Tumor Analysis System

Abstract

To provide a system for estimating cell composition of a tumor analyte including a normal cell, on the basis of cell composition of a cancer clone independently derived, and cell composition of a normal cell.SOLUTION: There is provided a system for analyzing cell composition in a tumor analyte, the system has at least a calculation part. At least one storage unit provided in the system or coupled to the system includes cell composition of a cancer clone which is acquired from cells included in the tumor analyte on the basis of information of gene mutation, and cell composition of a normal cell acquired from cells included in the tumor analyte on the basis of an amount of gene expression. The calculation part executes a step for adjusting an existence ratio of each cell species, about the cell composition of the cancer clone and cell composition of the normal cell read from the storage unit, while considering a fact that, out of the cell composition of the cancer clone, the normal cell is included in non-classified cells, and a step for outputting the cell composition of the adjusted tumor analyte.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a system and method for predicting the composition of cell types constituting a tumor specimen based on gene sequence information obtained from tumor tissue. [Background technology]

[0002] Tumors contain many cells that aid in the proliferation of cancer. These cells and the environment they create are called the tumor microenvironment. In order to clarify the state of the tumor microenvironment, attempts have been made to clarify the cellular composition of cancer clones that compose tumor specimens. For example, there is a method that uses a Bayesian clustering method to derive the cellular composition of cancer clones in tumor specimens based on information such as the allele frequency of single-base mutations and copy number abnormalities (Non-Patent Document 1).

[0003] However, samples taken from tumor sites of patients, such as clinical samples from cancer patients, contain not only tumor cells but also normal cells such as immune cells that do not have somatic mutations. It is known that normal cells that infiltrate into tumors, such as killer T cells that suppress the activity of cancer cells and macrophage cells that support the activity of cancer, are greatly involved in the formation of cancer. Therefore, in order to investigate the extent to which normal cells have infiltrated into tumor sites, a technology has been developed that derives the composition of normal cells based on the expression level, which is information indicating the degree to which genes known as marker genes, which are useful for characterizing various cell types, are active within the cells in tumor samples (Non-Patent Document 2). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2019-509282 [Non-patent literature]

[0005] [Non-Patent Document 1] A Roth et al., Nature Methods 11, 396-398 (2014) [Non-Patent Document 2] AM Neuman et al., Nature Biotechnology 37, 773-782 (2019) Summary of the Invention [Problem to be solved by the invention]

[0006] On the other hand, no technology is known for simultaneously estimating the cellular composition of cancer clones and normal cells. Information on gene mutations can be used to derive the cellular composition of cancer clones. The information on gene mutations includes allele frequency, copy number variation (CNV), ploidy, gene fusion, and the like. On the other hand, gene expression levels can be used to derive the cellular composition of normal cells, and in particular, the expression levels of marker genes that are characteristically expressed in each cell type are often used. Therefore, the present invention provides a technology for simultaneously estimating cancer clones and normal cells, which uses different information.

[0007] That is, the present invention aims to predict the cellular composition of a tumor specimen containing normal cells based on the cellular composition of a cancer clone and the cellular composition of normal cells that are independently derived. [Means for solving the problem]

[0008] As a result of thorough investigation into the above-mentioned problems, the present inventors discovered that by adjusting the cellular composition of a cancer clone and the cellular composition of normal cells, taking into account that unclassified cells in the cellular composition of a cancer clone include normal cells, it is possible to derive a cellular composition consisting of cancer clones and normal cells contained in a tumor specimen, and thus completed the present invention.

[0009] In one aspect, the present invention provides a system for analyzing cellular composition in a tumor specimen, comprising: At least a calculation unit is provided, At least one storage device included in the system or connected to the system includes a cell composition of a cancer clone obtained based on information on gene mutations from cells contained in a tumor specimen, and a cell composition of a normal cell obtained based on gene expression levels from cells contained in a tumor specimen, The calculation unit: a step of adjusting the abundance ratio of each cell type in the cell composition of the cancer clone and the cell composition of normal cells read from the storage device, taking into consideration that normal cells are included in unclassified cells in the cell composition of the cancer clone, so that the sum of the abundance ratios of each cell type becomes X (X is a constant between 0 and 1, and is a parameter designated by a user of the system); Outputting the cellular composition of the adjusted tumor specimen The present invention provides a system for performing the above steps.

[0010] In another aspect, the present invention provides a method for analyzing cellular composition in a tumor specimen, comprising the steps of: preparing a cellular composition of a cancer clone obtained from cells contained in a tumor specimen based on information on gene mutations, and a cellular composition of a normal cell obtained from cells contained in the tumor specimen based on gene expression levels; Regarding the cell composition of the cancer clone and the cell composition of normal cells, taking into consideration that unclassified cells in the cell composition of the cancer clone include normal cells, adjusting the abundance ratio of each cell type so that the sum of the abundance ratios of each cell type becomes X (X is a constant between 0 and 1); Obtaining the cellular composition of the prepared tumor specimen The method includes: Effect of the Invention

[0011] According to the system and method of the present invention, it is possible to derive the cell composition of a tumor specimen based on the cell composition of a cancer clone and the cell composition of normal cells that are derived independently. This makes it possible to obtain multiple types of information about a specific cell type in a single tumor specimen, making the present invention effective for, for example, predicting the efficacy of a drug or proposing cancer treatments. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 illustrates an example of a system configuration according to a first embodiment. [Diagram 2] FIG. 11 is a diagram showing an example of a setting screen displayed by an input unit according to the first embodiment. [Diagram 3] FIG. 2 is an explanatory diagram showing an example of the data structure of the cell composition of a cancer clone and the cell composition of a normal cell held by the system. [Figure 4] 1 is a flowchart of the first embodiment. [Diagram 5] FIG. 11 is a diagram showing an example of an output screen displayed by an output unit according to the first embodiment. [Figure 6] FIG. 13 is a diagram illustrating an example of a system configuration according to second and third embodiments. [Figure 7] FIG. 13 is a diagram showing an example of a setting screen displayed by an input unit according to the second embodiment. [Figure 8] FIG. 1 is an explanatory diagram showing an example of the data structure of gene mutation information data held by the systems of Examples 2 and 3, gene expression level data obtained from the entire tumor specimen, and gene expression level data for each known normal cell type. [Figure 9] 11 is a flowchart of a second embodiment. [Figure 10] FIG. 13 is a diagram showing an example of an output screen displayed by an output unit according to the second embodiment. [Figure 11] FIG. 1 is a diagram showing an example in which there may be multiple phylogenetic trees showing the parent-child relationships of clones. [Figure 12] FIG. 13 is a diagram showing an example of a setting screen displayed by an input unit according to the third embodiment. [Figure 13] 11 is a flowchart of a third embodiment. [Figure 14] FIG. 13 is a diagram showing an example of an output screen displayed by an output unit according to the third embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] The present invention relates to a system and method for analyzing the composition of cells in a tumor specimen. The tumor specimen is composed of cancer clones and normal cells. The cells contained in the tumor specimen can be classified into cancer clones having specific mutations and other unclassified cells (including normal cells) based on information on gene mutations, and the cell composition classified in this manner is referred to as the "cell composition of cancer clones" in this specification. The cells contained in the tumor specimen can be classified into specific normal cells (e.g., helper T cells, killer T cells, macrophages, etc.) and other unclassified cells (including cancer clones) based on gene expression levels, and the cell composition classified in this manner is referred to as the "cell composition of normal cells" in this specification. In this specification, the cell composition means the ratio of the presence of various cells, i.e., the proportion of presence.

[0014] In the present invention, the subject tumor specimen is not particularly limited, and specimens of any tumor (e.g., solid cancer, sarcoma, etc.) of any species (e.g., mammals, preferably humans, experimental animals) can be used. Collection and processing of tumor specimens can also be performed by methods conventional in the art.

[0015] In the present invention, the composition of cells contained in a tumor specimen is derived by using two separate cell compositions, the cell composition of a cancer clone and the cell composition of a normal cell, and adjusting the abundance ratio of cell types taking into consideration that unclassified cells in the cell composition of a cancer clone include normal cells. Alternatively, the composition of cells contained in a tumor specimen may be derived by adjusting the abundance ratio of cell types taking into consideration that unclassified cells in the cell composition of normal cells include cancer clones.

[0016] In one aspect, the present invention relates to a method for analyzing the cellular composition of a tumor specimen, the method comprising: preparing a cellular composition of a cancer clone obtained from cells contained in a tumor specimen based on information on gene mutations, and a cellular composition of a normal cell obtained from cells contained in the tumor specimen based on gene expression levels; Regarding the cell composition of the cancer clone and the cell composition of normal cells, taking into consideration that unclassified cells in the cell composition of the cancer clone include normal cells, adjusting the abundance ratio of each cell type so that the sum of the abundance ratios of each cell type becomes X (X is a constant between 0 and 1); Obtaining the cellular composition of the prepared tumor specimen may include.

[0017] The cellular composition of the cancer clone is obtained based on the genetic mutation information of the cancer clone. The genetic mutation information varies depending on the type of the target tumor specimen and the type of tumor, but a person skilled in the art can routinely obtain the genetic mutation information and obtain the cellular composition of the cancer clone of the tumor specimen. In a preferred embodiment, the cellular composition of the cancer clone is obtained based on the genetic mutation information obtained from the gene sequences of the tumor specimen and healthy tissue of the same patient.

[0018] The cell composition of normal cells is obtained based on the gene expression levels of each known normal cell type. The gene expression levels of each known normal cell type also vary depending on the type of normal cell to be classified, but a person skilled in the art can obtain the gene expression levels and obtain the cell composition of normal cells in a tumor specimen. In addition, even if it is a cancer clone, if the gene expression level is known, it may be subject to processing using the gene expression level in the same way as normal cells.

[0019] Among the cell compositions of cancer clones, the cell population that is not classified as a cancer clone (unclassified cells) contains normal cells. Taking this into consideration, the abundance ratio of each cell type is adjusted for the cell compositions of cancer clones and normal cells so that the sum of the abundance ratios of each cell type becomes X (X is a constant that satisfies 0≦X≦1, and is a parameter provided by the user).

[0020] In one embodiment, when the sum of the abundance ratios of cancer clones in the cell composition of cancer clones is a and the sum of the abundance ratios of classified normal cells in the cell composition of normal cells is b, If a + b < X, the cell population that cannot be classified into either cancer clones or normal cells is regarded as unclassified cells, and its abundance is derived as X - (a + b). If a + b ≥ X, assuming that the abundance of unclassified cells, which is difficult to estimate, is Y (Y is a constant satisfying 0 ≤ Y ≤ X and is a parameter given by the user), so that the sum of the abundances of classified normal cells and cancer clones and the abundance of unclassified cells becomes X, the abundances of classified normal cells and cancer clones are multiplied by (X - Y) / (a + b), and the abundances of each cell type can be adjusted.

[0021] In another embodiment, when the sum of the abundances of cancer clones in the cell composition of cancer clones is a, the sum of the abundances of normal cells in the cell composition of normal cells is b, and X - a - Y > 0 is satisfied (Y is a constant satisfying 0 ≤ Y ≤ X and is a parameter given by the user), If a + b < X, the abundance of unclassified cells is derived as X - (a + b). If a + b ≥ X, assuming that the abundance of unclassified cells, which is difficult to estimate, is Y, the abundance of normal cells is X - a - Y Thereby, the abundances of each cell type can be adjusted.

[0022] In yet another embodiment, when the sum of the abundances of cancer clones in the cell composition of cancer clones is a, the sum of the abundances of classified normal cells in the cell composition of normal cells is b, and X - b - Y > 0 is satisfied (Y is a constant satisfying 0 ≤ Y ≤ X and is a parameter given by the user), If a + b < X, the abundance of unclassified cells is derived as X - (a + b). If a + b ≥ X, assuming that the abundance of unclassified cells, which is difficult to estimate, is Y, the abundance of cancer clones is X - b - Y Thereby, the abundances of each cell type can be adjusted.

[0023] Also, the method according to the present invention preparing information on gene mutations in cells contained in the tumor specimen, gene expression levels for each known normal cell type, gene expression levels obtained from the entire tumor specimen, and gene expression levels in each normal cell of an arbitrary specimen; obtaining a first cellular composition of cells contained in the tumor specimen; obtaining a second cell composition by updating the first cell composition based on the gene expression levels obtained from the entire tumor specimen and the gene expression levels for each of the known normal cell types; Obtaining a third cell composition by updating the second cell composition based on the information of the gene mutation; Repeating the steps of obtaining a second cellular composition and obtaining a third cellular composition by regarding the third cellular composition as the first cellular composition. This method may further include the steps of: (a) determining whether a cell composition of a tumor specimen is a normal cell composition; (b) determining whether a cell composition of a tumor specimen is a normal cell composition; and (c) determining whether a cell composition of a tumor specimen is a normal cell composition. This method may further include the steps of: (a) determining whether a cell composition of a tumor specimen is a normal cell composition; (b) determining whether a cell composition of a tumor specimen is a normal cell composition; and (c) determining whether a cell composition of a tumor specimen is a normal cell composition. This method may further include the steps of: (a) determining whether a cell composition of a tumor specimen is a normal cell composition;

[0024] In one embodiment, the number of calculations n (n is an integer) specified as a parameter can be used to determine the end of the repeating step of the above method. In another embodiment, the error ε(k) is calculated using the following formula I:

number

[0025] In yet another embodiment, the error ε(k) is calculated according to the following formula II:

number

[0026] The method according to the present invention further comprises the steps of: Prepare multiple cell compositions of cancer clones, For each of the cell compositions of the cancer clones, the step of obtaining the prepared cell composition as a first cell composition together with the cell composition of normal cells, the step of obtaining a second cell composition, the step of obtaining a third cell composition, and the step of repeating the steps are performed; The method may further include selecting the cell composition of the cancer clone that is most consistent with the gene expression levels obtained from the entire tumor specimen, the gene expression levels for each known normal cell type, and the information on the gene mutation. This method makes it possible to select an appropriate cell composition when there are multiple candidates for the cell composition of the cancer clone. The cell compositions of each cancer clone are processed sequentially, and the cell composition of the tumor specimen is calculated from the cell composition of the normal cells and the cell composition of the cancer clone as described above, and the cell composition with the smallest error is obtained.

[0027] In the above method, the error ε(k) is a function represented by the above formula I or formula II, and this error ε(k) can be used to select the cell composition of the cancer clone in the above method.

[0028] The method according to the present invention described above can be implemented more simply and quickly by using a system with an appropriate configuration. Therefore, in another aspect, the present invention is a system for analyzing the cell composition in a tumor specimen, and the system includes at least a calculation unit, at least one storage device included in or connected to the system includes the cell composition of cancer clones obtained based on gene mutation information from cells contained in a tumor specimen and the cell composition of normal cells obtained based on gene expression levels from cells contained in a tumor specimen, the calculation unit Regarding the cell composition of the cancer clones and the cell composition of normal cells read from the storage device, considering that normal cells are included in the unclassified cells in the cell composition of the cancer clones, the sum of the abundance ratios of each cell type becomes X (X is a constant satisfying 0 ≤ X ≤ 1, a parameter specified by the user of the system), adjusting the abundance ratio of each cell type, Outputting the cell composition of the adjusted tumor specimen is executed.

[0029] In one embodiment of the above system, when the sum of the abundance ratios of cancer clones in the cell composition of cancer clones is a and the sum of the abundance ratios of classified normal cells in the cell composition of normal cells is b, the calculation unit if a + b < X, the abundance ratio of unclassified cells is derived as X - (a + b), if a + b ≥ X, assuming that the abundance ratio of unclassified cells for which estimation is difficult is Y (Y is a constant satisfying 0 ≤ Y ≤ X, a parameter given by the user), and multiplying the abundance ratios of classified normal cells and cancer clones by (X - Y) / (a + b), the abundance ratio of each cell type can be adjusted.

[0030] In another embodiment of the above system, when the sum of the abundance ratios of cancer clones in the cell composition of cancer clones is a, and the sum of the abundance ratios of classified normal cells in the cell composition of normal cells is b, and when X - a - Y > 0 (where Y is a constant satisfying 0 ≦ Y ≦ X, a parameter given by the user), the calculation unit if a + b < X, derives the abundance ratio of unclassified cells as X - (a + b), if a + b ≧ X, assumes the abundance ratio of unclassified cells, for which estimation is difficult, as Y, and sets the abundance ratio of normal cells as X - a - Y Thereby, the abundance ratio of each cell type can be adjusted.

[0031] In another embodiment of the above system, when the sum of the abundance ratios of cancer clones in the cell composition of cancer clones is a, and the sum of the abundance ratios of classified normal cells in the cell composition of normal cells is b, and when X - b - Y > 0 (where Y is a constant satisfying 0 ≦ Y ≦ X, a parameter given by the user), the calculation unit if a + b < X, derives the abundance ratio of unclassified cells as X - (a + b), if a + b ≧ X, assumes the abundance ratio of unclassified cells, for which estimation is difficult, as Y, and sets the abundance ratio of cancer clones as X - b - Y Thereby, the abundance ratio of each cell type can be adjusted.

[0032] Also, in the system according to the present invention, at least one storage device included in or connected to the system contains information on gene mutations of cells included in the tumor specimen, gene expression levels for each known normal cell type and gene expression levels obtained from the entire tumor specimen, and gene expression levels in each normal cell of an arbitrary specimen and includes the calculation unit outputs a first cell composition of cells included in the tumor specimen, outputting a second cell composition obtained by updating the first cell composition based on the gene expression levels obtained from the entire tumor specimen and the gene expression levels for each of the known normal cell types; outputting a third cell composition obtained by updating the second cell composition based on the information of the gene mutation; Repeating the steps of outputting the second cell composition and outputting the third cell composition by regarding the third cell composition as the first cell composition. may be executed.

[0033] In one embodiment, the calculation unit uses whether or not the number of iterations reaches n, where n is an integer, designated by a user, to determine the end of the repeating step. In a preferred embodiment, the number of iterations of the repeating step is displayed on an output screen provided in the system or connected to the system.

[0034] In one embodiment, the calculation unit sets the error ε(k) as a function expressed by the formula I or the formula II, and uses the error ε(k) to determine the end of the repeating step. In a preferred embodiment, the number of iterations of the repeating step and the value of the error ε(k) are displayed on an output screen provided in the system or connected to the system.

[0035] Further, in the system according to the present invention, the calculation unit reading out a plurality of cell compositions of cancer clones from the storage device, and executing the steps of inputting the cell composition of each cancer clone together with the cell composition of normal cells as a first cell composition, outputting the second cell composition, outputting the third cell composition, and repeating the steps of using the third cell composition as the first cell composition; A step of selecting and outputting the cell composition of the cancer clone that is most consistent with the gene expression levels obtained from the entire tumor specimen, the gene expression levels for each known normal cell type, and the information on the gene mutation may be further carried out.

[0036] In one embodiment, the calculation unit sets the error ε(k) as a function represented by formula I or formula II, and uses the error ε(k) to select the cell composition of the cancer clone.

[0037] As described above, the cell type composition of cancer clones and normal cells in a tumor specimen obtained or output according to the present invention can be useful for proposing treatment. Biochemical information that stimulates inside and outside the cell is called a signal, and is related to the amount of gene expression. Signals of specific cell types that have infiltrated into the tumor site are used as biomarkers for predicting the success of cancer treatment methods, but in tumor specimens containing a mixture of various cell types, only signals averaged over the constituent cell types can usually be seen. If the composition of cells contained in the specimen can be revealed and the gene expression levels of those cells can be estimated to predict the signals of specific cells in the tumor specimen, it is possible to predict the success of, for example, a drug. One example of a drug is an immune checkpoint inhibitor that targets the gene PD-1 (programmed cell death 1) and its ligand (PD-L1). Currently, various biomarkers are being investigated to determine the administration of immune checkpoint inhibitors, but it has been pointed out that it is difficult to accurately predict the success of immune checkpoint inhibitors alone, even with powerful biomarkers that are said to be related to the success of immune checkpoint inhibitors. For example, the expression level of PD-L1 in cancer cells is used as a biomarker for companion diagnostics of anti-PD-1 / L1 antibodies, but there are cases in which no correlation was observed between the expression of PD-L1 and the effect of the drug. If the present invention makes it possible to measure multiple signals of a specific cell type in one tumor specimen, it is believed that more accurate prediction of response will be possible without increasing the cost of measurement or the burden on patients. JP2019-509282A (Patent Document 1) is a technology for predicting the success of a therapy targeting PD-1 or PD-L1, etc., based on the amount of tumor gene mutation in cancers including melanoma. However, the technology described in Patent Document 1 cannot examine biomarkers other than the amount of tumor gene mutation. In the present invention, based on the derived cell composition, in addition to the amount of tumor gene mutation, information on the cell composition can be obtained, such as signals of specific cell types such as the expression level of PD-L1 in cancer cells, the abundance ratio of immune cells infiltrating the tumor site, and the cell composition of clones.Furthermore, based on this information, it is possible to predict the effects of drugs from multiple angles.

[0038] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. However, the present invention is not to be interpreted as being limited to the description of the embodiment shown below. It will be easily understood by those skilled in the art that the specific configuration can be changed without departing from the concept or purpose of the present invention.

[0039] In the configurations of the invention described below, the same or similar configurations or functions are given the same reference numerals, and duplicated explanations are omitted. EXAMPLES

[0040] In Example 1, the cell composition of the sample tissue is derived based on the cell composition of the cancer clone of the sample tissue, which is a tumor tissue, and the cell composition of the normal cell. Here, the cell composition of the cancer clone is the composition of cells of the sample tissue classified using the frequency of gene mutation as an index, and it is preferable that the cancer clone containing the mutation is classified, and the unclassified cancer clone and the cell group without the mutation (normal cell) are classified as one unclassified cell. That is, the cell composition of the cancer clone includes classified cancer clones and unclassified cells including normal cells. In addition, the cell composition of the normal cell is the composition of cells of the sample tissue classified using the gene expression level as an index, and it is preferable that the cells are classified by cell type, and the normal cells and cancer clones that could not be classified are collected as unclassified cells. That is, the cell composition of the normal cell includes cells classified into a specific cell type and unclassified cells including the cancer clone.

[0041] For two parameters X and Y (0 ≦ Y ≦ X ≦ 1) given by the user, data on the cell composition of cancer clones and the cell composition of normal cells are combined such that the sum of the compositions of each cancer clone and normal cells matches X. At this time, letting the sum of the compositions of cancer clones be a and the sum of the compositions of normal cells be b, the abundance ratio of unclassified cells is made to match X - (a + b) when a + b < X, and Y when a + b ≧ X. When it can be assumed that there are no unclassified cells, it is preferable to set the parameters X and Y to X = 1 and Y = 0. In the following description, the cell composition after combining the two input compositions (the cell composition of cancer clones and the cell composition of normal cells) is referred to as the total cell composition.

[0042] FIG. 1 is a diagram showing an example of the configuration of a computer system. This computer system is composed of one or more computers 101. In FIG. 1, for simplicity of explanation, a computer system composed of one computer 101 is shown.

[0043] The computer 101 has a CPU (Central Processing Unit) 102, a main memory device 103, an auxiliary storage device 104, and an interface (IF) 105. Each hardware is connected via an internal bus.

[0044] The CPU 102 executes a program stored in the main memory device 103. By the CPU 102 executing processing according to the program, it operates as a functional unit (module) that realizes a specific function. In the following description, when explaining the processing with the functional unit as the subject, it indicates that the CPU 102 is executing a program that realizes the module.

[0045] The main memory device 103 is a memory such as a RAM (Random Access Memory), and stores the program executed by the CPU 102 and the data necessary for the execution of the program.

[0046] The main memory device 103 in the first embodiment stores a program for implementing the calculation unit 120, the cell composition 107 of the cancer clone, the cell composition 108 of the normal cell, and the total cell composition 109.

[0047] The calculation unit 120 calculates the total cell composition 109 based on the cell composition 107 of the cancer clone and the cell composition 108 of the normal cell. The calculation unit 120 includes an input unit 121, a processing unit 122, and an output unit 123.

[0048] The input unit 121 reads data specified by a user. The processing unit 122 refers to the cell composition 107 of the cancer clone and the cell composition 108 of the normal cell inputted by the input unit 121, and calculates the total cell composition 109. The output unit 123 outputs the total cell composition 109 and the like.

[0049] Note that the functional units included in the calculation unit 120 may be configured such that multiple functional units are integrated into one functional unit, or one functional unit may be divided into multiple functional units for each function.

[0050] The cell composition of cancer clones 107 is data showing the abundance ratio of one or more cancer clones or cells without mutations, and the type of mutation contained in the cancer clone may be displayed. The cell composition of normal cells 108 is data showing the abundance ratio of one or more cell types of normal cells and unclassified cells. The two data 107 and 108 are preferably data on the same specimen.

[0051] The auxiliary storage device 104 is a storage device such as a hard disk drive (HDD) or a solid state drive (SSD), and stores data permanently.

[0052] The interface (IF) 105 is an interface for connecting to or communicating with an external device. The interface 105 is, for example, an I / O interface, a network interface, etc. The computer 101 can have an interface 105 that connects to an external storage device 125 via a network 124, and an interface 105 that connects to an input / output device 126. Note that a terminal having the input / output device 126 may be connected to the computer 101 via the network 124.

[0053] The external storage device 125 is a device that provides a storage area for storing data handled by the computer 101. The network 124 connected to the external storage device 125 is, for example, a LAN (Local Area Network) or the Internet. However, the present invention is not limited to the type of the network 124. The connection method of the network 124 may be either wired or wireless. Note that, instead of the external storage device 125 and the network 124, a storage medium such as a CD or DVD that is detachable from the computer 101 may be used.

[0054] The input / output device 126 is a device used by a user to operate the computer 101. The input / output device 126 is composed of a keyboard, a mouse, a touch panel, a display, etc. The computer 101 may have the input / output device 126 built in.

[0055] The cell composition 107 of the cancer clone, the cell composition 108 of the normal cell, and the total cell composition 109 stored in the main memory device 103 may be stored in either the auxiliary memory device 104 or the external memory device 125. The symbols (A), (B), and (C) in Fig. 1 are used to distinguish information stored in different locations.

[0056] In this embodiment, when the computer 101 is started up, the CPU 102 reads out the cell composition 107 of the cancer clone, the cell composition 108 of the normal cell, and the total cell composition 109 stored in either the auxiliary storage device 104 or the external storage device 125, and loads them into the main storage device 103. In addition, when the computer 101 is turned off or when the free space of the main storage device 103 is insufficient, the computer 101 may move or copy information from the main storage device 103 to another storage device.

[0057] The process executed in Example 1 will be described in detail below. First, input data is set on the setting screen in FIG. 2. FIG. 3 shows the structure of data to be input to this system. This system is executed according to the flow explained in FIG. 4 to derive the total cell composition. Furthermore, an example of the output screen of the total cell composition output by this system is shown in FIG. 5.

[0058] 2 shows an example of a setting screen 201 when a user selects and sets input data using the system of the present invention. The setting screen 201 is displayed on the input / output device 126 using the input unit 121 of the calculation unit 120.

[0059] The user can input a specimen ID in the first line 202 and set it to specify the cell composition 107 of the cancer clone and the cell composition 108 of the normal cell in the specimen. Note that the user can also specify the file paths of the cell composition 107 of the cancer clone and the cell composition 108 of the normal cell in the second line 203 and the third line 204.

[0060] When the above settings are made and a setting button 205 is pressed, the set information is sent to the input unit 121.

[0061] FIG. 3 is an explanatory diagram showing an example of the data structure of the cell composition 107 of the cancer clone and the cell composition 108 of the normal cell held by the computer system of the system.

[0062] First, the cell composition 107 of cancer clones includes, for each of one or more cancer clones, the mutations possessed by that clone and the abundance ratio of that clone. Note that the last line representing the total of all clone-cell types may be omitted. Here, the illustrated sum a is the sum of the compositions of the cancer clones excluding unclassified cells in the cell composition of the cancer clones.

[0063] The cell composition 108 of normal cells includes the abundance ratio for one or more cell types of normal cells. Note that the last line representing the total of all clone-cell types may be omitted. Here, the illustrated sum b is the sum of the compositions of the cells excluding unclassified cells in the cell composition of the normal cells.

[0064] FIG. 4 is a flowchart for explaining a typical processing procedure of deriving the total cell composition S301 executed by the system of Example 1. While referring to FIGS. 3 and 4, the processing flow of Example 1 will be described.

[0065] First, the processing in the input unit 121 of the calculation unit 120 will be described. Input data is selected and read from the setting screen of FIG. 2 by the input / output device 126 (step S302).

[0066] Next, the processing in the processing unit 122 of the calculation unit 120 will be described. Based on the data read in step S302, the sum a of the compositions of the cancer clones and the sum b of the compositions of the normal cells are calculated (step S303). Next, it is determined whether a + b < X is satisfied for the calculated a and b (step S304). If it is determined that it is satisfied, the abundance ratio of the unclassified cells is set to X - (a + b) (step S305). If it is determined that it is not satisfied, the compositions of the classified cells and the cancer clones are multiplied by (X - Y) / (a + b), and the abundance ratio of the unclassified cells is set to Y (step S306).

[0067] For example, when X = 1 in the data structure example of FIG. 3, in step S304, since a = 0.5, b = 0.45, and a + b = 0.95 < X, the process proceeds to step S305, and the unclassified cells are calculated as X - 0.95 = 0.05. Note that the ratios (abundance ratios) of the cell types of the cancer clone and normal cells are directly inherited as the values described in FIG. 3.

[0068] Here, in step S306, for example, if it is known in advance that the cell composition of the cancer clone among the two input compositions is obtained with better accuracy, the composition of the cancer clone may be fixed, and only the composition of the classified normal cells may be set to X - b - Y. Similarly, the composition of the classified normal cells may be fixed, and only the composition of the cancer clone may be set to X - a - Y.

[0069] Finally, the process in the output unit 123 of the calculation unit 120 will be described. Based on the abundance ratios of the cell types calculated and updated in step S305 or S306, the total cell composition 109 is created (step S307). Finally, the created data is transmitted to the input / output device 126 (step S308), and the process ends (step S309).

[0070] FIG. 5 shows an example 401 of the output screen displayed by the output unit of Example 1. The total cell composition 109 transmitted in step S308 is output to the input / output device 126 as shown in the output screen 401. The total cell composition 109 includes the cell types of normal cells (including unclassified cells) and the compositions for each cancer clone.

Example

[0071] In Example 2, the total cell composition is derived based on the expression levels and gene mutation information in the tumor specimen. Example 2 is characterized by inputting the initial value of the total cell composition and performing iterative calculations to update the total cell composition. Hereinafter, Example 2 will be described focusing on the differences from Example 1.

[0072] 6 is a diagram showing an example of the configuration of the system of Example 2. The difference is that the main memory device 103, the external memory device 125, and the auxiliary memory device 104 contain information on gene mutation 141, gene expression levels 142 obtained from the entire tumor specimen, and gene expression levels 143 for each known normal cell type, instead of the cell composition 107 of the cancer clone and the cell composition 108 of the normal cell. Also, the process executed by the processing unit 122 of the calculation unit 120 is different.

[0073] FIG. 7 is an example of a setting screen 701 displayed by the input unit of the second embodiment. The difference is that, in the input data at the top, instead of the file input field for the cell composition 107 of the cancer clone and the file input field for the cell composition 108 of the normal cell, a file input field 221 for the information on gene mutation 141, a file input field 222 for the gene expression level 142 obtained from the entire tumor specimen, and a file input field 223 for the gene expression level 143 for each known normal cell type are included. In addition, in the parameters at the bottom, the input field 224 on the first line specifies the path of the file of the initial value of the total cell composition 109. As in the input field 225 on the second line, it may be possible to select whether to set the allowable error ε. In that case, the allowable error ε can be set by checking the input field 225. In the input field 228 on the next line, it may be possible to continuously select whether to prioritize time or accuracy. Depending on the selection in the input field 228, values ​​may be automatically input into the allowable error ε and the maximum number of calculations n in the input fields 226 and 227. Alternatively, it may be possible to directly input numerical values ​​into the input fields 226 and 227. Note that the input fields 225, 226, 227, and 228 do not necessarily have to be provided.

[0074] FIG. 8 is an explanatory diagram showing an example of the data structure of gene mutation information 141, gene expression data 142 obtained from the entire tumor specimen, and gene expression data 143 for each known normal cell type, which are held by the system of the second embodiment. First, in the gene mutation information 141, a gene mutation ID and a numerical value related to the allele frequency are included in one gene mutation information. A gene mutation ID is a code assigned to one gene mutation, and different gene mutations are assigned different gene mutation IDs. Here, different gene mutations refer to differences in either the gene in which the mutation occurred, the position of the mutation in the sequence, or the base type after the mutation. In the gene expression data 142 obtained from the entire tumor specimen, a gene ID and a numerical value related to the expression level are included in one gene information. In the gene expression data 143 for each known normal cell type, a gene and its expression level data are included in one gene information for each cell type. Note that the cell type of normal cells listed in the table is an example.

[0075] Fig. 9 is a flow chart for explaining a typical processing procedure of the processing for improving the total cell composition executed by the system of Example 2. The flow of the processing of Example 2 will be explained with reference to Figs.

[0076] First, the processing in the input unit 121 of the calculation unit 120 will be described. Input data and parameters are selected and read from the setting screen in FIG. 7 by the input / output device 126 (step S502). Here, the initial value of the total cell composition, which is a parameter, is stored in the system as P(0). If the specification of the initial value is omitted, an arbitrarily determined value such as a random number may be used as the initial value. Note that the following description assumes that the allowable error ε is set in the input field 225. Note that if it is not set, the allowable error ε can be given in advance as a parameter of a fixed value.

[0077] Next, the processing in the processing unit 122 of the calculation unit 120 will be described. First, the iteration parameter k is initialized to 0 (step S503). Next, 1 is added to k (step S504), and the initial value of the total cell composition is set as P(k-1), and the cell composition of the cancer clone is derived based on the mutation data (step S505).

[0078] Examples of the method used in step S505 include, but are not limited to, principal component analysis, non-negative matrix factorization, random forest, neural network, support vector regression, etc.

[0079] What is derived in step S505 is the cell composition of cancer clones. Therefore, the normal cell composition part of the whole cell composition P(k - 1) is combined with the cell composition of the cancer clones derived in step S505 and normalized to derive the whole cell composition P(k) (step S506). This step S506 is performed according to the procedures of steps S303 to S307 in Example 1 of FIG. 4.

[0080] Regarding the iteration parameter k, it is determined whether k < n holds based on the maximum number of calculations n (step S507). The maximum number of calculations n is a parameter input from the setting screen 701 in step S502.

[0081] Next, using P(k) obtained in step S506 as the initial value of the whole cell composition, the whole cell composition is derived based on the gene expression level data obtained from the entire tumor specimen and the gene expression levels of known normal cell types, and this is set as P(k + 1) (step S508).

[0082] Examples of the algorithm used in step S508 include, but are not limited to, principal component analysis, non-negative matrix factorization, random forest, neural network, support vector regression, etc., similar to step S505.

[0083] Then, 1 is added to k (step S509). Next, regarding the iteration parameter k and the error ε(k), it is determined whether k < n and ε(k) > ε hold (step S510). The maximum number of calculations n and the allowable error ε are parameters input from the setting screen 701 in step S502. Here, as the error ε(k), any function considered appropriate by the user can be used, and for example, the function represented by the above formula I may be used. If step S510 is true, the process returns to step S504, and the calculation is carried out in the same manner.

[0084] If the judgment formulas in steps S507 and S510 are not satisfied, it is judged whether k=n for the iteration parameter k (step S511). If the condition is satisfied, P(m) for i=m that gives the minimum value for the error ε(i) (i=1, ..., n) in each of the n calculation results calculated in the above steps is selected as the whole-cell composition P' (step S512). If the condition in step S511 is not satisfied, the last calculated whole-cell composition P(k) is selected as the whole-cell composition P' (step S513).

[0085] Finally, the process in the output unit 123 of the calculation unit 120 will be described. Output data is created based on the total cell composition P' set in step S512 or S513 and the error ε' in the total cell composition, and is transmitted to the input / output device 126 (step S514), and the process ends (step S515).

[0086] FIG. 10 is a diagram showing an example of an output screen displayed by the output unit of the second embodiment. The input / output device 126 outputs the total cell composition 109 transmitted in step S514 as shown in the output screen 1001. The total cell composition 109 includes compositions for each of the cell types of normal cells (including unclassified cells) and cancer clones. The error ε' 421 in the total cell composition is output at the bottom. In order to make the error 421 easier to interpret, the result of evaluating the accuracy 422 of the output in three stages may be displayed. In that case, it is preferable to present the user in advance with a guideline for interpretation, such as, for example, if the evaluation of the accuracy is "high", the calculation result can be used, if it is "medium", the calculation result can be used but attention should be paid to the accuracy, and if it is "low", further repeated iteration of the calculation or confirmation of the accuracy of the input data is required. In addition, in order to accurately grasp the progress of the error in the iterative calculation, the result may be displayed in a graph 423. EXAMPLES

[0087] Example 3 describes a method for selecting the optimal total cellular composition when multiple cellular compositions are possible for a single tumor specimen.

[0088] An example of a situation in which such a method is required will be described. FIG. 11 shows an example 601 of a mutation in a tumor specimen and a list 602 of clone phylogenetic trees that can reproduce the mutation. Note that a phylogenetic tree in this specification is a tree-like diagram that illustrates the relationship obtained by connecting clones that have evolved directly due to the occurrence of a mutation. For example, when three types of mutations are contained in the cells constituting a specimen at a rate of 10%, 40%, and 50%, respectively, a phylogenetic tree showing the evolution of six types of cancer clones can be considered as shown in 602. If there is no other information available, the true phylogenetic tree cannot be determined from the six types of phylogenetic trees. This embodiment provides a method for selecting an appropriate cell composition when there is a possibility that multiple cell compositions exist.

[0089] Fig. 12 is a diagram showing an example of a setting screen 1201 displayed by the input unit of Example 3. The difference is that the setting screen 701 of Example 2 in Fig. 7 does not include a field for inputting an initial value of the total cell composition 109.

[0090] Fig. 13 is a flow chart for explaining a typical processing procedure of improving the total cell composition S701 executed by the system of Example 3. The flow of processing of Example 3 will be explained with reference to Figs. 13, 9 and 4.

[0091] First, the process in the input unit 121 of the calculation unit 120 will be described. Input data and parameters are selected and read from the setting screen 1201 in Fig. 12 by the input / output device 126 (step S702). Note that the following description will be given on the assumption that the allowable error ε is set in the input field 225.

[0092] Next, the processing in the processing unit 122 of the calculation unit 120 will be described. First, the cell composition of normal cells is derived based on the gene expression levels obtained from the entire input tumor specimen and the gene expression levels of each known normal cell type (step S703). Next, one or more cell compositions of cancer clones are derived based on the input gene mutation information (step S704).

[0093] The algorithm used in steps S703 and S704 is not limited, as in step S505 in Example 2 (FIG. 9). The cell composition of the cancer clone derived in step S704 may be multiple. In this example, multiple cases are mainly assumed.

[0094] Next, a variable c corresponding to the cellular composition of each cancer clone is initialized to 0 (step S705). Then, 1 is added to c to obtain one cellular composition of the cancer clone derived in step S704 (step S706). The cellular composition of the normal cells derived in step S703 and the cellular composition of the cancer clone obtained in step S706 are combined and normalized to derive the total cellular composition P(0) (step S707). This step S707 is performed in the order of steps S303 to S307 of Example 1 in FIG. 4.

[0095] Next, the whole cell composition is updated using the whole cell composition P(0) derived in step S707 as an initial value (step S708). This step S707 is performed in the order of steps S503 to S513 in the second embodiment of FIG. 9. The whole cell composition P' derived in step S512 or S513 is updated by P c Then, it is determined whether there is a cell composition of a cancer clone that has not yet been obtained (step S710), and if there is, the process proceeds to step S706.

[0096] If there is no unobtained cell composition of the cancer clone in step S710, the error ε c Calculate the error ε c The list of cell compositions of cancer clones is sorted in ascending order of error εc The ranks sorted in ascending order are called composition rank 441.

[0097] Finally, a description will be given of the process in the output unit 123 of the calculation unit 120. The output data created in step S711 is sent to the input / output device 126 (step S712), and the process ends (step S713).

[0098] Fig. 14 is a diagram showing an example of an output screen displayed by the output unit of Example 3. The whole cell composition 109 transmitted in step S712 is output to the input / output device 126 as in output screen 1401. The screen differs from Fig. 10 in that it includes a field 441 for selecting the whole cell composition to be viewed at the top of the screen. Fig. 14 is an example of the screen when the whole cell composition with the number 1 composition rank 441 (i.e. the whole cell composition estimated to have the smallest error) is selected.

[0099] The above-mentioned configurations, functions, processing units, processing means, etc. may be partly or entirely realized in hardware, for example, by designing them as integrated circuits. The present invention can also be realized by software program code that realizes the functions of the embodiments. In this case, a storage medium on which the program code is recorded is provided to a computer, and a processor included in the computer reads the program code stored in the storage medium. In this case, the program code itself read from the storage medium realizes the functions of the above-mentioned embodiments, and the program itself and the storage medium storing it constitute the present invention. Examples of storage media for supplying such program code include flexible disks, CD-ROMs, DVD-ROMs, hard disks, SSDs (Solid State Drives), CD-Rs, DVD-Rs, magnetic tapes, non-volatile memory cards, ROMs, etc.

[0100] Furthermore, the program code for realizing the functions described in this embodiment can be implemented in a wide range of program or script languages, such as, for example, assembler, C / C++, Perl, Python, and the like.

[0101] Furthermore, the program code of the software that realizes the functions of the embodiments may be distributed over a network and stored in a storage means such as a computer's hard disk or memory, or in a storage medium such as a CD-RW or CD-R, and the processor of the computer may read out and execute the program code stored in the storage means or storage medium.

[0102] In the above-mentioned embodiment, the control lines and information lines are shown as those considered necessary for the explanation, and not all the control lines and information lines are shown in the product. All the components may be connected to each other.

[0103] The present invention is not limited to the above-mentioned embodiment, but includes various modified examples. For example, the above-mentioned embodiment describes the configuration in detail in order to easily explain the present invention, and the present invention is not necessarily limited to the configuration including all the described configurations. Also, it is possible to add, delete, or replace a part of the configuration of each embodiment with another configuration. [Explanation of symbols]

[0104] 101 Calculator 102 CPU 103 Main storage 104 Auxiliary storage 105 Interface (IF) 107 Cellular composition of cancer clones 108 Cellular composition of normal cells 109 Whole cell composition 120 Calculation Department 121 Input section 122 Processing section 123 Output section 124 Network 125 External storage device 126 I / O Devices 141 Gene Mutation Information Gene expression levels from 142 tumor samples 143 Gene expression levels for known normal cell types 201, 701, 1201 setting screen 401, 1001, 1401 Output screen

Claims

1. 1. A system for analyzing a cellular composition in a tumor specimen, comprising: At least a calculation unit is provided, At least one storage device included in the system or connected to the system includes a cell composition of a cancer clone obtained based on information on gene mutations from cells contained in a tumor specimen, and a cell composition of a normal cell obtained based on gene expression levels from cells contained in a tumor specimen, The calculation unit: a step of adjusting the abundance ratio of each cell type with respect to the cell composition of the cancer clone and the cell composition of normal cells read from the storage device, taking into consideration that unclassified cells in the cell composition of the cancer clone include normal cells, so that the sum of the abundance ratios of each cell type becomes X (X is a constant between 0 and 1, and is a parameter designated by a user of the system); Outputting the cellular composition of the adjusted tumor specimen Run (1) The sum of the abundance ratios of the cancer clones in the cell composition of the cancer clones is a, When the sum of the abundance ratios of the classified normal cells in the cell composition of the normal cells is b, the calculation unit If a+b<X, a cell population that is not classified as either a cancer clone or a normal cell is defined as an unclassified cell, and its abundance ratio is calculated as X-(a+b); If a+b≧X, the abundance ratio of the classified normal cells and cancer clones is multiplied by (X−Y) / (a+b) and the abundance ratio of the unclassified cells is set to Y (Y is a constant between 0 and X, and is a parameter designated by a user of the system), thereby adjusting the abundance ratio of each cell type; or (2) the sum of the abundance ratios of the cancer clones in the cell composition of the cancer clones is a, When the sum of the abundance ratios of classified normal cells in the cell composition of the normal cells is b, and X-a-Y>0 is satisfied (Y is a constant not less than 0 and not more than X, and is a parameter designated by a user of the system), The calculation unit: If a+b<X, a cell population that is not classified as either a cancer clone or a normal cell is defined as an unclassified cell, and its abundance ratio is calculated as X-(a+b); If a+b≧X, the abundance ratio of the normal cells is set to X−a−Y, and the abundance ratio of the unclassified cells is set to Y, thereby adjusting the abundance ratio of each cell type; or (3) the sum of the abundance ratios of the cancer clones in the cell composition of the cancer clones is a, When the sum of the abundance ratios of classified normal cells in the cell composition of the normal cells is b, and X-b-Y>0 is satisfied (Y is a constant not less than 0 and not more than X, and is a parameter designated by a user of the system), The calculation unit: If a+b<X, a cell population that is not classified as either a cancer clone or a normal cell is regarded as an unclassified cell, and its abundance ratio is calculated as X-(a+b); If a+b≧X, the abundance ratio of the cancer clones is set to X−b−Y, and the abundance ratio of the unclassified cells is set to Y, thereby adjusting the abundance ratio of each cell type. A system characterized by:

2. 2. The system of claim 1, At least one storage device provided in or connected to the system, Information on genetic mutations in cells contained in the tumor sample; The gene expression levels of each known normal cell type and the gene expression levels obtained from the entire tumor sample, The gene expression level in each normal cell of any sample and Including, The calculation unit: outputting a first cellular composition of cells contained in the tumor specimen; outputting a second cell composition obtained by updating the first cell composition based on the gene expression levels obtained from the entire tumor specimen and the gene expression levels for each of the known normal cell types; outputting a third cell composition obtained by updating the second cell composition based on the information of the gene mutation; Repeating the steps of outputting the second cell composition and outputting the third cell composition by regarding the third cell composition as the first cell composition. A system for executing the above.

3. 3. The system of claim 2, The system is characterized in that the calculation unit uses whether or not the number of iterations has reached n, where n is a calculation number n (n is an integer) specified by a user, to determine the end of the repeating step.

4. 4. The system of claim 3, A system characterized in that the number of iterations of the repeating step is displayed on an output screen provided by the system or connected to the system.

5. 3. The system of claim 2, The calculation unit calculates the error ε(k) according to the following formula I: [0010] [wherein k means the kth iteration, T ｉ is the gene expression level obtained from the whole tumor sample, and C ｉｊ (k) is the gene expression level for each cell type, and P ｊ (k) is the abundance ratio for each cell type. The subscript i is a given gene, and the subscript j is a given cell type. and ε(k) is used to determine whether the repeating step is completed.

6. 6. The system of claim 5, The system further comprises an output screen provided in the system or connected to the system, the output screen displaying the number of iterations of the repeating step and the numerical value of the error ε(k).

7. 3. The system of claim 2, The calculation unit: reading out a plurality of cell compositions of cancer clones from the storage device, and executing the steps of inputting the cell composition of each cancer clone together with the cell composition of normal cells as a first cell composition, outputting the second cell composition, outputting the third cell composition, and repeating the steps of using the third cell composition as the first cell composition; The system further comprises a step of selecting and outputting a cell composition of a cancer clone that is most consistent with the gene expression levels obtained from the entire tumor specimen, the gene expression levels for each known normal cell type, and the information on the gene mutation.

8. 8. The system of claim 7, The calculation unit calculates the error ε(k) according to the following formula I: [0025] [wherein k means the kth iteration, T ｉ is the gene expression level obtained from the whole tumor sample, and C ｉｊ (k) is the gene expression level for each cell type, and P ｊ (k) is the abundance ratio for each cell type. The subscript i is a given gene, and the subscript j is a given cell type. and using ε(k) to select the cell composition of a cancer clone.

9. 1. A method for analyzing a cellular composition in a tumor specimen, comprising: preparing a cellular composition of a cancer clone obtained from cells contained in a tumor specimen based on information on gene mutations, and a cellular composition of a normal cell obtained from cells contained in the tumor specimen based on gene expression levels; Regarding the cell composition of the cancer clone and the cell composition of normal cells, taking into consideration that normal cells are included in unclassified cells in the cell composition of the cancer clone, adjusting the abundance ratio of each cell type so that the sum of the abundance ratios of each cell type becomes X (X is a constant between 0 and 1); Obtaining the cellular composition of the prepared tumor specimen wherein in the adjusting step, (1) The sum of the abundance ratios of the cancer clones in the cell composition of the cancer clones is a, When the sum of the abundance ratios of the classified normal cells in the cell composition of the normal cells is b, if a+b<X, a cell population that is not classified as either a cancer clone or a normal cell is regarded as an unclassified cell, and its abundance ratio is calculated as X-(a+b); If a+b≧X, the abundance ratios of the classified normal cells and cancer clones are multiplied by (X−Y) / (a+b) and the abundance ratio of the unclassified cells is set to Y (Y is a constant between 0 and X, and is a parameter designated by the user), thereby adjusting the abundance ratios of the cell types; or (2) the sum of the abundance ratios of the cancer clones in the cell composition of the cancer clones is a, When the sum of the abundance ratios of classified normal cells in the cell composition of the normal cells is b, and X-a-Y>0 is satisfied (Y is a constant not less than 0 and not more than X, and is a parameter designated by the user), If a+b<X, a cell population that is not classified as either a cancer clone or a normal cell is defined as an unclassified cell, and its abundance ratio is calculated as X-(a+b); If a+b≧X, the abundance ratio of the normal cells is set to X−a−Y, and the abundance ratio of the unclassified cells is set to Y, thereby adjusting the abundance ratio of each cell type; or (3) the sum of the abundance ratios of the cancer clones in the cell composition of the cancer clones is a, When the sum of the abundance ratios of the classified normal cells in the cell composition of the normal cells is b, and X-b-Y>0 is satisfied (Y is a constant not less than 0 and not more than X, and is a parameter designated by the user), If a+b<X, a cell population that is not classified as either a cancer clone or a normal cell is regarded as an unclassified cell, and its abundance ratio is calculated as X-(a+b); When a+b≧X, the abundance ratio of the cancer clones is set to X−b−Y, and the abundance ratio of the unclassified cells is set to Y, thereby adjusting the abundance ratio of each cell type.

10. preparing information on gene mutations in cells contained in the tumor specimen, gene expression levels for each known normal cell type, gene expression levels obtained from the entire tumor specimen, and gene expression levels in each normal cell of an arbitrary specimen; obtaining a first cellular composition of cells contained in the tumor specimen; obtaining a second cell composition by updating the first cell composition based on the gene expression levels obtained from the entire tumor specimen and the gene expression levels for each of the known normal cell types; Obtaining a third cell composition by updating the second cell composition based on the information of the gene mutation; Repeating the steps of obtaining a second cellular composition and obtaining a third cellular composition by regarding the third cellular composition as the first cellular composition. The method of claim 9 further comprising:

11. Prepare multiple cell compositions of cancer clones, For each of the cell compositions of the cancer clones, the step of obtaining the prepared cell composition as a first cell composition together with the cell composition of normal cells, the step of obtaining a second cell composition, the step of obtaining a third cell composition, and the step of repeating the steps are performed; The method according to claim 10, further comprising selecting a cell composition of a cancer clone that is most consistent with the gene expression levels obtained from the entire tumor specimen, the gene expression levels for each known normal cell type, and the information on the gene mutation.

12. The method according to claim 9, wherein the cell composition of the cancer clone is obtained based on information of gene mutations obtained from gene sequences of a tumor specimen and healthy tissue of the same patient.

Images