and the ions derived from the test object are immediately subjected to mass analysis.

  The mass spectrometer 36 analyzes the ions generated by the ionization means 301 in real time. In the third embodiment, any one of a quadrupole mass spectrometer, a Fourier transform mass spectrometer, an ion trap mass spectrometer, and a time-of-flight mass spectrometer is used as a mass spectrometer.

(2. Analysis device and analysis method)
The analysis apparatus of the present invention has determination means and further has other means as needed.
The analysis method of the present invention includes a determination step, and further includes other steps as necessary.

The analysis method of the present invention can be suitably implemented by the analysis device of the present invention, the determination step can be performed by the determination means, and the other steps can be performed by other means.
The processing by the mass spectrometry analysis program is executed using a computer having a control unit that constitutes the analyzer.

Based on the result of mass spectrometry obtained by the mass spectrometer of the present invention by the analyzer of the present invention, the possibility of determining whether the subject corresponds to a disease derived from the state of existence of the test object in blood It can be determined.
For example, when the test object is CTC, by detecting CTC, clinical information such as early detection and recurrence prediction of cancer and sensitivity of cancer cells to an anticancer agent can be obtained. Knowing the number of CTCs per unit of blood may lead to a diagnosis of cancer or a diagnosis of prognosis.
In addition to CTC, for example, when the test object is a lipid, if the presence state of the lipid in the blood is known, the possibility of being a lipid-derived disease such as hyperlipidemia can be determined. . Further, for example, when the test object is a carbohydrate, if the presence state of the carbohydrate in the blood is known, there is a possibility that the possibility of being applicable to a disease derived from a carbohydrate such as diabetes may be determined.

<Judgment process and judgment means>
In the determination step, at least one of mass spectrum information and information of analysis result of mass spectrum obtained by learning the mass spectrum information is used to determine the existence state of the test object in blood. , It is determined whether the predetermined allowable range is exceeded or the mass spectrum exhibits a disease-positive pattern,
When it is determined that the presence state of the test object exceeds the predetermined allowable range or the mass spectrum exhibits a disease-positive pattern, the subject is derived from the presence state of the test object It is a process of determining the possibility of falling under the disease to
Examples of the disease-positive pattern include a tumor-positive pattern.

In the analyzer of the present invention, an appropriate representative mass spectrum is created based on the raw scan data output from the mass spectrometer. The representative mass spectrum may be plural. Using this representative mass spectrum data, the shape of the entire mass spectrum data, or a marker peak specific to the object to be examined (for example, the peak indicated by the arrow in FIG. 8I) as shown in (i) of FIG. By determining these peaks, it is possible to grasp the existence state of the test object in the blood.
For example, since there is a difference in the shape of a mass spectrum between a mass spectrum indicated by a normal analyte sample not including the examination object and a mass spectrum indicated by a non-normal analyte sample including the examination object, these mass spectra By comparing the information to be shown with the mass spectrum of the measured result, it is possible to know the existence state of the test object in the measured object.
For example, the mass spectrum shown by a patient who has an abnormal test subject suffering from a disease, which is different from the shape of the mass spectrum or the marker peak shown by a healthy person, where the presence state of the test subject in blood exceeds a predetermined tolerance. When the measurement result close to the shape of the marker or the marker peak is obtained, it can be judged that the subject is highly likely to be a relevant disease.
In this case, depending on the shape of the mass spectrum and the degree of closeness of the measurement result to the information indicated by the marker peak, not only the binary judgment of whether or not the patient has the disease but also the likelihood of developing the disease. (Probability) can also be determined.

Further, in the analysis apparatus of the present invention, not only the information of the representative mass spectrum described above but also the analysis result of the mass spectrum obtained by learning the information of the mass spectrum as shown in (ii) of FIG. The information of (1) may be used to grasp the existence state of the test object in the blood.
As shown in FIG. 8, the information on the mass spectrum shown in (i) and the information on the analysis result of the mass spectrum obtained by learning shown in (ii) are used together to grasp the existence state of the test object in blood You may

  Further, the analysis method of the present invention not only determines whether or not a single measurement sample of blood of a subject falls under one disease, but also whether or not it falls under a plurality of diseases at one time. It can be determined. That is, based on the mass spectrometry results obtained for a single measurement sample of the blood of the subject by the analysis method of the present invention, it may be determined whether or not it corresponds to a plurality of diseases. In this case, the possibility of corresponding to a plurality of diseases can be determined by one analysis.

Hereinafter, a specific embodiment of determining the possibility that the subject falls under a disease derived from the state of the subject to be examined in blood using the analysis method of the present invention will be described below.
(A) For example, mass spectrometry results of blood from cancer patients are obtained in advance, and those data are analyzed to prepare an index for determining the presence of CTC in blood. Here, examples of the index include a unique marker peak of a mass spectrum, an overall peak shape of a mass spectrum, and a standard for distinguishing the distribution of profiles of healthy persons from the distribution of profiles of cancer patients.
Then, the possibility of the presence of CTC of the subject can be determined by comparing the index prepared in advance with the measurement result of the subject.
In addition, as mass spectrometry results of cancer patients prepared in advance, mass spectrometry results of primary and metastatic tumor tissues can also be used without being limited to mass spectrometry results of CTC.
(B) In addition, since there is a difference between the data shown by the mass spectrum of a healthy person and the mass spectrum data of a patient having an abnormal examination subject, data as an indicator in advance as in the above (a) Even if it does not hold, if measurement results different from the mass spectrum data shown by healthy people are obtained, it may be suspected that it may correspond to any disease.
(C) The previously prepared data stored in the database can be updated sequentially if new measurement results are obtained. By increasing the accumulated data and updating the index information, it is possible to construct a database that allows more accurate determination.
(D) In addition, the measurement results of the subject may be accumulated, and the measurement data of the subject may be compared with the past data of the subject. As a result, it is possible to follow up on the subject of examination in the subject, which can lead to early detection of the subject's disease and selection of an optimal treatment.

  As a method of preparing the mass spectrum mentioned above and analyzing the mass spectrum by learning, for example, the method disclosed in Japanese Patent Application Laid-Open No. 2016-200435 can be used.

A mass spectrum analysis method for automatically creating a representative mass spectrum appropriately based on raw scan data output from a mass spectrometer and an analysis method for analyzing the created mass spectrum using each learning method will be described in detail below. explain.
In the following description, although the measurement target is not described in the example using the test target (for example, CTC) in the blood which is the target of the present invention, the operation of the analysis system is the same. The operation of the computer system described below can be applied to the analysis device of the present invention as well.

<Specific analysis system of analysis device of the present invention>
In the analysis apparatus of the present invention, in order to create a representative mass spectrum, for example, measurement data input processing, best time zone detection processing, and representative mass spectrum creation processing are performed.
Also, in order to learn and analyze the created mass spectrum, each learning method described below is used.

<< Measurement data input process and measurement data input part >>
The measurement data input process is a process of inputting three-dimensional measurement data of mass-to-charge ratio, ion intensity, and measurement time, and is performed by the measurement data input unit.
Three-dimensional measurement data is generally referred to as an ion chromatogram obtained by a mass spectrometry unit (mass spectrometer of the present invention), and indicates the relationship between mass-to-charge ratio and ion intensity at predetermined intervals (scanning intervals). Data (data that can be expressed as a spectrum) is obtained. Therefore, the measurement time can be expressed by the number of scans.

  Three-dimensional measurement data is generally converted into data (chromatogram) representing the total ion intensity on a time axis (scan number axis). The chromatogram includes total ion intensity (TIC) integrated over all mass-to-charge ratios in the measured data, separate ion intensity (MIC) integrated ion intensities over a specific range of mass-to-charge ratio, specific There is a single peak ion intensity (EIC) that indicates the ion intensity of the mass-to-charge ratio. The specific range of mass-to-charge ratio in MIC and the specific peak in EIC may be specified by the user, or the range or peak showing the maximum value may be determined automatically. These ranges and specific peaks may be changed to perform trial and error.

<< Best time zone detection processing and best time zone detection unit >>
The best time zone detection process is a process of calculating a time zone in which the sum of ion intensities is maximum with respect to the designated mass-to-charge ratio of the input measurement data, and is implemented by the best time zone detection unit.

<< Representative mass spectrum creation processing and representative mass spectrum creation part >>
The representative mass spectrum creation process is a process for creating a representative mass spectrum based on the ion intensity of the input measurement data in the best time zone detected, and is performed by the representative mass spectrum creation unit.

Based on the measurement data of the time zone that is considered to be the most stable and the best representative of the characteristics of the sample, by the best time zone detection process and the best time zone detection unit and the representative mass spectrum preparation process and the representative mass spectrum preparation unit. A representative mass spectrum (generally, the mass-to-charge ratio is taken on the horizontal axis and the sum (or average value) of ion intensities in the above time zone is taken on the vertical axis), thus ensuring appropriate analysis thereafter Ru.
When the above TIC, MIC, EIC (chromatogram) calculated from the process of creating a representative mass spectrum or a representative spectrum obtained from them is displayed on the display, the user can view the above range of mass-to-charge ratio The MIC (in the case of MIC), the particular peak (in the case of EIC), the above mentioned time zone, and other parameters can be changed as appropriate.

  Preferably, the analysis apparatus further includes a condition setting unit that sets a condition for defining a data range to be a target for detecting the best time zone from the input measurement data.

Preferably, the analysis apparatus further includes a mass spectrum display unit for displaying the detected best time zone and the generated representative mass spectrum.
Furthermore, it is preferable that the analysis apparatus further include a mass spectrum storage unit that adds label information to the created representative mass spectrum and stores it.

  Only the date and the identification code (ID) are added to the measurement data output from the mass spectrometer (the mass spectrometer of the present invention). Therefore, if a user can understand and add rich label information that can be easily understood to the created representative spectrum and store it, mass spectrum management, editing (grouping etc.) and reading can be easily performed later. Will be able to The label information may be anything that can be understood by the user, but information on the user's organization, plans, actions, information on samples, persons who provided the samples, objects, places, dates, times, measurement conditions in the mass spectrometer, It includes information on measurement environment etc.

  The analyzer has an evaluation function of the created (or already stored) mass spectrum. The evaluation function may be provided in an analyzer having a function of automatically (semi-automatically) creating a representative mass spectrum as described above, or may be provided in an analyzer not having a function of automatically creating a representative spectrum (semi-automatic). .

  For mass spectrum evaluation, in addition to the mass spectrum group (including one mass spectrum) to be evaluated, it is premised that there are mass spectrum groups for creating an internal standard to be the evaluation standard. It is. There is a mass spectrum storage that stores these mass spectra. The mass spectrum to be evaluated may be stored in a storage unit (for example, a work area of a computer) different from the storage unit of the mass spectrum group for creating the internal symbol. All of these storage devices are collectively called a mass spectrum storage unit.

  The analyzer designates a mass spectrum storage unit storing a large number of created mass spectra, and a mass spectrum of a specific first group for internal mark creation from the mass spectra accumulated in the mass spectrum storage unit. The first designated part, or 1 or the ion intensity is high and the fluctuation is small (the ion intensity is higher than (or higher than the predetermined value and the fluctuation is within the predetermined range) based on the mass spectrum of the designated first group 1 or An internal symbol candidate creation unit for selecting a plurality of peaks, and a second designation for designating a mass spectrum (including one mass spectrum) of a second group to be evaluated from mass spectra accumulated in a mass spectrum storage unit Pass / fail judgment that determines the quality of each mass spectrum in the second group with one or more of the internal symbol candidates created by the unit and the internal symbol candidate creation unit as internal symbols Parts in which comprises a.

  Preferably, the first group of mass spectra is a collection of mass spectra determined to be good by the user or by the analysis device. When a specific peak or peaks are defined as an internal standard, the corresponding peak of the mass spectrum of the second group is based on the internal standard, and if the ion intensity is equal to or higher than a predetermined value and its variation is within an allowable range The mass spectrum is judged to be good. Others are bad. The mass spectrum determined to be good can be used as a target of management, editing and statistical analysis to be described later. It is preferable that the products determined to be defective are reproduced again by returning to the preparation of the representative mass spectrum described above.

  Next, the statistical analysis function is described. With respect to any of the analyzers equipped with both the representative mass spectrum creation function and the mass spectrum evaluation function described above, the analyzers equipped with one of the functions, and the analyzers not equipped with any of the functions below, The statistical analysis function described in can be provided.

  Statistical analysis mainly includes significance test, dimensionality reduction, machine learning and verification. And there are many types of significance tests specifically, and many types of dimension reduction, machine learning, and verification have already been developed.

  First, an analysis apparatus having a significant difference test function selectively displays a plurality of kinds of significant difference test methods and a mass spectrum storage unit storing data of a large number of mass spectra, and a selected significant difference test method A statistical analysis method input unit capable of inputting a desired significance level of data, a data set to which a selected significant difference test method is to be applied from a mass spectrum stored in a mass spectrum storage unit (a collection of a plurality of mass spectrums ), And a program routine for executing a plurality of types of significant difference test displayed in the statistical analysis method input unit, and for the specified data set, the selected significant difference test is specified. The statistical analysis execution unit executes a statistical analysis execution unit, which selects a peak determined to have a significant difference between the groups.

  Next, an analysis apparatus having a machine learning function has a statistical analysis method input unit that displays a plurality of machine learning methods in a selectable manner, and a program routine that executes the machine learning method displayed on the statistical analysis method input unit. And a statistical analysis execution unit that executes the selected machine learning method on a predetermined data set.

  The machine learning method selectively displays a plurality of types of machine learning methods, receives an input of the selected machine learning method, and selects a machine learning method selected from program routines for executing the displayed machine learning method. Program routine for the predetermined data set.

  In this way, the user can select a desired one from a plurality of machine learning methods to execute learning. The user can select multiple learning methods and compare the results. The verification method mentioned later can be used for this comparison.

  There are several types of predetermined data sets to be subjected to machine learning.

  The first one is to select and specify a data set to which a learning method to be selected is to be applied, from mass spectrum storage units storing a large number of mass spectrum data.

  The second one is to apply a learning method to a peak determined to be significant by the above-mentioned significant difference test.

  Third, the data set selected from the mass spectrum storage unit or the data determined to be significant by the significant difference test is contracted by a predetermined dimensionality reduction method into data of a contracted score. The learning method is applied to this.

  The reduction method can also display a plurality of reduction methods and allow the user to select one or more of them. In particular, since the reduction method and machine learning are related, it is preferable to display a plurality of types of reduction methods and a plurality of types of learning methods, and select and display the ones selected by the user on the display screen.

  At least one cross validation method may be displayed in a selectable manner, and learning results of the selected machine learning method may be verified by the selected cross validation method.

  A plurality of cross validation methods may be displayed to be selected by the user.

  A significant difference test method, a machine learning method, a contraction method, and a cross validation method are displayed so as to be selectable by a plurality of types, and the user is allowed to select these combinations. You can also judge It is also possible to have the computer execute all (or by default) combinations and verify which combination is the best by a verification method. The result is displayed. In this way, it is possible to select an objective combination regardless of the user's subjectivity, and it is also possible to present distinguishable peaks.

  Here, FIG. 9 is a block diagram showing an example of the analysis apparatus of the present invention. The analyzer of FIG. 9 is implemented by a computer system programmed as described below.

  The analysis device 320 is functionally divided into a processing unit 321 that executes creation of a representative mass spectrum, mass spectrum evaluation based on an internal standard, labeling, various statistical analysis processing, verification processing, etc. An input unit 322 for inputting mass spectrum data obtained by the analysis unit 310, a processing result such as an analysis process, an intermediate progress, and the like, and an output unit 323 used as a user interface, and a storage unit 326. . The mass analysis unit 310 corresponds to the mass spectrometer of the present invention.

  The input unit 322 is a medium reader for reading data stored in a USB memory, a CD-ROM, etc. in addition to a normal input device such as a keyboard, a mouse, etc. Data (including instructions) by communication regardless of wired or wireless Communication device etc. which receive the

  The output unit 323 may share part of the display screen with the input unit 322. The output unit 323 displays graphs of various data (including mass spectra) and other data in an easy-to-see format, and also displays various inputs as a user interface, a display unit 324 that displays a setting screen, and various data and processing results. Although the printer 325 has a printer 325 for outputting data, and further explicit illustration is omitted, a medium writer for writing various data and the like to a storage medium, a communication device for outputting by communication (transmission) (a communication device of the input unit 322 May be shared.

  The storage unit 326 temporarily stores input mass (mass) spectrum data, stores representative mass spectrum data (database shown in FIG. 10), various processes (shown in FIGS. 11 and 12 and will be described later). It is used for storing programs for manual generation, automatic generation of representative mass spectra, data editing and management, mass spectrum evaluation, statistical analysis processing, verification processing, etc., and provides a work area. The storage unit 326 is realized by a semiconductor memory, a hard disk or the like.

  The processing unit 321 is a main body of a computer, and executes the processing shown in FIGS. 11 and 12 on mass spectrum data input through the input unit 322 in accordance with various programs stored in the storage unit 326. At this time, necessary information (user interface screen) is displayed on the display unit 324, and data during processing and data of the processing result are stored in the storage unit 326.

FIG. 10 is a data structure diagram showing the measurement data obtained from the mass analysis unit 310 and the database created in the storage unit 326.
FIGS. 11 and 12 are flowcharts showing the procedure of processing executed by the processing unit 321 in accordance with the program stored in the storage unit 326. These figures are referred to sequentially in the following description.
The processing unit 321 has means for realizing the functions represented by these flowcharts.

  The analysis device 320 can be applied to measurement data obtained from many types of mass analysis units 310 having a scan function.

  FIG. 13 shows an example of text format raw data (text data) output from the mass analysis unit 310. This is obtained by repeating scanning from a low mass side to a high mass side at a constant time interval (for example, 0.05 seconds to 0.5 seconds as an example) by a scanning method for one sample. . The ion intensity (arbitrary scale, the same applies hereinafter) data obtained by each scanning corresponds to the value of the mass-to-charge ratio (m / z) (only one example of the range of 700 to 800 is shown) They are arranged as numerical values in columns 2, 3, 4 and so on. This is three-dimensional data (text file) of mass-to-charge ratio (m / z), ion intensity and measurement time (data representing scanning order).

  FIG. 14 represents the data shown in FIG. 13 in the form of a mass spectrum for each scan. The horizontal axis represents mass-to-charge ratio (m / z), the vertical axis represents ion intensity, and the depth direction (indicated by an arrow) represents time or the number of scans.

  The input unit 322 of the analysis device 320 acquires such raw data from the mass analysis unit 310 (S11 in FIG. 11) (measurement data input unit). The data may be transmitted from the mass analysis unit 310 and may be received by the input unit 322, or the mass analysis unit 310 may store the data in a storage medium such as a USB memory, and the input unit 322 may read the data. Good.

  FIG. 15 shows a chromatogram of TIC, MIC or EIC (described below) using the raw data described above. The horizontal axis is time (number of scans), and the vertical axis is ion intensity. The ion intensity on the vertical axis takes different values depending on TIC, MIC and EIC, but it should be understood that Fig. 15 is that of EIC (Since the ion intensity is an arbitrary scale, the value itself has no significant meaning. , MIC and EIC do not have any problem). A chromatogram is a plot of the sum of ion intensities (as summarized below, somehow organized or extracted) over time (along the time axis, ie in the order of the scans) .

  TIC is an abbreviation for total or total ion intensity (or current) (Total Ion Current), and the ion intensity (corresponding to all mass-to-charge ratios) (corresponding to all mass-to-charge ratios) of all peaks included in the acquired mass spectrum Means the sum of

  TICC stands for Chromatogram of TIC (draw TIC over time).

  MIC is an abbreviation of Merged Ion Current (or Merged Ion Current), and (m / m) represents a peak in a specific mass-to-charge ratio (hereinafter simply referred to as m / z) range. It means the sum of the ionic strengths (corresponding to z). MICC represents a chromatogram of MIC.

  EIC is an abbreviation for single (or extracted) peak ion intensity (or current) (Extracted Ion Current), and represents the ion intensity of a specific peak (corresponding to m / z). EICC is a chromatogram of EIC.

  A mode selection screen (not shown) is displayed on the display unit 324 of the analysis device 320 for creation of a representative mass spectrum, and the user selects either the manual mode or the automatic mode according to the mode selection screen. When the manual mode is selected, the processing unit 321 executes a process of manually generating a representative mass spectrum (S12 in FIG. 11) described next.

  In the manual generation process, one of TIC, MIC and EIC is selected by the user. In the case of TIC, raw data of all acquired m / z ranges are used, but in the case of MIC, the range of m / z to be used is specified by the user, and in the case of EIC, a specific peak is specified. Ru. A chromatogram (as shown in FIG. 15) of the TIC, MIC or EIC designated by the user in this manner is created by the processing unit 321 and displayed on the display unit 324.

  The user inputs a time range that is considered to represent the analysis result best on this chromatogram, using an input device such as a cursor (included in the input unit 322). The time range is determined, for example, by specifying the lower limit PL and the upper limit PH. When the time range is determined, the processing unit 321 adds up (or takes an average value) for each peak having the same m / z value of the mass spectrum of each scan within the time range to create a representative mass spectrum, Displayed on the display unit 324. An example of the representative mass spectrum created is shown in FIG. The representative mass spectrum data is text data consisting of two variables of mass-to-charge ratio (m / z) and ion intensity, and is stored in the database of the storage unit 326.

  When the user selects automatic on the mode selection screen relating to representative mass spectrum creation, a condition setting screen as shown in FIG. 17 is displayed on the display unit 324 (S131 in FIG. 11) (condition setting unit). The user can set desired conditions using this screen.

  In the condition setting screen, the target file is a file storing the raw data (measurement data) acquired from the mass spectrometric unit 310 in S11, and the file name assigned by the mass spectrometric unit 310 is displayed in a box.

  The TIC threshold, the MIC threshold, and the EIC threshold are thresholds for removing noise and the like in TICC, MICC, and EICC, respectively (see FIG. 17), and only data having a value exceeding the threshold is used in the following calculation. The user checks and selects one of TIC, MIC, and EIC that he / she wants to use for calculation (it may check all as shown in FIG. 17), and inputs a threshold for the selected one. . If there is no threshold input, a default value (prescribed value) is used.

  Since MIC is the sum of ion intensities in a particular m / z range, an input in the m / z range is required when MIC is selected. Moreover, since EIC represents the ion intensity of a specific peak, when EIC is selected, it is necessary to input the m / z value (designated spectrum) of the specific peak.

When the above input (that is, condition setting) is completed and the user presses the “execute” button (click, the same applies hereinafter), TIC, MIC, and EIC are selected (specified by putting a check) An operation for obtaining the following equation (1) is executed by the processing unit 321 for the chromatogram related to the.

The above equation (1) means to find the largest one of S (t), and in the case of TIC, S (t) is given by the following equation (2).

  In the case of MIC and EIC, MIC and EIC may be used instead of TIC.

  τ represents the horizontal axis of TIC, ie, time. The above equation (2) means that τ obtains the sum of TIC in the time width from t to t + Δt, and by the above equation (1), t (or t + Δt) at which the sum of this chromatogram becomes maximum is shown The time (time or sample instant) or time zone is determined. Δt may be set in advance (for example, several seconds to several tens of seconds), or may be input by the user on the condition setting screen (see FIG. 17).

  The time zone in which the sum of the ion intensities in the chromatogram is the highest is determined by the calculation of the equations (1) and (2) described above (S132 in FIG. 11) (best time zone detector). The fact that the sum of ion intensities is the highest is considered to be the most stable mass spectrum and the best characterization of the sample.

  A representative mass spectrum is created for a selected one of TIC, MIC, and EIC using ion intensity data of a time zone of t to t + Δt satisfying the above-mentioned formulas (1) and (2) (FIG. 11) S133) (representative mass spectrum creation unit). Then, the screen shown in FIG. 18 is displayed.

  In the screen shown in FIG. 18, for each of TIC, MIC, and EIC (these are all selected), the chromatogram is displayed on the left side, and in each chromatogram, the above-mentioned equation (1), A time zone satisfying (2) is indicated by a long rectangular box having a broken line. On the right side of the screen of FIG. 18, a representative created by the sum (or average value) for each mass / charge ratio m / z of the ion intensity data of the time zone satisfying the above-mentioned equations (1) and (2) A mass spectrum is shown (mass spectrum display). In FIG. 18, the representative mass spectrum is simply referred to as a representative spectrum or spectrum. Also, mass spectra are simply referred to as spectra.

  Any one of these representative mass spectra is stored in the storage unit 326 as the one most representative of the mass analyzed sample. This representative mass spectrum is text data consisting of two variables of mass-to-charge ratio and ion intensity. In FIG. 18, the user is required to select one to be stored in the storage unit 326 as “a spectrum to be adopted” in the box on the left side of each chromatogram. In FIG. 18, the user checks and selects MIC. Note that binary format data is also stored for recreating the representative mass spectrum, which will be described later. The check button will be described later.

  When creating a representative mass spectrum again, it returns to the condition setting shown in FIG. 17 to reinput the threshold value, the m / z range, the designated spectrum, and the value of the time zone Δt if necessary.

  As described above, the representative mass spectrum for one sample created in the processing unit 321 is registered in the database of the storage unit 326. At this time, in order to enable easy management and editing of data, labels for measurement condition information and sample information are attached. This label is preferably created such that the collection of data has a hierarchical structure (hierarchical structure). In this example, the highest label is the project name.

  Therefore, on the display screen of the display unit 324, a project name selection screen as shown in FIG. 19 is displayed. The user selects and inputs one of the project names displayed in a pull-down manner. Here, it is assumed that “human cancer sample” is selected as the project name. Then, the user presses (clicks on) the “register” button.

  Then, as shown in FIG. 20, a screen for inputting label information is displayed on the display unit 324. The project name has already been entered. As the label information, it is preferable for the user to be easily understandable and to clearly indicate the origin, attribute, characteristic and the like of the sample. In this example, the label uses the file name, the gender and age of the person who provided the sample, and the disease name, stage (progression degree) and tissue type of the sample. Of course, the project name may be included in the label information. Also, measurement condition information may be added. Then, when these contents are input as shown in FIG. 20 and the “store” button is pressed, this representative mass spectrum data is stored in the database together with the input label information (S 141 in FIG. 11). (Mass spectrum storage unit).

  Here, the data structure will be described with reference to FIG.

The measurement data d1 of one sample obtained by the measurement (mass spectrometry) in the mass spectrometric unit 310 differs from the format of the raw data by the mass spectrometric unit 310. Given to The mass spectrum data is attached with an ID (identification code) assigned by the mass analysis unit 310 or the user.
In the analysis device 320, the text data is converted into data d3 in binary format in order to display a mass spectrum in the above-described processing of manual generation and automatic generation of representative mass spectrum (S12 and S13 in FIG. 11). . When representative mass spectrum data (text format) d4 is obtained, label information and the like are input, and label information is added to the representative mass spectrum data instead of or in addition to the ID given by the mass analysis unit 310 (FIG. 11). S141). Then, data conversion to a text file for relational database (RDB) construction is performed, and the project is accumulated in the database. This database is shown in particular in FIG.

  The representative mass spectrum data thus stored in the mass spectrum database 327 is used for various purposes. One of them is a statistical analysis process described later, and the other is an internal symbol search process described next. For each of these processes, a group (data set) of data is selected from the accumulated mass spectral data. This is the folder of the selected mass spectral data set indicated at 328 in FIG.

As an example, FIG. 21 shows an example of a display screen for selecting one of the mass spectrum data already stored in the mass spectrum database 327, which belongs to a specific group, in order to perform statistical analysis. The project name (human cancer sample) and the range of label information that defines the group, that is, gender (not specified), age (50 to 80), disease (liver cancer), stage (1 to 3), tissue type ( The primary HCC) is input by the user. When the "select" button is pressed, the mass spectrum database 327 is searched, and mass spectrum data satisfying the above project name and label range are extracted (S142 in FIG. 11). The label information of the extracted data is displayed in the form of a list as shown in the lower part of FIG.
When the "Export" button is pressed, it is exported (S143 in FIG. 11). That is, the extracted mass spectrum data is transferred from the database to a predetermined storage location of the storage unit 326, and statistical analysis processing can be used (creation of the mass spectrum database 328) (data set designation unit).

  Thus, by adding label information to all representative mass spectrum data, grouping, search, extraction (selection), etc. of data becomes easy using terms and concepts (labels) that can be understood by the user.

  All mass spectra created in the representative mass spectrum creation process (S12, S13 in FIG. 11) (especially, in the automatic creation process (S13 in FIG. 11)) described above are those with high quality (quality) Not exclusively. The quality of the mass spectrum created or already accumulated in the mass spectrum database 327 is evaluated as follows. Although this evaluation process is executed by designating on the menu screen (not shown), it is possible to proceed to this evaluation process even if the "check" button is pressed on the display screen of FIG.

  First, an index for evaluation (referred to as an internal label) is determined (searched and selected). Next, using this internal label, the quality of a specific mass spectrum is judged (discriminated) (individual spectrum judgment) or the quality of a mass spectrum in a specific group is judged (discriminated) (spectrum batch in folder) Judged).

  The determination of the internal label is performed using a large number of stored mass spectral data. As mass spectrum data used to determine the internal standard, it is preferable to use a group determined to be a good mass spectrum in the evaluation of mass spectra performed in the past, or a collection of objects that the user visually determined to be a good mass spectrum .

  First, on the display screen shown in FIG. 22, a large number of mass spectrum data to be used for determining the internal standard are designated, and conditions are set. Here, one of the selected mass spectrum data set folders (reference numeral 328 in FIG. 10) is designated as the target folder (first designation means). It is also possible to input label information and specify a target folder or the like. Further, as the conditions, the lower limit value of the detection intensity and the upper limit value of the variation coefficient are input. The mass spectrum has a number of peaks corresponding to specific m / z values. The internal label is determined by selecting one or more of these peaks (with less variation) which appear stably. That is, the internal label is a specific stable peak. The lower limit value of the detection intensity defines the lower limit value of the ion intensity (average intensity) of the peak adopted as the internal standard. That is, a peak having a value with an average intensity higher than the lower limit can be a candidate for the internal label. The coefficient of variation is a value obtained by dividing the variance by the average value, assuming that a set of peak values (ion intensities) corresponding to a specific m / z value follows a normal distribution. A peak having a variation coefficient lower than the input variation coefficient upper limit may be a candidate for the internal symbol. These two conditions are AND conditions.

  When the target file is specified, the detection strength lower limit and the variation coefficient upper limit conditions are input, and the “execute” button is pressed, a set of peak values corresponding to m / z values for all mass spectra in the target folder The coefficient of variation and the average ion intensity are calculated with respect to and those among the obtained results that satisfy the above condition are displayed as the internal symbol candidate, arranged in ascending order of the coefficient of variation, as shown in the upper half of FIG. Be done (internal symbol candidate creation unit). What is displayed is the order, the coefficient of variation, m / z, and the average ion intensity. The user selects an appropriate internal symbol for mass spectrum evaluation from among the internal symbol candidates, and checks the corresponding box. In the example of FIG. 23, m / z is selected as the internal symbol of the 208 peak. If it is not selected by the user, the peak with the smallest coefficient of variation is automatically selected as the internal standard (S151 in FIG. 16).

  Evaluation of mass spectra using this internal label is based on the following concept. That is, since the internal standard is a peak that appears stably in the mass spectrum as described above, the corresponding peaks (peaks with the same m / z value) have similar ion intensities even in the mass spectrum to be evaluated. It can be expected to have Therefore, as an example, an allowable range is set up and down centering on the average ion intensity of the peak which is the internal standard, and if the ion intensity of the corresponding peak of the mass spectrum to be evaluated is within this allowable range If it is outside the allowable range, it is judged as a bad mass spectrum.

  In the lower half of the screen of FIG. 23, the user inputs a folder including mass spectral data files to be evaluated. The mass spectrum data to be evaluated may be designated by the label information described above (as described above, “in-folder spectrum batch determination). Also, one mass spectrum created by the representative mass spectrum automatic creation process of S13 in FIG. The quality of the data may be determined (individual spectrum determination) (as described above, the second designation unit) This representative spectrum determination is representatively shown in FIG. It is performed when the "check" button next to the sentence "Please press here" is pressed.

  When the "Execute" button is pressed, it is judged whether the ion intensity of the corresponding peak is within the upper and lower tolerance of the average ion intensity of the internal standard, based on the internal standard. If the mass spectrum is out of the allowable range, it is judged to be a defective mass spectrum (S152 in FIG. 11) (goods / nonconforming judging means). At the bottom of the screen of FIG. 23, the file names of the determined good spectrum and the bad spectrum are listed.

  The basic mass spectrum data group (target folder designated in FIG. 22) used to determine the internal standard and the mass spectrum data of the quality determination target are obtained from mass analysis of the same sample. A sample of the same species is a cell of the same disease such as liver cancer, the same organ as a specific animal or human liver, a cell of the same site, a peak of the same m / z value such as part of a living organism. Are expected to appear (peaks of different m / z values may be included). When the peak determined as the internal standard is used, the internal standard may not be correct if the number determined as the good spectrum is smaller than that determined as the defective spectrum. The determination process may be retried (such as changing basic mass spectral data), or a peak (m / z value) having a second or later variation coefficient may be used as an internal standard. When a plurality of internal marks (peaks) are determined, final judgment results can be obtained by AND logic or OR logic of the quality judgment results based on each internal mark.

  When it is determined that the created mass spectrum is defective in the individual spectrum determination, the process returns to the automatic creation of the representative spectrum (S13 in FIG. 11) and changes the threshold in the screen of FIG. 17 (relaxes the condition by the threshold), By selecting not MIC but TIC or EIC on the screen of FIG. 18, it is possible to redo creation (re-creation) of a representative mass spectrum.

  Although there are various total analysis methods, they are roughly classified into four, and a few representative methods included therein are described for each classification.

1) Significance test · Welch t-test (Welch's t-test)
Under the null hypothesis that "the averages of the two populations are equal," a two-sided test is performed without assuming equal variance.
・ WRST (Wilcoxon rank sum test) (Wilcoxon rank sum test)
A nonparametric test is performed based on the null hypothesis that "both samples are extracted from the same population".
・ ANOVA (Analysis of variance) (analysis of variance)
A multigroup parametric test is performed based on the null hypothesis that there is no difference in the population mean of all groups.

  The significant difference test is performed by the analyzer for each mass spectrum peak (m / z value), and it is useful for searching for peaks having a significant difference between mass spectrum groups (groups). It can be used to elucidate the molecular mechanism. Peaks (m / z) that are considered to be significantly different between the groups are selected, and the corresponding data can be used in machine learning to be described later.

2) Dimension reduction Dimension reduction reduces many variables to a few variables (scores).
・ PCA (Principal component analysis) (principal component analysis)
It is an unsupervised dimensionality reduction method.
・ PLS (Partial least squares) (partial least squares method)
It is a supervised dimensionality reduction method.
-OPLS (Orthogonal Partial Least squares) (Orthogonal PLS)
This is a modified version of PLS that separates and analyzes orthogonal components of explanatory variables.
-KPLS (kernel partial least squares) (kernel PLS)
Since the PLS is nonlinearly extended using the kernel method, separation performance is improved.

  In the analysis apparatus, since many peaks (variables) included in the mass spectrum can be contracted to a few small scores (principal components), the correlation between the scores can be determined, and the results can be used for machine learning. The number of scores can be specified by the user.

3) Machine learning method · LDA (Linear discriminant analysis) (linear discriminant analysis)
Construct a discriminant function based on straight lines and hyperplanes.
・ QDA (Quadratic discriminant analysis) (secondary discriminant analysis)
Construct a discriminant function with curves and hypersurfaces.
・ SVM (Support vector machine) (Support vector machine)
A non-linear identification method in which the identification surface that maximizes the margin is configured in the feature space.
・ LR (Logistic regression) (Logistic regression)
A regression model that assumes that log likelihood ratios of posterior probabilities are represented in linear form.
・ RF (Random forest) (Random forest)
A group learning algorithm that uses decision trees as weak learners.

  The analyzer can diagnose an unknown mass spectrum using the discriminant function generated by these learning methods. Therefore, it can be used for diagnosis and determination of treatment plan.

4) Verification method This is to verify the diagnostic accuracy of the machine learning method. This verification result also makes it possible to automatically select the most suitable machine learning method.

K-fold CV (k-fold cross validation) (k-fold cross validation)
A verification method in which a sample group is divided into k pieces, one of which is a test sample and the remaining is a training sample.
・ LOOCV (Leave one out cross validation) (Leave-one-out cross validation)
A verification method in which only one sample is extracted from a sample group and used as a test sample, and the remaining as a training sample.

  A program (routine) (means) for executing each individual verification method (Welch t-test, WRST, ANOVA, etc.) included in the above-described significant difference test method is stored in the storage unit 326 of the analysis device 320. Programs (routines) (means) that respectively execute all individual reduction methods (PCA, PLS, OPLS, KPLS, etc.) included in the method, all individual learning methods (LDA, QDA included in machine learning methods) , (SVM, LR, RF, etc.) programs (routines) (means) and programs (routines) (means) respectively executing all individual verification methods (k-fold CV, LOOCV, etc.) included in the verification method Is stored, and the processing unit 321 individually performs each statistical analysis method and verification method according to these programs. Or can be performed simultaneously.

  That is, as shown in FIG. 24, all the statistical analysis methods and verification methods described above are displayed on the display screen of the display unit 324, and the user can select any one of the displayed analysis methods and verification methods. The above can be selected. All analysis and validation methods can also be selected. In addition, one or more of the verification methods, one or more of the reduction methods, one or more of the learning methods, and one or more of the verification methods can be selected in combination, and among the combinations You can also remove one or more from That is, it becomes possible to select any combination (S21 in FIG. 12) (statistics analysis method input unit). Then, the processing unit 321 can execute the selected analysis method and verification method individually, in parallel, or in order according to the combination (S30, S40, S50, S60 in FIG. 12) (Statistics) Analysis execution unit). The results of these processes (various numerical information obtained as a result of statistical analysis, reference numeral 329 in FIG. 10) are displayed on the screen of the display unit 324 and printed by the printer 325 or in the form of data. Or output to a storage medium (S31, S41, S52, S61 in FIG. 12).

  As a result of execution of the selected test method, data corresponding to the selected peak is passed to the selected reduction method to be used as its processing (reduction) target data or to the selected learning method. It can also be used as processing (learning) target data (S 32 in FIG. 12). Similarly, the score data processed and output by the selected reduction method can be passed to the selected learning method and used as processing (learning) target data (S42 in FIG. 12). In particular, the combination of the selected reduction method and the selected learning method is displayed on the display screen as a line connecting them (see FIG. 24). The selected verification method verifies the diagnostic accuracy of the selected learning method (S50, S60 in FIG. 12). Diagnosis processing of unknown data is also possible by the discriminant function determined by the selected learning method (S51 in FIG. 12), and the diagnosis result is presented (S52 in FIG. 12). The selected analysis method and verification method are executed in response to the press (click) of the “execute” button (see FIG. 23).

Thus, since there are many types of statistical analysis methods and program routines for executing verification methods, the user can select from among these general-purpose to advanced program routines ( One or more can be performed. It is possible to install additional program routines for statistical analysis and verification methods that have not yet been provided (highly diffusive). From the test method, the reduction method, and the learning method, an appropriate combination (a combination other than the test method may be set) can be set, and the trial can be performed on the mass spectrum group to be analyzed. You can choose the best combination for you. At this time, a verification method can be used to determine whether the set combination is appropriate.
As in the following example, a combination of a plurality of statistical analysis methods is used to manually or automatically change meaningful changes, for example, molecules (markers) that specifically change in a disease, for a mass spectrum group of interest. It can be extracted (including semiautomatic). The screen (user interface) shown in FIG. 24 is simple to operate and is easy to understand, and the working time can be shortened.

  As shown in FIG. 24, a specific example in the case where ANOVA is selected as a test method, PCA as a reduction method, LDA as a learning method, and LOOCV as a verification method will be described below.

  The folder of the data set used for analysis (the box labeled "Select a folder containing the data set to be used") contains mass spectrum data of rabbit plasma of three groups (groups) (FIG. 11, S142, S143) (data set designation unit). The three groups of rabbits are as follows.

10 normal rabbits (placed as control): abbreviated as C0 10 rabbits with added cholesterol load through food: abbreviated as C16 10 rabbits with genetic abnormality in cholesterol metabolism: abbreviated as W

  Thus, 30 mass spectral data of these 30 rabbit plasmas are the subject of statistical analysis.

The m / z range defines the range of m / z values used for analysis in a mass spectrum, and a range of 10.0 to 1000.0 is specified here.
The Bin size indicates the possible interval (width) of the m / z value, and here, 1 is set. Therefore, mass spectra having ion intensities corresponding to m / z values that change one by one, such as 10.0, 11.0, 12.0,..., 999.0, 1000.0, are used as target data. If the Bin size of mass spectral data in the data set is not 1, take the average value or the added value (if the Bin size is small), or interpolate (if the Bin size is large), etc. It is processed to be

  As statistical analysis methods, ANOVA, PCA and LDA are selected as described above, and LOOCV is selected as a verification method. Since the reduction result by PCA is used in the learning method LDA, blocks of these characters are connected by a line. The significance level, multiple test correction, coefficient of variation range and mean intensity range are with respect to the test ANOVA, which is described in the next section on marker search. In the reduction method PCA, the number of scores output as a result of reduction is set to 2. After the above setting (S21 in FIG. 12), when the execution button is pressed, the set statistical analysis and verification are executed.

  Data processing is performed prior to execution of the statistical analysis routine and the verification routine (S22 in FIG. 12). The mass spectrum data of the data set read from the mass spectrum database 328 (see FIG. 10) is processed so as to be in the set m / z range and to have the set Bin size. Also, with respect to the 30 mass spectra, normalization (northization) of ion intensity is performed for each group of C0, C16, and W. That is, the average intensity of each spectrum is calculated, and normalization is performed by dividing the value of each peak by this average intensity.

  The purpose of marker search is to find marker substances useful for identification of three groups, among m / z = 10-1000.

  The marker search is performed by narrowing down (searching for) substances (m / z) that satisfy the following conditions (i), (ii) and (iii).

(I) Select m / z whose intensity (ion intensity) differs greatly between groups. This is achieved by the test method ANOVA, which is described in detail below.
(Ii) Select m / z whose intensity is sufficiently large. The average strength is 1.0 to inf. It is (infinite). Since 1.0 is normalized by the average value, it means the average value. The condition (ii) is to select m / z corresponding to a peak whose intensity exceeds 1 in any group.
(Iii) Select m / z with sufficiently small variation in strength within each group. The variation is defined by the variation coefficient range (0.0 to 0.3) set on the screen of FIG. The coefficient of variation is determined by the value of the variance when the intensity of each peak is 1 (the value of the variance varies with the intensity of the peak, so normalization is performed). The condition (iii) is to select m / z corresponding to a peak for which the above-mentioned dispersion is less than 0.3 in each group.

  An ANOVA method for selecting m / z satisfying the above condition (i) will be described.

  Under the null hypothesis that the average intensity of each group is equal (if the average intensity of each group is μ1, μ2 and μ3 is μ1 = μ2 = μ3), the ANOVA method is for each peak (for each m / z value) Calculate P value. The P value indicates the probability that the null hypothesis is established. The larger the P value, the smaller the difference in intensity for the same m / z value between groups (null hypothesis is correct).

  The significance level of the P value is set to 0.05 (screen in FIG. 24). The m / z whose P value calculated as described above is smaller than this significance level 0.05 is selected as m / z (peak) whose intensity is different between the groups. Note that, in FIG. 24, the multiple test correction uses the BF (Bonferroni) method to set the significance level of the P value to 0.05 / N based on the Bonferroni method when the test is repeated N times (N> 2). means.

  M / z value (peak) selected as satisfying the above conditions (i), (ii) and (iii), the corresponding P value (represented by -log), coefficient of variation (CV_C0 CV_C16 and CV_W indicate groups C0, C16 and W, respectively, and average intensities (M_C0, M_C16 and M_W indicate groups C0, C16 and W, respectively) are shown as output files in FIG.

  Although m / z satisfying the above conditions (i), (ii) and (iii) is selected as described above (see FIG. 25) in the marker search, the number of variables (m / z) is large, so Reduce the number of variables by the reduction method. By reducing the amount of information by the dimensionality reduction method, the accuracy of learning / diagnosis may be improved.

  Principal component analysis (PCA) is set for the dimensionality reduction method, and the number of scores is 2 (see FIG. 24).

  The results of reducing the ion intensities of all the peaks in the 30 mass spectral data to two main components (first and second main components) by PCA are shown in FIG. The ion intensities of all the selected m / z peaks shown in FIG. 25 may be reduced.

  PC1 and PC2 are first and second main components, respectively. jw-0W, JW-16W, and WHHL correspond to the groups C0, C16, and W, respectively. It is understood that these groups are highly likely to be separated (discriminated) by their main components PC1 and PC2.

  All of the original three groups of data (all m in addition to dimensional contraction using m / z selected as shown in FIG. 25 as satisfying the above conditions (i), (ii) and (iii) Dimension reduction can also be performed using / z).

  It is an object of the present invention to perform machine learning using teacher data, and to estimate a group to which an unknown spectrum belongs with high accuracy based on the learning result (discrimination function). Most preferably, the above-described statistical test, dimensionality reduction method, and machine learning method can be combined to construct a more accurate classifier.

  As an example, a discriminator (PCA-LDA) for estimating an objective variable (group) using the first and second principal components as explanatory variables based on the result obtained by the reduction method shown in FIG. Can. In this case, even if learning is performed on the basis of the result of contraction using the results of the test (ANOVA) (FIG. 23), the contraction is performed using the original three groups of mass spectral data. The learning may be based on either one or the other.

  The following four types of methods are available to estimate the group to which an unknown spectrum belongs.

1) Machine learning: Train teacher data using all m / z as explanatory variables (object variable is group name).
2) Statistical test → machine learning: Train teacher data using m / z regarded as important for identification in the test as an explanatory variable.
3) Dimension reduction method → machine learning: The information of all m / z is reduced to fewer variables (principal component and PLS score), and learning of teacher data is performed using them as explanatory variables.
4) Statistical test → dimensional reduction → machine learning: m / z which is regarded as important for discrimination in the test is reduced to smaller variables (scores) and learning of teacher data is made as explanatory variables Do.

  In the screen shown in FIG. 24, Leave-one-out cross validation (LOOCV) is set as a verification method. This is the process of "Can one sample be taken out of all samples for verification and training be performed on the remaining samples, and the group to which the verification sample belongs be correctly estimated?" Output the rate.

  FIG. 27 shows the result of checking the correct answer rate estimated by the above-mentioned discriminator (PCA-LDA) by LOOCV (correct errata). [1] is a correct answer, [0] is an incorrect answer, and the correct answer rate is 27/30 = 90%. In addition to the above, verification results can also be output as performance data such as ROC curves and AUCs.

  The combination of the reduction method and the learning method is changed, and in some cases, the combination with the test method is also changed, and the accuracy of estimation based on each learning result is determined as the verification method (the verification method may be changed) Semi-automatically (or the user selects a combination) automatically (or a combination of test method, reduction method, and learning method) that can learn / diagnose data with higher accuracy based on the result of estimation accuracy of Can be selected.

  FIG. 28 shows the results of statistical analysis and verification shown in FIG. 25, FIG. 26, and FIG. 27 collectively on one display screen. By displaying such a screen, the user can comprehensively view the results of a series of analysis and verification.

(Test example using the mass spectrometry system of the present invention)
Using the mass spectrometry system of the present invention, a detection confirmation test of cancer cells in blood was performed.
SAS was analyzed on a sample in which 2 mL of human whole blood was mixed with varying concentrations of human squamous cell carcinoma cells (SAS). The sample No. 1 is a sample of whole blood free of cancer cells. Samples No. 2 to No. 5 had different concentrations of SAS.
Sample No. 2: 1 × 10 ⁵ SAS / 2 mL blood
No. 3 sample: 0.5 × 10 ⁵ SAS / 2 mL blood
Sample No. 4: 1 × 10 ⁴ SAS / 2 mL blood
Sample No. 5: 1 × 10 ³ SAS / 2 mL blood
Sample No. 6 is a positive control of cancer cells only.

The test was performed according to the following procedure.
i) Blood was collected.
ii) The mononuclear cell separation process was performed by Ficoll.
iii) Mass spectrometry was performed using the mass spectrometer of the present invention having an ionization means using the remote sampling electrospray ionization method described in the second embodiment.
iv) The analysis device of the present invention (including the determination unit and the statistical analysis unit including the machine learning) is used to output the result of mass analysis.

The test results are shown in FIG. 29 and FIG.
From the results of FIG. 29 and FIG. 30, it was confirmed that No. 2 to No. 4 show the same peak shape as No. 6 at 740, 758 and 764 (m / z).
It could be confirmed that it was possible to detect 5,000 tumor cells mixed in 1 mL of blood.

The aspect of the present invention is, for example, as follows.
<1> An ionization means for ionizing the component to be examined in the blood of the subject
An analysis means for mass analyzing ions generated by the ionization means;
It is a mass spectrometer characterized by having.
<2> The mass spectrometer according to <1>, wherein an antibody (excluding an antibody derived from the same species as the subject) is not added to the test target.
<3> The mass spectrometer according to any one of <1> to <2>, wherein the test target is any of cells, proteins, peptides, nucleic acids, lipids, and carbohydrates.
<4> The mass spectrometer according to <3>, wherein the cell is a circulating tumor cell (CTC).
<5> The mass spectrometer according to any one of <1> to <4>, wherein the test target is obtained by mononuclear cell separation of the blood.
<6> The ionization means is
A needle-like member which contacts or pierces the test object to collect the test object;
A power supply for applying a voltage to the needle-like member in order to generate ions derived from the inspection object from the tip of the needle-like member by electrospray;
An ion collection member for collecting generated ions;
The mass spectrometer according to any one of <1> to <5>, wherein
<7> The ionization means is
A first tubular member, one end of which is used to contact the subject to be examined to collect the subject to be examined, and the other end of which is used to generate ions derived from the subject to be examined by electrospray;
A second tubular member concentrically inserting the first tubular member;
A gas flow generating member for flowing a gas to the second tubular member;
A voltage is applied to the counter electrode such that a potential difference is generated between the other end of the first tubular member and the counter electrode opposed to the other end disposed in the region where the electrospray is generated. Power supply,
An ion induction member for guiding generated ions to the mass analysis means;
The mass spectrometer according to any one of <1> to <5>, wherein
<8> The ionization means is
A linear member for transporting droplets of the sample solution using the sample solution to be tested prepared using a limiting dilution method;
An ionization member that generates ions derived from the inspection object from the droplets attached to the linear member;
The mass spectrometer according to any one of <1> to <5>, wherein
<9> An ionization step of ionizing a component to be examined with respect to a subject to be examined in blood of a subject;
An analysis step of mass analyzing ions generated by the ionization step;
It is a mass spectrometry method characterized by having.
<10> The mass spectrometric method according to <9>, wherein an antibody (excluding an antibody derived from the same species as the subject) is not added to the test object.
<11> The mass spectrometer according to any one of <1> to <8>,
Information of at least one of mass spectrum information created based on mass spectrometry results obtained by the analysis means, and mass spectrum analysis result information obtained by learning the mass spectrum information To determine whether the state of presence of the test object in blood exceeds a predetermined tolerance or whether the mass spectrum exhibits a disease-positive pattern, by using
When it is determined that the presence state of the test object exceeds the predetermined allowable range or the mass spectrum exhibits a disease-positive pattern, the subject is derived from the presence state of the test object An analysis device having a determination unit that determines the possibility of falling under the condition
It is a mass spectrometry system characterized by having.
<12> The mass spectrometry system according to <11>, wherein an antibody (except for an antibody derived from the same species as the subject) is not added to the test object.
<13> The analysis device determines the possibility of falling under a plurality of diseases at one time, based on mass spectrometry results obtained for a single measurement sample of the blood of the subject. It is a mass spectrometry system in any one of 11> to <12>.
<14> Information of mass spectrum created based on the result of mass analysis obtained by ionizing the component to be examined in the blood of the subject and examining the generated ions And / or using at least one of the information of the analysis result of mass spectrum obtained by learning the information of the mass spectrum, whether the existence state of the test object in blood exceeds a predetermined allowable range Determine whether or not
When it is determined that the presence state of the test object exceeds the predetermined allowable range, the possibility of the subject corresponding to a disease derived from the presence state of the test object is determined.
It is a mass spectrometry analysis program characterized by causing a computer to execute processing.
<15> The mass spectrometry analysis program according to <14>, wherein an antibody (except for an antibody derived from the same species as the subject) is not added to the test target.

  The mass spectrometer according to any one of the above <1> to <8>, the mass spectrometric method according to any one of the above <9> to <10>, any one of the above <11> to <13> According to the mass spectrometry system described in and the mass spectrometry analysis program described in any one of the above <14> to <15>, the above-mentioned various problems in the prior art are solved and the above object of the present invention is achieved. Can.

DESCRIPTION OF SYMBOLS 1 Needle-like member 7 Power supply 8 Ion collection member 12 Mass spectrometer 20 Power supply 21 1st tubular member 22 2nd tubular member 25 Gas flow generation member 28 Ion induction member 29 Mass spectrometer 33 Linear member 35 Ionization member 36 Mass Analyzer 100 mass spectrometer 101 ionization means 102 analysis means 200 mass spectrometry 201 ionization means 202 analysis means 300 mass spectrometer 301 ionization means 302 analysis means 310 mass spectrometry unit 320 analysis device 321 processing unit 322 input unit 323 output unit 324 display Part 325 Printer 326 Memory

Claims (15)

An ionization means for ionizing the component to be examined in the blood of the subject;
An analysis means for mass analyzing ions generated by the ionization means;
A mass spectrometer characterized by having:   The mass spectrometer according to claim 1, wherein an antibody (except for an antibody derived from the same species as the subject) is not added to the test object.   The mass spectrometer according to any one of claims 1 to 2, wherein the test object is any of a cell, a protein, a peptide, a nucleic acid, a lipid and a carbohydrate.   The mass spectrometer according to claim 3, wherein the cells are circulating tumor cells (CTC) in blood.   The mass spectrometer according to any one of claims 1 to 4, wherein the test object is obtained by mononuclear cell separation of the blood. The ionization means
A needle-like member which contacts or pierces the test object to collect the test object;
A power supply for applying a voltage to the needle-like member in order to generate ions derived from the inspection object from the tip of the needle-like member by electrospray;
An ion collection member for collecting generated ions;
The mass spectrometer according to any one of claims 1 to 5, which has The ionization means
A first tubular member, one end of which is used to contact the subject to be examined to collect the subject to be examined, and the other end of which is used to generate ions derived from the subject to be examined by electrospray;
A second tubular member concentrically inserting the first tubular member;
A gas flow generating member for flowing a gas to the second tubular member;
A voltage is applied to the counter electrode such that a potential difference is generated between the other end of the first tubular member and the counter electrode opposed to the other end disposed in the region where the electrospray is generated. Power supply,
An ion induction member for guiding generated ions to the mass analysis means;
The mass spectrometer according to any one of claims 1 to 5, which has The ionization means
A linear member for transporting droplets of the sample solution using the sample solution to be tested prepared using a limiting dilution method;
An ionization member that generates ions derived from the inspection object from the droplets attached to the linear member;
The mass spectrometer according to any one of claims 1 to 5, which has An ionization step of ionizing the component to be examined in the blood of the subject;
An analysis step of mass analyzing ions generated by the ionization step;
A mass spectrometry method characterized by having:   The mass spectrometry method according to claim 9, wherein no antibody (except for an antibody derived from the same species as the subject) is added to the test object. The mass spectrometer according to any one of claims 1 to 8.
Information of at least one of mass spectrum information created based on mass spectrometry results obtained by the analysis means, and mass spectrum analysis result information obtained by learning the mass spectrum information To determine whether the state of presence of the test object in blood exceeds a predetermined tolerance or whether the mass spectrum exhibits a disease-positive pattern, by using
When it is determined that the presence state of the test object exceeds the predetermined allowable range or the mass spectrum exhibits a disease-positive pattern, the subject is derived from the presence state of the test object An analysis device having a determination unit that determines the possibility of falling under the condition
The mass spectrometry system characterized by having.   The mass spectrometry system according to claim 11, wherein no antibody (except for an antibody derived from the same species as the subject) is added to the test object.   The analysis device determines the possibility of falling under a plurality of diseases at one time based on the result of mass spectrometry obtained for a single measurement sample of the blood of the subject, as described above. The mass spectrometry system according to any one of the above. Information of a mass spectrum created based on the result of mass analysis obtained by ionizing the component to be examined with respect to the test object in the blood of the subject and mass analyzing the generated ions, and the above By using at least one of the information of the analysis result of the mass spectrum obtained by learning the information of the mass spectrum, whether the existence state of the test object in the blood exceeds a predetermined allowable range or not Judge
When it is determined that the presence state of the test object exceeds the predetermined allowable range, the possibility of the subject corresponding to a disease derived from the presence state of the test object is determined.
A mass spectrometry analysis program that causes a computer to execute a process.   The mass spectrometry analysis program according to claim 14, wherein no antibody (except for an antibody derived from the same species as the subject) is added to the test object.