Method for detecting de novo mutation in incomplete trio with missing parent information

Abstract

This method for detecting de novo mutation comprises the steps in which: an analysis device acquires family genomic observation information about an analysis subject from which genomic information of at least one of the parents is missing; the analysis device analyzes a genome inheritance state on the basis of the family genomic observation information; the analysis device determines haploid information of an analysis subject family on the basis of the analyzed genomic inheritance state; the analysis device replaces the missing genomic information of the at least one of the parents with genomic information of a sibling of the analysis subject on the basis of the determined haploid information of the analysis subject family; and on the basis of parent genomic information including the replaced genomic information and genomic information of the analysis subject, the analysis device searches for whether a de novo mutation occurs.

Description

A method for detecting de novo mutations in incomplete trios with missing parental information.

[0001] The technique described below describes a method for detecting de novo mutations.

[0002] De novo mutations are genetic mutations that do not occur in the parental generation and arise in the offspring generation. De novo mutations are the cause of congenital diseases and a driving force behind evolution. The factors that influence the incidence of de novo mutations are not fully understood. However, it is known that the occurrence of de novo mutations is influenced by factors such as the parents' age.

[0003] [Prior Art Literature]

[0004] [Non-patent literature]

[0005] Open paper: Acuna-Hidalgo, R., Veltman, JA & Hoischen, A. New insights into the generation and role of de novo mutations in health and disease. Genome Biol 17, 241 (2016). https: / / doi.org / 10.1186 / s13059-016-1110-1

[0006] De novo mutations are identified through whole-genome sequencing. Specifically, de novo mutations are identified by sequencing the entire genomes of the parents (father, mother) and the child and comparing them. However, in reality, it is difficult to determine de novo mutations using parental genomic information. This is due to the difficulty of obtaining parental genomic information. For example, if the proband is elderly, it may be difficult to obtain their genomes because the parents are no longer alive.

[0007] The technology described below aims to address the aforementioned issues. Specifically, the technology described below aims to identify de novo mutations by utilizing sibling genetic information instead when parental genetic information is unavailable.

[0008] A de novo mutation detection method comprises the steps of: obtaining, by an analysis device, family genome observation information of a subject of analysis whose genome information of at least one parent is missing; analyzing, by the analysis device, a genome inheritance state based on the family genome observation information; determining, by the analysis device, haploid information of the subject family of analysis based on the analyzed genome inheritance state; replacing, by the analysis device, sibling genome information of the subject of analysis with the missing genome information of at least one parent based on the determined haploid information of the subject family of analysis; and searching, by the analysis device, for the presence or absence of a de novo mutation based on the genome information of the parents including the replaced genome information and the genome information of the subject of analysis.

[0009] The technology described below can be used to detect de novo mutations in a subject of analysis. In particular, the technology described below can detect de novo mutations even when all or part of the subject's parents' genomic information is not available.

[0010] Figure 1 shows the overall process of the analysis device (100) performing the de novo mutation detection method.

[0011] Figure 2 is a flowchart (200) of one embodiment of a de novo mutation detection method.

[0012] Figure 3 shows the number of possible genetic inheritance states that can be analyzed when two siblings inherit a non-recombinant haploid from their parents.

[0013] Figure 4 shows the number of cases of family genome observation information that can be observed from one sibling.

[0014] Figure 5 shows the familial genome observation information that can be observed when two siblings inherit a non-recombinant haploid from their parents.

[0015] Figure 6 is a table showing all cases of genetic inheritance states that can arise from a linear trio and the haploid information of the family that can be determined accordingly.

[0016] Figure 7 is a table showing all cases of genetic inheritance status that can arise from the inverted trio and the haploid information of the family that can be determined accordingly.

[0017] Figure 8 shows one example of searching for de novo mutations in a linear trio.

[0018] Figure 9 shows one example of searching for de novo mutations in the reverse trio.

[0019] Figure 10 shows an example of dividing a genome into several regions and analyzing the genome inheritance status for each region in the process of analyzing the genome inheritance status.

[0020] Figure 11 shows one embodiment of setting an analyzable region in a linear trio and an inverse trio.

[0021] Figure 12 shows the overall structure of the sibling HMM and the inverse trio HMM.

[0022] Figure 13 shows the priority among family genome observation information.

[0023] Figure 14 shows the results of analyzing the accuracy of sibling HMM and inverse trio HMM.

[0024] Figure 15 shows the results of measuring whether the sibling HMM and trio HMM can detect de novo mutations well in artificially created incomplete trios targeting 33 CEPH families with many children.

[0025] Figure 16 shows the results of estimating the number of samples required to detect a statistically significant difference in de novo mutation rate compared to the control group based on the accuracy of the current method.

[0026] Figure 17 shows that as the number of siblings increases under the Linear Trio and Reverse Trio conditions, the number of de novo mutations that can be detected by the de novo mutation detection method increases.

[0027] Figure 18 shows the results of applying the de novo mutation detection method to data on patients and families with undiagnosed rare diseases.

[0028] Figure 19 shows the results of applying the de novo mutation detection method in a case where there is no information on both parents and there are five children.

[0029] Fig. 20 is a configuration of one embodiment of an analysis device (300).

[0030] The technology described below is susceptible to various modifications and embodiments. Specific embodiments of the technology described below may be illustrated in the drawings of the specification. However, these are intended to illustrate the technology described below and are not intended to limit the technology described below to any specific embodiments. Therefore, it should be understood that all modifications, equivalents, or alternatives that fall within the spirit and scope of the technology described below are encompassed by the technology described below.

[0031] Terms such as first, second, A, and B may be used to describe various components. However, these terms are only used to distinguish one component from other components and are not intended to limit the components. For example, without departing from the scope of the technology described below, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. The term "and / or" includes any combination of multiple related listed items or any one of multiple related listed items.

[0032] In the terms used hereinafter, singular expressions should be understood to include plural expressions unless the context clearly dictates otherwise, and terms such as "comprises" should be understood to mean the presence of a described feature, number, step, operation, component, part, or combination thereof, but not to exclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0033] Before going into a detailed description of the drawings, it should be made clear that the division of components in this specification is merely a division based on the main function of each component. That is, two or more components described below may be combined into one component, or one component may be further divided into two or more components with more detailed functions. In addition to its own main function, each component described below may additionally perform some or all of the functions of other components, and of course, some of the main functions of each component may be exclusively performed by other components.

[0034] Additionally, in performing a method or method of operation, each process constituting the method may occur in a different order than the stated order, unless the context clearly indicates a specific order. That is, each process may occur in the same order as the stated order, may be performed substantially simultaneously, or may be performed in the opposite order.

[0035] First, the terms used in the technology described below will be explained.

[0036] In the technology described below, siblings may refer to children born from the same parents, regardless of gender. That is, in the technology described below, siblings may include male-male, male-female, and female-female children born from the same parents.

[0037] In the technology described below, an incomplete trio may refer to a case where genetic information of at least one parent is missing. For example, an incomplete trio may refer to a case where there are three children without parents. Alternatively, an incomplete trio may refer to a case where there are one parent and two children. In the technology described below, types of incomplete trios may include a linear trio and a reverse trio. In the technology described below, a linear trio may refer to a case where genetic information of both parents is missing. For example, a linear trio may refer to a case where there are at least three children, including the proband. In the technology described below, a reverse trio may refer to a case where genetic information of one of the two parents is missing. For example, a reverse trio may refer to a case where there are one parent and at least two children, including the proband.

[0038] Below we examine the overall process by which the analysis device performs the de novo mutation detection method.

[0039] Figure 1 shows the overall process of the analysis device (100) performing the de novo mutation detection method.

[0040] The analysis device (100) can be physically implemented in various forms. For example, the analysis device (100) can take the form of a PC, a laptop, a smart device, a server, or a data processing-dedicated chipset.

[0041] The analysis device (100) may be singular or multiple. That is, the above-described de novo mutation detection method may be performed by at least one analysis device (100).

[0042] The analysis device (100) may be a device that performs a de novo mutation detection method. Specifically, the analysis device (100) may obtain family genome observation information of an analysis subject whose genome information of at least one parent is missing. The analysis device (100) may analyze the inheritance state of the genome based on the family genome observation information. The analysis device (100) may determine haploid information of the analysis subject's family based on the analyzed genome inheritance state. The analysis device (100) may replace the sibling genome information of the analysis subject with the genome information of at least one parent who is missing based on the determined haploid information of the analysis subject's family. The analysis device (100) may detect whether a de novo mutation exists based on the genome information of the parents including the replaced genome information and the genome information of the analysis subject.

[0043] Below, the de novo mutation detection method is described in detail.

[0044] Figure 2 is a flowchart (200) of one embodiment of a de novo mutation detection method.

[0045] The analysis device can obtain family genome observation information of an analysis subject whose genome information of at least one parent is missing (210).

[0046] The subject of analysis may be a person suspected of having a de novo mutation. For example, the subject may have a trait not present in the parents, and it may be desired to determine whether this trait is due to a de novo mutation.

[0047] Family genome observation information may include information on incomplete trios. For example, family genome observation information may include a linear trio (genomic information of three siblings, including the subject) or an inverse trio (genomic information of one parent and two siblings, including the subject).

[0048] Family genome observation information may include information on whether identical or different base sequences are observed compared to a reference genome (or reference genome). In one example, the reference genome may be a genome representing a person's genetic information. Therefore, family genome observation information may include information on whether identical or different genetic information is observed compared to the genome representing the person's genetic information.

[0049] Family genome observation information may include observations of each individual haploid within a person compared to a reference genome. For example, family genome observation information may include information on how many of the individual's two haploids have the same sequence as the reference genome. In this case, the family genome observation information may include three values: i) both haploids are different from the reference genome, ii) only one of the two haploids is identical to the reference genome, and iii) both haploids are identical to the reference genome.

[0050] Family genome observation information can be analyzed separately for each haploid region of a person. For example, family genome observation information may include information on whether the genomes of two haploid anterior regions of a person are identical or different from the reference genome. Furthermore, family genome observation information may include information on whether the genomes of two haploid posterior regions of a person are identical or different from the reference genome.

[0051] Family genome observation information can be input into an analysis model. As described below, the analysis model can analyze genetic inheritance status based on the input values.

[0052] The analysis device can analyze the genetic inheritance state based on family genetic observation information (220).

[0053] Haplotype inheritance status can refer to how haplotypes are shared between siblings. Specifically, the genetic inheritance status can include information on which haplotypes are identical or non-identical among the haplotypes possessed by two siblings. In one embodiment, the genetic inheritance status can be one of the following: all haplotypes of two siblings are identical, only one is identical, or none are identical. In this case, only one is identical, meaning that one identical haplotype between two siblings is inherited from the father or from the mother.

[0054] Genomic inheritance states can be defined between two siblings. Therefore, for a linear trio, a total of three genetic inheritance states (subject-sibling 1, subject-sibling 2, and sibling 1-sibling 2) can be analyzed. For an inverse trio, a total of one inheritance state (subject-sibling 1) can be analyzed.

[0055] Genomic inheritance status can be analyzed separately for each haploid region. This is because, as mentioned above, familial genome observation data can be analyzed for each individual haploid region. Furthermore, this is because genetic recombination, such as crossing over, can occur during meiosis. For example, genome inheritance status may include information that the anterior region of a haploid is identical between two siblings, but the posterior region is not.

[0056] Specific information on family genome observations and genetic inheritance status will be provided later.

[0057] Genomic inheritance status can be analyzed using an analysis model. Specifically, the analysis device can analyze genetic inheritance status based on family genome observation data using the analysis model. The analysis model can be a model that receives family genome observation data as input and outputs genetic inheritance status as output.

[0058] In one example, the analysis model may be a machine learning (ML)-based model. A machine learning model may be a model that learns from training data to understand specific objects or conditions or identify patterns in data. Specifically, the analysis model may include decision trees, random forests (RF), k-nearest neighbors (KNN), naive Bayes, support vector machines (SVMs), and artificial neural networks (ANNs).

[0059] In one embodiment, the analysis model may be a Hidden Markov Model (HMM). A Hidden Markov Model (HMM) may include a model analyzing a linear trio or a model analyzing an inverse trio. For convenience, a HMM analyzing a linear trio may be referred to as a sibling HMM, and a HMM analyzing an inverse trio may be referred to as an inverse trio HMM.

[0060] The analysis device can determine haploid information of the target family based on the analyzed genetic inheritance status (230).

[0061] Haploid information can include information about the haploid composition of the family under analysis, based on the haploid sharing status between siblings identified by genetic inheritance status. Specifically, because genetic inheritance status is analyzed between two siblings, it does not reflect the haploid status of the entire family under analysis. Therefore, to determine the haploid status of the family under analysis, it is necessary to utilize all analyzed genetic inheritance statuses.

[0062] The analysis device can replace the missing genetic information of at least one parent with the genetic information of the sibling of the analysis subject based on the haploid information of the determined analysis subject family (240).

[0063] As mentioned above, in the case of an incomplete trio, at least one parent's genomic information is missing. Therefore, the missing parental genomic information can be replaced with the sibling genomic information from the haploid family of the subject. This allows for the detection of de novo mutations even when the genomic information of the subject's parents is unavailable.

[0064] At this time, the analysis device can search for an analyzable region among the genome of the analysis target, and replace the missing genome information of at least one parent with the genome information of the sibling of the analysis target only for the searched analyzable region.

[0065] An analyzable region can refer to a region where the haploid combination of the subject can be found in other family members. Specifically, an analyzable region can refer to a region where the haploid combination of the subject can be found in other members of an incomplete trio. This is because finding the haploid combination of the subject in other family members provides a comparison target for detecting de novo mutations.

[0066] For example, if the haploid for chromosome 12 of the analysis subject in a linear trio is AB, and sibling 1 has A and sibling 2 has B, the entire chromosome 12 of the analysis subject can be searched as an analyzable region. Alternatively, if the anterior region of the haploid for chromosome 12 of the analysis subject in a linear trio is AB, the posterior region is CD, and sibling 1 has A and sibling 2 has B, only the anterior region of chromosome 12 of the analysis subject can be searched as an analyzable region.

[0067] The analysis device can detect whether there is a de novo mutation based on the genetic information of the parents including the replaced genetic information and the genetic information of the analysis subject (250).

[0068] In one embodiment, if the device detects a mutation that exists in the genomic information of the subject of analysis but does not exist in the parental genomic information including the replaced genomic information, it can identify this as a de novo mutation.

[0069] Furthermore, the analysis device can output whether or not it has detected a de novo mutation.

[0070] Below we take a closer look at genetic inheritance status and family genetic observation information.

[0071] Figure 3 shows the possible genetic inheritance states that can be analyzed when two siblings inherit a non-recombinant haploid from their parents. In other words, Figure 3 shows the possible ways in which a haploid can be shared between two siblings.

[0072] As shown in Fig. 3, the genomic inheritance status can be analyzed between two siblings. There can be a total of four cases of genomic inheritance status between two siblings. Specifically, the genomic inheritance status can include a case where two siblings share two identical haploids (Identical (TWOID)), a case where they share one identical haploid and one different haploid (Haploidentical (HAPID)), and a case where they share two different haploids (Nonidentical (NONID)). In this case, HAPID can include a case where the shared identical haploid is inherited from the mother (Maternal identical (MATID)) and a case where the shared identical haploid is inherited from the father (Paternal identical (PATID)).

[0073] Figure 4 shows the number of cases of family genome observation information that can be observed from one sibling.

[0074] As shown in Figure 4, family genome observation information can be obtained by comparing each human haploid with reference genome information. The reference genome information may be a genome representing the human genetic information.

[0075] As shown in Fig. 4, if the genomes of both haploids of one sibling are identical to the reference genome, the family genome observation information may be {00}. Alternatively, if only one of the genomes of the two haploids of one sibling is identical to the reference genome, the family genome observation information may be {01}. Alternatively, if the genomes of both haploids of one sibling are different from the reference genome, the family genome observation information may be {11}.

[0076] Figure 5 shows the familial genome observation information that can be observed when two siblings inherit a non-recombinant haploid from their parents.

[0077] As shown in Fig. 5, the family genome observation information may include observation information of sibling 1 and sibling 2.

[0078] In Fig. 5, we consider the case where the haploid of sibling 1 and sibling 2 is AC / AC (row 1, column 1). In this case, the genome information of sibling 1 and sibling 2 is identical. If the genome information of sibling 1 and sibling 2 and the reference genome information are identical, {00, 00} can be observed. Conversely, if the genome information of sibling 1 and sibling 2 and the reference genome information are both different, {11, 11} can be observed. In this case, {11, 11} and {00, 00} are virtually the same, so {11, 11} can be converted to {00, 00}. If only one haploid among the genome information of sibling 1 and sibling 2 is identical to the reference genome information, {01, 01} can be observed. In this case, since the genomes of sibling 1 and sibling 2 are identical, {00, 01} or {00, 11} cannot be observed. In summary, if the haplotypes of siblings 1 and 2 are AC / AC, then the family genome observation information of {00, 00} or {01, 01} can be obtained.

[0079] In Figure 5, we consider the case where the haploids of siblings 1 and 2 are AD / AC (row 1, column 2). In this case, only one of the haploids of siblings 1 and 2 (A) is the same. At this time, siblings D of siblings 1 and C of siblings 2, inherited from the mother, may have different genotypes, but may also have the same genotype. In this case, if the genome information and reference genome information of haploid A shared by siblings 1 and 2 are identical, the genome information and reference genome information of haploid D of siblings 1 are identical, and the genome information and reference genome information of haploid C of siblings 2 are identical, then the family genome observation information of {00, 00} can be obtained. In addition, if the genome information and reference genome information of haploid A shared by siblings 1 and 2 are identical, the genome information and reference genome information of haploid D of siblings 1 are identical, and the genome information and reference genome information of haploid C of siblings 2 are different, then the family genome observation information of {00, 01} can be obtained. And if the genome information and reference genome information of haploid A shared by siblings 1 and 2 are identical, the genome information and reference genome information of haploid D of sibling 1 are different, and the genome information and reference genome information of haploid C of sibling 2 are different, then the family genome observation information of {01, 01} can be obtained. And if the genome information and reference genome information of haploid A shared by siblings 1 and 2 are different, the genome information and reference genome information of haploid D of sibling 1 are identical, and the genome information and reference genome information of haploid C of sibling 2 are identical, then the family genome observation information of {11, 01} can be obtained. In this case, since {11, 01} and {00, 01} are virtually identical, {11, 01} can be converted to {00, 01}. In summary, if the haploids of siblings 1 and 2 are AD / AC, then the family genome observation information of {00, 00} {00, 01} or {01, 01} can be obtained.

[0080] In Figure 5, we consider the case where the haplotypes of siblings 1 and 2 are BD / AC (row 1, column 3). In this case, there is no haplotype shared by siblings 1 and 2. Therefore, in this case, since the haplotypes of siblings 1 and 2 are independently compared with the reference genome information, all familial genome observation information can be obtained. In summary, if the haplotypes of siblings 1 and 2 are BD / AC, the genome observation information of {00, 00}, {00, 01}, {01, 01}, or {00, 11} can be obtained.

[0081] Below, we will look at a table that can be used to determine haploid information for the family under analysis based on the analyzed genetic inheritance status.

[0082] Figure 6 is a table showing all possible genetic inheritance states that can arise from a linear trio and the resulting family haplotype information that can be determined accordingly. Figure 6 shows all possible genetic inheritance states that can arise when the subject's haplotypes are fixed as A and D, and the parental haplotypes are fixed as A and B, and C and D.

[0083] As mentioned above, a total of three genetic inheritance states can be analyzed in a linear trio (subject-sibling 1, subject-sibling 2, and sibling 1-sibling 2). Notably, in linear trios, PATID and MATID cannot be distinguished due to the lack of parental genetic information. Therefore, if two siblings in a linear trio share only one haploid, this can be analyzed as a HAPID state.

[0084] Using the table in Figure 6, the haploid information of a family can be determined based on the analyzed genetic inheritance status in the linear trio. This is because the table in Figure 6 maps the genetic inheritance status and haploid information identified in the linear trio.

[0085] Figure 7 is a table showing all cases of genetic inheritance status that can arise from the inverted trio and the haploid information of the family that can be determined accordingly.

[0086] Figure 7 shows all possible genetic inheritance states that can arise when the subject's haploid is fixed as A and D, and the parent's haploid is fixed as A and B. As mentioned above, in the inverse trio, a single genetic inheritance state (subject-sibling 1) can be analyzed. Unlike the inverse trio, the linear trio can distinguish between PATID and MATID because it includes parental genetic information.

[0087] Using the table in Figure 7, the haploid information of a family can be determined based on the analyzed genetic inheritance status in the inverse trio. This is because the table in Figure 7 maps the genetic inheritance status and haploid information identified in the inverse trio.

[0088] Below we will look at an example of detecting de novo mutations.

[0089] Figure 8 shows one example of searching for de novo mutations in a linear trio.

[0090] As shown in Fig. 8, the family genome observation information of the linear trio can be obtained by comparing it with the reference genome information (1). At this time, the genome inheritance status can be analyzed based on the three family genome observation information (analysis subject-sibling 1, analysis subject-sibling 2, and sibling 1-sibling 2) (2)(3). If necessary, a hidden Markov model (HMM) can be used. Based on the analyzed genome inheritance status, the haplotype information of the analysis subject family can be determined (4). If necessary, the table described in Fig. 6 can be used. Based on the determined haplotype information, the genome information of the father / mother with missing genome information can be replaced with the genome information of sibling 1 and sibling 2 (5). By comparing the genome information of the parents, including the replaced genome information, with the genome information of the analysis subject, de novo mutations can be detected (6).

[0091] Figure 9 shows one example of searching for de novo mutations in the reverse trio.

[0092] As shown in Fig. 9, the family genome observation information of the reverse trio can be obtained by comparing it with the reference genome information (1). At this time, the genome inheritance status of one family genome observation information (analysis subject - sibling 1) can be analyzed (2)(3). If necessary, a hidden Markov model (HMM) can be used. Based on the analyzed genome inheritance status, the haploid information of the analysis subject family can be determined (4). If necessary, the table described in Fig. 7 can be used. Based on the determined haploid information, the maternal genome information with missing genome information can be replaced with part of the genome information of sibling 1 (5). By comparing the genome information of the parents that include this replaced genome information with the genome information of the analysis subject, de novo mutations can be detected (6).

[0093] Below, we will look at an example in which the haploid is divided into several regions and the genetic inheritance status of each region is analyzed.

[0094] Figure 10 shows an example of dividing a genome into several regions and analyzing the genome inheritance status for each region in the process of analyzing the genome inheritance status.

[0095] As shown in Figure 10, Child 1 can inherit both the father and mother's haplotypes (Haplotype A, Haplotype D). On the other hand, Child 2 can inherit a recombinant haplotype rather than the father and mother's haplotypes. Therefore, when analyzing the genetic inheritance status between Child 1 and Child 2, the genome can be divided into several regions, and the genetic inheritance status for each region can be analyzed.

[0096] Below, we will look at an example of detecting an analyzable region among the genomes of the analysis target.

[0097] Figure 11 shows one embodiment of setting an analyzable region in a linear trio and an inverse trio.

[0098] As shown in Figure 11, the upper region of the target genome in both linear and inverse trios can be found in other family members. Conversely, the lower region of the target genome in both linear and inverse trios cannot be found in part or all of the other family members.

[0099] Therefore, only the upper region of the genome of the analysis target is set as an analyzable region, and it is possible to determine whether there is a de novo mutation only in the analyzable region.

[0100] Below, we examine the results of constructing a hidden Markov model and evaluating the performance of the constructed Markov model.

[0101] Hidden Markov models analyze genetic inheritance based on familial genome observations. Hidden Markov models include the Sibling HMM, a model applicable to linear trios, and the Reverse Trio HMM, a model applicable to reverse trios.

[0102] Figure 12 shows the overall structure of the sibling HMM and the inverse trio HMM.

[0103] Sibling HMMs and inverse trio HMMs analyze genetic inheritance status based on observed familial genome observations. Inverse trio HMMs, since siblings are a pair, only one genetic inheritance status exists. Therefore, only the results of the inverse trio HMM are required for inverse trio analysis. However, to increase accuracy, sibling HMMs were also used for inverse trio analysis, and only cases where the results of both the sibling HMM and inverse trio HMM matched were considered for analysis.

[0104] The state values, observation values, starting probability, transition probability, and emission probability of the sibling HMM are as shown in Equations 1 to 3 and Tables 1 and 2.

[0105] Mathematical expression 1 represents the state value (genetic inheritance state) of the sibling HMM.

[0106] [Mathematical Formula 1]

[0107] States Q = {PATID, MATID, TWOID, NONID}

[0108] Mathematical expression 2 represents the observation value (family genome observation information) of the sibling HMM.

[0109] [Mathematical Formula 2]

[0110] Observations O = {{00, 00}, {00, 01}, {01, 01}, {00, 11}}

[0111] Mathematical expression 3 represents the starting probability of the sibling HMM.

[0112] [Mathematical Formula 3]

[0113] Start: {PATID: 0.25, MATID: 0.25, TWOID: 0.25, NONID: 0.25}

[0114] Table 1 shows the transition probabilities of sibling HMMs. In Table 2, r represents the recombination rate and is set to 1.0e-08.

[0115] PATIDMATIDTWOIDNONIDPATID1 - 2*r0rrMATID01 - 2*rrrTWOIDrr1 - 2*r0NONIDrr01 - 2*r

[0116] Table 2 shows the emission probabilities of the sibling HMM. In Table 2, e was used to account for Mendelian inheritance errors and was set to 0.9995.

[0117] (00, 00)(01, 01)(00, 01)(00, 11)PATID0.3*e0.3*e0.4*e1-eMATID0.3*e0.3*e0.4*e1-eTWOID0.5*e0.5*e(1-e) / 2(1-e) / 2NONID0.2*e0.3*e0.4*e0.1*e

[0118] The state values, observation values, starting probability, transition probability, and emission probability of the inverse trio HMM are as shown in Equations 4 to 5 and Tables 3 and 4.

[0119] Mathematical expression 4 represents the state value (genetic inheritance state) of the inverse trio HMM.

[0120] [Equation 4]

[0121] States Q = {PATID, MATID, TWOID, NONID}

[0122] Equation 5 represents the observations (family genome observation information) of the inverse trio HMM. The observations of the inverse trio consist of pairs of parental genome types and sibling genome types.

[0123] [Equation 5]

[0124] Observation O = {{'01', {'00','00'}}, {'01', {'00','01'}}}, {'01', {'00','11'}}}, {'01', {'01','01'}}, {'00', {'00','00','00'}}, {'00', {'00','01'}}}, {'00', {'00','11'}}}, {'00', {'01','01'}}, {'11', {'00','00'}}, {'11', {'00','01'}}}

[0125] Mathematical expression 6 represents the starting probability of the inverse trio HMM.

[0126] [Equation 6]

[0127] Start: {PATID: 0.25,MATID: 0.25,TWOID: 0.25,NONID: 0.25}

[0128] Table 3 shows the transition probabilities of the inverse trio HMM. In Table 3, r represents the recombination rate and is set to 1.0e-08.

[0129] PATIDMATIDTWOIDNONIDPATID1 - 2*r0rrMATID01 - 2*rrrTWOIDrr1 - 2*r0NONIDrr01 - 2*r

[0130] Table 4 shows the emission probabilities of the inverse trio HMM. In Table 4, e was used to account for Mendelian inheritance errors and was set to 0.9995.

[0131] (01,{00, 00})(01,{00, 01})(01,{00, 11})(01,{01, 01})(00,{00, 00})(00,{00, 01})(00,{00, 11})(00,{01, 01})(11,{00, 00})(11,{00, 01})PATID0.1*e0.2*e(1-e) / 40.1*e0.2*e0.2*e(1-e) / 40.2*e(1-e) / 4(1-e) / 4MATID(1-e) / 70.4*e(1-e) / 7(1-e) / 70.3*e(1-e) / 7(1-e) / 70.3*e(1-e) / 7(1-e ) / 7TWOID0.2*e(1-e) / 6(1-e) / 60.2*e0.3*e(1-e) / 6(1-e) / 60.3*e(1-e) / 6(1-e) / 6NONID(1-e) / 40.2*e0.1*e0.1*e0.2*e0.2*e(1-e) / 40.2*e(1-e) / 4(1-e) / 4

[0132] The analysis model was input with only one selected set of multiple genomic observations.

[0133] Figure 13 illustrates the priorities among family genome observation information. While family genome observation information can utilize single-measured values, to reduce noise in the analysis model's output, multiple genome observation information can be acquired and only one of them can be selected for input into the analysis model. As illustrated in Figure 13, priority can be set to select a single genome observation information.

[0134] As shown in Fig. 13(i), in the case of sibling HMMs, {00, 11} has the highest priority. Therefore, if {00, 11} exists in multiple genome observation information, {00, 11} is selected and input into the sibling HMM model. Similarly, if {00, 11} does not exist in multiple genome observation information and {00, 01}, which has the second highest priority, {00, 01} can be selected and input into the sibling HMM model.

[0135] In the case of the inverse trio HMM as in Fig. 13(ii), {01, {00, 00}} has a high priority. Therefore, if {01, {00, 11}} is not present in the multiple genome observation information but {01, {00, 00}} is present, {01, {00, 00}} can be selected and input into the inverse trio HMM.

[0136] Figure 14 shows the results of an accuracy analysis of the sibling HMM and the inverted trio HMM. Long-read sequencing data from a family consisting of two parents and two children, which allowed for high-resolution identification of the locations of chromosomal recombination in the children, was used as the gold standard. Comparing the correct data with the sibling HMM and inverted trio HMM, we confirmed a high accuracy of approximately 98%.

[0137] Figure 15 shows the results of measuring whether the sibling HMM and trio HMM can detect de novo mutations well in artificially created incomplete trios targeting 33 CEPH families with many children.

[0138] Specifically, the sibling HMM achieved 76.36% precision and 89.23% recall while covering 52.28% of the genome area in the linear trio. The inverse trio HMM achieved 70.02% precision and 91.07% recall while covering 43.87% of the genome area in the inverse trio.

[0139] Figure 16 shows the estimated number of samples needed to detect a statistically significant difference in de novo mutation rate compared to the control group, based on the accuracy of the current method. In a study design comparing the number of de novo mutations in a control group exposed to a mutation-inducing influence and a control group not exposed, a 15% difference in de novo mutation rate between the two groups would require 39 linear trios or 46 inverse trios.

[0140] Figure 17 shows that as the number of siblings increases under the Linear Trio and Reverse Trio conditions, the number of de novo mutations that can be detected by the de novo mutation detection method increases.

[0141] As shown in Figure 17, applying the de novo mutation detection method increases the likelihood of finding de novo mutations as the number of siblings increases. As shown in Figure 17, when the number of siblings is four or more, applying the de novo mutation detection method can detect most de novo mutations across 80% of the genome.

[0142] Figure 18 shows the results of applying the de novo mutation detection method to data on patients and families with undiagnosed rare diseases.

[0143] In Figure 18, Complete refers to a case where all parent information is available. Incomplete refers to an incomplete trio where at least one parent information is missing.

[0144] As shown in Figure 18, the de novo mutation detection method can detect disease-associated mutations in approximately 30% of patients in the Complete group. As shown in Figure 18, the de novo mutation detection method can detect disease-associated mutations in approximately 20% of patients in the Incomplete group.

[0145] As can be seen in Figure 18, the de novo mutation detection method shows that an additional diagnosis of about 20% is possible even when analysis is impossible (incomplete) with conventional methods due to missing at least one of the parental information.

[0146] Figure 19 shows the results of applying the de novo mutation detection method in a case where there is no information on both parents and there are five children.

[0147] In Figure 19, three of the five children have the disease. As shown in Figure 19, when genetic information from the patient and four siblings was used to identify de novo mutations, two candidate mutations associated with the disease were identified.

[0148] The following describes the analysis device.

[0149] Fig. 20 is a configuration of one embodiment of an analysis device (300).

[0150] The analysis device (300) may correspond to the analysis device (100) described in Fig. 1. That is, the analysis device (300) may be a device that performs the de novo mutation detection method described above.

[0151] The analysis device (300) may include at least one input device (310), storage device (320), operation device (330), output device (340), interface device (350), and communication device (360).

[0152] The input device (310) can receive data, information, or models necessary for performing the aforementioned de novo mutation detection method. The input device (310) can receive family genome observation information. The input device (310) can receive an analysis model.

[0153] The input device (310) may include a device for inputting a certain command or data (such as a keyboard, mouse, touch screen, joystick, trackball, touchpad, scanner, webcam, etc.). The input device (310) may also include a configuration for receiving data through a separate storage device (such as a USB, CD, hard disk, etc.). The input device (310) may also receive data through a separate measuring device or a separate database. The input device (310) may also receive data through a communication device (360) via a wired or wireless connection. The input device (310) may also receive a control signal for controlling the analysis device (300).

[0154] The storage device (320) can store data, information, models, etc. required to perform the aforementioned de novo mutation detection method. The storage device (320) can store family genome observation information. The storage device (320) can store an analysis model. The storage device (320) can also be a device that stores certain data, information, models, etc. The storage device (320) can store data, information, models, etc. input through the input device (310). The storage device (320) can store commands that cause the calculation device (330) to perform operations required for the de novo mutation detection method. The storage device (320) can store information generated during the calculation process of the calculation device (330). That is, the storage device (320) can include a memory. For example, the storage device can include a hard disk drive (HDD), a solid state drive (SSD), a ROM, a RAM, a CD-ROM, a magnetic tape, or a floppy disk.

[0155] The computing device (330) can perform the calculations necessary to perform the aforementioned de novo mutation detection method. The computing device (330) can analyze the inheritance state of the genome based on the family genome observation information. The computing device (330) can determine the haploid information of the family to be analyzed based on the analyzed genome inheritance state. The computing device (330) can replace the sibling genome information of the analysis subject with the genome information of at least one or more parents based on the determined haploid information of the analysis subject. The computing device (330) can search for the presence of a de novo mutation based on the genome information of the parents including the replaced genome information and the genome information of the analysis subject.

[0156] The computing device (330) may be a device such as a processor, an application processor (AP), or a chip embedded with a program that processes data and performs certain operations. For example, the computing device (330) may include a central processing unit (CPU), a graphics processing unit (GPU), or a neural processing unit (NPU). The computing device (330) may generate a control signal that controls the analysis device (300). The computing device (330) may generate a control signal that controls the input device (310), the storage device (320), the output device (340), the interface device (350), and the communication device (360) included in the analysis device (300).

[0157] The output device (340) may be a device that outputs certain data, information, and models. The output device (340) may be a device that outputs certain data, information, and models to the outside of the analysis device (300). The output device (340) may output interfaces, input data, analysis results, etc. required for the data processing process. The output device (340) may include a device that outputs data, etc. through tactile, visual, auditory, gustatory, and olfactory methods. The output device (340) may output the results of a de novo mutation search. The output device (340) may be physically implemented in various forms, such as a display, a speaker, a vibration motor, or a document output device. The output device (340) may output data, information, or models stored in the storage device (320). The output device (340) may output data, information, and models generated during the operation of the operation device (330). The output device (340) can output the result of the calculation performed by the calculation device (330).

[0158] The interface device (350) may be a device that receives certain commands and data from the outside. The interface device (350) may receive a control signal for controlling the analysis device (300). The interface device (350) may output the results analyzed by the analysis device (300). The interface device (350) may receive information necessary for performing the aforementioned de novo mutation detection method from a physically connected input device or an external storage device.

[0159] The communication device (360) can receive information necessary for performing the aforementioned de novo mutation detection method. The communication device (360) can receive a model necessary for performing the aforementioned de novo mutation detection method. The communication device (360) can transmit and receive family genome observation information. The communication device (360) can transmit and receive an analysis model. The communication device (360) can receive a control signal necessary for controlling the analysis device (300). The communication device (360) can transmit the results analyzed by the analysis device (300). The communication device (360) can refer to a configuration that receives and transmits certain data, information, models, etc. through a wired or wireless network. The communication device (360) can perform network communication such as Wi-Fi (Wireless Fidelity), Wi-Fi Direct, Bluetooth, UWB (Ultra-Wide Band), NFC (Near Field Communication), USB (Universal Serial Bus), HDMI (High Definition Multimedia Interface), LAN (Local Area Network), etc.

[0160] The above-described de novo mutation detection method can be implemented as a program (or application) including an executable algorithm that can be run on a computer.

[0161] The above program may be provided stored on a non-transitory computer readable medium.

[0162] The above-mentioned temporarily readable medium refers to various RAMs such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous DRAM (Synclink DRAM, SLDRAM), and direct Rambus RAM (DRRAM).

[0163] The above non-transitory readable medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, the various applications or programs described above may be stored and provided in a non-transitory readable medium, such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM (read-only memory), PROM (programmable read only memory), EPROM (Erasable PROM, EPROM), EEPROM (Electrically EPROM), or flash memory.

[0164] The present embodiment and the drawings attached to the present specification only clearly illustrate a part of the technical idea included in the above-described technology, and it will be obvious that all modified examples and specific embodiments that can be easily inferred by a person skilled in the art within the scope of the technical idea included in the specification and drawings of the above-described technology are included in the scope of the rights of the above-described technology.

Claims

1. A step in which an analysis device acquires family genome observation information of an analysis subject whose genetic information of at least one parent is missing; A step in which the above analysis device analyzes the genetic inheritance state based on family genetic observation information; A step in which the above analysis device determines haploid information of the family to be analyzed based on the analyzed genetic inheritance status; A step in which the analysis device replaces the missing at least one parental genome information with sibling genome information of the analysis subject based on the determined haploid information of the analysis subject family; and A method for detecting a de novo mutation, comprising: a step of the analysis device searching for a de novo mutation based on the genetic information of the parent including the replaced genetic information and the genetic information of the analysis subject.

2. In paragraph 1, The above family genome observation information is a linear trio or a reverse trio, The above linear trio includes the genetic information of at least three siblings including the subject of the analysis, A de novo mutation detection method, wherein the above reverse trio includes the genetic information of one parent and the genetic information of at least two siblings including the subject of the analysis.

3. In paragraph 1, A de novo mutation detection method, wherein the analysis of the above-mentioned genetic inheritance status includes analyzing each region of the haploid separately.

4. In paragraph 1, Analyzing the above genetic inheritance status includes analyzing using an analysis model, The above analysis model is a de novo mutation detection method that receives the above family genome observation information as input and infers the genome inheritance status.

5. In paragraph 4, The above analysis model is a de novo mutation detection method based on machine learning (ML).

6. In paragraph 4, The above analysis model is a de novo mutation detection method based on Hidden Markov Models (HMM).

7. In paragraph 1, The above analysis device further includes a step of searching for an analyzable region among the genomes of the analysis target; Replacing the above parental genome information means replacing the missing genome information of at least one parent with the sibling genome information of the analysis subject only for the searched analyzable region. A de novo mutation detection method, wherein the analyzable region includes a region in which the haploid combination possessed by the analyzable region can be found in other family members.

8. An input device for obtaining family genome observation information of a subject of analysis in which at least one parent's genome information is missing; A computing device that analyzes the inheritance state of a genome based on the above family genome observation information, determines haploid information of a family to be analyzed based on the analyzed genome inheritance state, replaces the missing genome information of at least one parent with sibling genome information of the subject of analysis based on the determined haploid information of the subject of analysis, and searches for de novo mutations based on the genome information of the parents including the replaced genome information and the genome information of the subject of analysis; and A method for detecting de novo mutations, comprising an output device for outputting whether or not the de novo mutation is present.

Images