Method for non-invasively evaluating zygotic property and fetal fraction of double-fetal pregnant fetuses

Abstract

The invention provides a method for non-invasively evaluating the zygotic property and the fetal fraction of a double-fetal pregnant fetus. Specifically, the method provided by the invention constructs a method for noninvasive evaluation of fetal zygote by detecting and calculating relative abundance of fetal specific allele and fetal specific allele fraction of cfDNA in a blood sample. By adopting the method disclosed by the invention, the zygotic property of the fetus can be efficiently and accurately determined.

Description

Technical Field

[0001] The present invention relates to the field of medical diagnosis, and in particular, to a method for non-invasively evaluating fetal zygosity and fetal fraction in twin pregnancies. Background Art

[0002] Approximately 3% of pregnant women have a multifetal pregnancy (MP), which can lead to significant risks to maternal and fetal health, including gestational diabetes, intrauterine growth restriction, preeclampsia, and preterm birth. The overall MP rate has been increasing with the use of ovulation drugs in assisted reproductive technology. Twin pregnancies account for 97%-98% of all MPs, which are classified as monozygotic (MZ) and dizygotic (DZ) twins based on the fetal zygote. MZ twins are produced by the splitting of a single zygote into two embryos with essentially identical embryonic structures, while DZ twins are produced by the fertilization of two pairs of eggs and sperm (Bai, et al., 2020). Twin pregnancies can usually be detected early in pregnancy by ultrasound and followed by special prenatal care.

[0003] Chorionicity is the most important risk factor for adverse pregnancy outcomes in twin pregnancies (ACOG, 2016). Perinatal morbidity and mortality are approximately four times greater in monochorionic twin pregnancies than in dichorionic twin pregnancies. Because DZ twins are usually dizygotic, establishing fetal zygosity of the twins has prognostic value for appropriate prenatal care. Chorionicity can be identified early in pregnancy by ultrasonography, but its accuracy is affected by gestational age at the time of evaluation, clinician experience, and the ultrasonographic variables used. Another limitation of ultrasonographic twin pregnancies is the identification of those affected by vanishing twin syndrome, in which one twin ceases to develop in utero, resulting in spontaneous reduction to a singleton pregnancy. Vanishing twin syndrome occurs in 36% of pregnancies with two gestational sacs and is associated with an increased risk of preterm birth, low birth weight, and perinatal mortality.

[0004] With the advancement of next-generation sequencing (NGS) technology, NIPS involving cell-free fetal DNA (cfDNA) sequence analysis has been shown to be an accurate and effective method to identify fetuses at increased risk of aneuploidy. However, the presence of low FF twins or vanishing twins reduces the accuracy of NIPS. FF can be inferred by detecting SNPs using low-depth whole-genome sequencing (WGS). NIPS (non-invasive prenatal screening) using targeted NGS can also determine dizygosity and fetal fraction (FF) by analyzing more than 12,000 SNPs.

[0005] Because fetuses with aneuploidy in DZ twins often have lower fetal fractions (FF) than fetuses in singleton pregnancies, erroneous results may be generated when NIPS cannot identify samples from twin pregnancies. Similarly, twin death and twin reduction can interfere with NIPS determination because cfDNA from the undeveloped fetus remains in the maternal circulation and accounts for a significant portion of the total cfDNA. However, the current understanding of changes in FF after twin reduction is limited, which has hindered the improvement of NIPS in such pregnancies.

[0006] Therefore, there is an urgent need in the art to develop a new method for more accurately and efficiently detecting the zygosity and fetal fraction of twin pregnancies. Summary of the invention

[0007] The present invention provides a new method for more accurately and efficiently detecting the zygosity and fetal fraction of twin pregnancy.

[0008] In a first aspect of the present invention, a detection device for determining fetal zygosity is provided, the device comprising:

[0009] (a) a data input module, the data input module is used to input nucleic acid data of a sample to be analyzed, the nucleic acid data including sequence data based on fetal free deoxyribonucleic acid in the peripheral blood of a pregnant woman;

[0010] (b) an analysis and calculation module, the analysis and calculation module being configured to analyze the nucleic acid data of the sample to be analyzed, thereby determining (i) the relative heterozygous abundance of fetal-specific alleles and (ii) the fetal-specific allele fraction or its relative fluctuation;

[0011] (c) an evaluation module, the evaluation module being configured to evaluate the relative heterozygous abundance of the fetal-specific alleles and the fetal-specific allele fraction or the relative fluctuation thereof, thereby obtaining a detection result of fetal zygosity; and

[0012] (d) an output module, which is used to output the detection result of the fetal zygosity.

[0013] In another preferred example, the SNP frequency of a target population of a specific single nucleotide polymorphism (SNP) is denoted as p, the total number of queried SNPs is denoted as N, the number of SNPs in which the mother is homozygous is N0, the number of heterozygous SNPs in singletons or monozygotic (MZ) twins among the SNPs in which the mother is homozygous is N1, and the number of heterozygous SNPs in dizygotic (DZ) twins is N2;

[0014] The number of informative SNPs in DZ or MZ twin pairs was calculated by the following formula, where lowercase letters are observed values, uppercase letters are derived values, and italic uppercase letters are used only for symbols:

[0015] N0=[(p 2 +(1-p) 2 )]*N

[0016] N1=[p 2 (1-p)+p(1-p) 2 ]*N=p(1-p)*N

[0017]

[0018] In another preferred embodiment, in the analysis and calculation module, the B allele fraction (BAF) value is calculated.

[0019] In another preferred embodiment, the calculation of the B allele fraction (BAF) value includes:

[0020] For maternally homozygous SNPs, they were divided into two groups according to the maternal genotype (BB or AA), where the median BAFs of these SNPs were M1 and M2;

[0021] Calculate the standard deviation of BAF for each group, i.e., σ1 and σ2;

[0022] Take the weighted average σ of σ1 and σ2 ave , which is the weighted number of SNPs included in the statistics for each group, and the median of the heterozygous SNP depth is expressed as M dep ;

[0023] The corrected BAF change σ is calculated as follows: cor

[0024]

[0025] In another preferred embodiment, the analysis and calculation module further includes a fetal score calculation submodule, and the fetal score calculation submodule is configured to analyze the nucleic acid data of the sample to be analyzed, so as to determine the fetal score;

[0026] And the assessment module also includes a twin pregnancy risk assessment submodule, and the twin pregnancy risk assessment submodule is configured to evaluate the fetal score to obtain a twin pregnancy risk value.

[0027] In another preferred embodiment, the calculation of the fetal fraction includes: calculating the FF of each twin based on equations 1, 2, 18 and 19:

[0028]

[0029]

[0030] In the formula, the total haploid genome copies of fetus 1, fetus 2, and mother are R1, R2, and M present in the sample, respectively;

[0031]

[0032]

[0033] Among them, bafm1, bafm2, and bafm3 are calculated according to formulas 13-15 respectively;

[0034]

[0035]

[0036]

[0037] Where BAF is the B allele fraction; R1 is the haploid genome copy number from fetus 1; R2 is the haploid genome copy number from fetus 2; and M is the haploid genome copy number from the mother.

[0038] In another preferred embodiment, the calculation of the fetal fraction further includes obtaining R1 and R2 values ​​using equations 13, 14 and 15:

[0039]

[0040]

[0041] wherein bafm1, bafm2, and bafm3, R1, R2, and M are as defined above.

[0042] In another preferred embodiment, for the calculation of the fetal fractions of fraternal twins of different sexes, the fetal fractions of female fetus 1 and male fetus 2 are set to FF1 and FF2, respectively, and the median reading depths of the SNPs of the autosomes and Y chromosomes are RD a and RD y , the median paternal genetic read depth of informative SNPs on chromosome X in fetus 1 (female twin) was RD xf The maternal and fetal genome copies present in twins of different sexes are denoted as K a , K x and K y ;

[0043] And K x / K a =C1,K y / K a =C2;

[0044] The fetal fraction can be calculated by the following formula:

[0045]

[0046]

[0047] In another preferred example, the detection device further includes a control module, and the control module is configured to control the operation of each module.

[0048] In another preferred example, the sample to be analyzed is a blood sample.

[0049] In another preferred embodiment, the sequencing data is sequencing data of free DNA.

[0050] In another preferred example, the sequencing data includes NGS sequencing data.

[0051] In another preferred embodiment, the data input module is selected from the following group: a sequencer, a scanner, a keyboard, a tablet computer (PAD), a smart phone, or a combination thereof.

[0052] In another preferred embodiment, the detection device includes a processor and a storage device.

[0053] In another preferred example, the output module includes a display, a printer, a tablet computer (PAD), and a smart phone.

[0054] In another preferred embodiment, the modules are connected via wired or wireless means.

[0055] In another preferred embodiment, the zygosity includes identical and fraternal zygosity.

[0056] In another preferred embodiment, the fraternal egg includes: normal pregnancy and abnormal pregnancy.

[0057] In another preferred embodiment, the sequencing data is sequencing data of free DNA.

[0058] In another preferred example, the sequencing data includes NGS sequencing data.

[0059] In another preferred embodiment, the zygosity includes identical and fraternal zygosity.

[0060] In a second aspect of the present invention, a method for data analysis of fetal zygosity is provided, the method comprising the following steps:

[0061] (a) providing data, wherein the data includes nucleic acid data of a sample to be analyzed, wherein the nucleic acid data includes sequence data based on fetal free deoxyribonucleic acid in the peripheral blood of a pregnant woman;

[0062] (b) analyzing and calculating the nucleic acid data of the sample to be analyzed to determine (i) the relative heterozygous abundance of fetal-specific alleles and (ii) the fetal-specific allele fraction or its relative fluctuation; and

[0063] (c) evaluating the results, evaluating the relative heterozygous abundance of the fetal-specific alleles and the fetal-specific allele fraction or their relative fluctuations, thereby obtaining a detection result of the fetal zygosity.

[0064] In another preferred embodiment, the method is a non-diagnostic and non-therapeutic method.

[0065] In another preferred embodiment, the blood sample is a blood sample from a pregnant woman.

[0066] It should be understood that within the scope of the present invention, the above-mentioned technical features of the present invention and the technical features specifically described below (such as embodiments) can be combined with each other to form a new or preferred technical solution. Due to space limitations, they will not be described one by one here. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 Figure 2 shows the B allele fraction patterns of twins with different karyotypes; different B allele fraction patterns composed of different combinations of autosomal and X chromosome SNPs can be used to infer the karyotype of each fetus in different twin pregnancies; (A) 46,XX and 46,XX twins, each pair of twins inherited a different maternal X chromosome; (B) 46,XY and 46,XY twins, each pair of twins inherited a different maternal X chromosome; (C) 46,XX and 46,XX twins, each pair of twins inherited the same maternal X chromosome; (D) 46,XY (E) 46,XX and 46,XX twins, each pair of twins inherited the same maternal X chromosome; (F) 46,XX and 46,XY twins; (G) and (H) 46,XX and 46,XX identical twins; (I) 46,XY and 47,XY,+21 twins; Chromosome-specific SNP loci are marked with different colors, ChrRef: reference chromosome, Chr: chromosome, SNP: single nucleotide polymorphism.

[0068] Figure 2Figure 2 shows binary analysis of twin zygosity using the number of fetal-specific SNPs and the B allele fraction: (A) validation of singleton and twin pregnancies tested in 215 singletons, 90 DZ twins and 2 egg donor samples; X-axis: relative heterozygous abundance of informative fetal-specific SNPs, calculated as the ratio between the number of informative SNP loci in fetuses where the mother is homozygous and the number of fetal heterozygous variants for these SNPs; Y-axis: relative coefficient of variation of the B allele fraction of informative fetal-specific SNPs; (B) egg donor pregnancies; (C) samples with 5% contamination; (D) twins with hydatidiform mole; chromosome-specific SNP loci are marked with different colors, SNP: single nucleotide polymorphism, DZ: dizygous, ChrRef: reference chromosome, Chr: chromosome.

[0069] Figure 3 A comparison of fetal fraction calculation using the autosomal and sex chromosome methods is shown: (A) FF was calculated for 338 singleton male pregnancies based on the autosomal method and the sex chromosome method; (B) and (C) FF was calculated for each fetus in 9 dizygotic twin pregnancies based on the autosomal method and the sex chromosome method; FF: fetal fraction; FF1: fetal fraction of fetus 1; FF2: fetal fraction of fetus 2; CCC: concordance coefficient.

[0070] Figure 4 Figure 2 shows the dynamics of fetal fraction after twin reduction; Fetal fraction (FF) levels were measured in nine twin pregnancies before and after the reduction of one of the twins: (A) Change in FF in the non-developing twin; (B) Change in FF in the surviving twin; Nine pregnant women had one blood draw to quantify FF before twin reduction, and each had one or two additional blood draws to monitor changes in FF after twin reduction. FF was calculated by the sex chromosome method. GA: gestational age.

[0071] Figure 5 The pattern of B allele fractions on chromosomes for singleton and twin pregnancies is shown: (A) singleton pregnancy; (B) identical twin pregnancy; (C) fraternal twins inherit the same paternal alleles and the same maternal alleles; (D) fraternal twins inherit the same paternal alleles and different maternal alleles; (E) fraternal twins inherit different paternal alleles and the same maternal alleles; (F) fraternal twins inherit different paternal and maternal alleles; X-axis: different loci on chromosomes, Y-axis: B allele fractions.

[0072] Figure 6Detection of complicated pregnancies using the novel SNP-based NIPS is shown: (A) egg donor pregnancy; (B) singleton pregnancy sample (46, XY) with 2% contamination from another sample (46, XX); (C) singleton pregnancy sample (46, XY) with 1% contamination from another sample (46, XY); chromosome-specific SNP loci are marked with different colors; SNP: single nucleotide polymorphism, NIPS: non-invasive prenatal screening, ChrRef: reference chromosome, Chr: chromosome. DETAILED DESCRIPTION

[0073] After extensive and in-depth research, the inventors unexpectedly developed a method for detecting fetal zygosity with high accuracy and sensitivity for the first time. Based on the inventors' research, using only one of the indicators (i) relative heterozygous abundance of fetal-specific alleles and (ii) relative fluctuation of fetal-specific allele fractions cannot distinguish cfDNA sequencing data from blood samples of mothers with DZ and MZ; however, when (i) relative heterozygous abundance of fetal-specific alleles and (ii) relative fluctuation of fetal-specific allele fractions are combined for analysis, DZ and MZ can be distinguished very accurately and efficiently and accurate fetal zygosity can be obtained. The present invention was completed on this basis.

[0074] the term

[0075] As used herein, the term "read depth" refers to the target number of sequencing reads obtained per sample.

[0076] As used herein, the terms "fetal-specific allele relative heterozygous abundance" and "fetal-specific allele relative abundance" are used interchangeably.

[0077] As used herein, the terms "fetal-specific allele fraction (BAF)", "B allele fraction" are used interchangeably and refer to the overall BAF of a fetal-specific SNP.

[0078] The detection device of the present invention

[0079] The present invention provides a device for detecting fetal zygosity and / or fetal fraction based on fetal free deoxyribonucleic acid in the peripheral blood of pregnant women.

[0080] Typically, the detection device of the present invention is as described in the first aspect of the present invention.

[0081] In the present invention, nucleic acid data can be input manually or automatically. Typically, the nucleic acid data input module is selected from the group consisting of a sequencer, a scanner, a keyboard, a tablet computer (PAD), a smart phone, or a combination thereof.

[0082] Preferably, in the present invention, the detection device comprises a processor and a storage device, wherein the storage device stores threshold data for determining fetal zygosity. In addition, the storage device may also store calculation data for analytical calculation processing.

[0083] In the present invention, representative output modules include (but are not limited to): a display, a printer, a tablet computer (PAD), and a smart phone.

[0084] The detection device of the present invention can be in the form of an integrated device or a distributed device. For example, the nucleic acid data input module can be independent, and the collected or input nucleic acid data can be transmitted to a local processing module by wire or wireless means, or uploaded to a non-local processing module (for example, a server in a remote center) by wireless or communication means, thereby realizing remote screening or data analysis.

[0085] In one embodiment, after the remote processing module has identified the nucleic acid data, the auxiliary screening results (i.e., the test results of fetal zygosity and / or fetal score) can be wirelessly transmitted back to an output device connected to the network, such as a tablet computer (PAD) or a smart phone, thereby achieving rapid NIPS screening.

[0086] Method of the present invention

[0087] In the present invention, the fetal free DNA (cfDNA) in maternal plasma during pregnancy can be detected by known methods (including sequencing, PCR, etc.), thereby obtaining the genetic information of the fetus in the fetal free DNA.

[0088] Typically, in the present invention, two independent parameters, the relative heterozygous abundance of fetal alleles and the fetal-specific allele fraction (or its relative fluctuation (variation)), can be used as the X-axis and Y-axis in a two-dimensional coordinate system, and many points of sequencing data based on cfDNA from blood samples of MZ and DZ can be plotted in the coordinate system, and two clearly separated discrete sample clusters can be obtained, namely, the DZ sample cluster (red) located in the upper right and the MZ sample cluster located in the lower left. A dividing line can be drawn between the DZ sample cluster (red) and the MZ sample cluster, and the samples above the dividing line belong to the DZ samples, and the samples below the dividing line belong to the MZ samples, thereby accurately distinguishing identical and fraternal eggs.

[0089] The method of the present invention is further described below in conjunction with a preferred embodiment of the present invention. Figure 1 and Figure 2 A, which shows a schematic flow chart of a preferred embodiment of the present invention.

[0090] In a preferred embodiment, fetal zygosity can be measured or determined based on a two-dimensional coordinate system. Typically, two independent parameters, the relative abundance of fetal alleles and the fetal-specific allele fraction, can be used as the X-axis and the Y-axis, respectively; in this way, each sample point is based on the corresponding sample point P (x, y) in the two-dimensional coordinate system, wherein x is the relative abundance (or quantity) of the fetal alleles of the sample, and y is the relative fluctuation (or variation) of the fetal-specific allele fraction or BAF of the sample. Based on the research of the present invention, the cfDNA sequencing data from the blood samples of mothers pregnant with DZ and MZ cannot be distinguished by using any one of the indicators (i) relative heterozygous abundance of fetal-specific alleles and (ii) fetal-specific allele fraction alone; however, the combined analysis of (i) relative heterozygous abundance of fetal-specific alleles and (ii) fetal-specific allele fraction can distinguish DZ and MZ very accurately and efficiently, thereby obtaining accurate fetal zygosity. The inventors unexpectedly found that by plotting the numerous points of the sequencing data of cfDNA from blood samples of MZ and DZ in the coordinate system, two distinctly separated discrete sample clusters can be obtained. For any sample to be tested, the fetal zygosity corresponding to the sample to be tested (or sample data) can be determined based on the discrete sample clusters described therein.

[0091] In another preferred embodiment, the method of the present invention may include the following steps:

[0092] (S1) providing nucleic acid data of a sample to be analyzed;

[0093] (S2) analyzing the nucleic acid data of the sample to be analyzed to determine the relative abundance of fetal-specific alleles (or the number of SNPs) and the fetal-specific allele fraction (or BAF);

[0094] (S3) establishing a two-dimensional coordinate system for the relative abundance of fetal-specific alleles and the fetal-specific allele fraction, with the relative abundance of fetal-specific alleles as the X-axis and the fetal-specific allele fraction as the Y-axis, and depicting the position of the sample data to be tested in the coordinate system (i.e., the sample point);

[0095] (S4) judging the position of the sample point and the threshold line according to a pre-set fitting threshold line of fetal zygosity; if the sample point is below the threshold line, the fetus is identical; if the sample point is above the threshold line, the fetus is fraternal.

[0096] The main advantages of the present invention are:

[0097] (a) Compared with traditional biochemical methods and ultrasound screening, NIPS is more accurate in detecting fetal aneuploidy by analyzing cfDNA in fetal plasma.

[0098] (b) The method of the present invention has high accuracy and versatility in detecting twin zygosity even in pregnancies with different analytical confounding factors.

[0099] (c) The method of the present invention can effectively determine zygosity, which helps to determine chorionicity and thus facilitates the prenatal management of high-risk twin pregnancies.

[0100] (d) The method for calculating the fetal fraction of the present invention can be used to monitor changes in FF after twin reduction to manage high-risk twin pregnancies.

[0101] The present invention will be further described below in conjunction with specific examples. It should be understood that these examples are intended to illustrate the present invention only and are not intended to limit the scope of the present invention. The experimental methods in the following examples where specific conditions are not specified are usually performed under conventional conditions, such as those described in Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989), or under conditions recommended by the manufacturer. Unless otherwise indicated, percentages and parts are weight percentages and weight parts.

[0102] Materials and methods

[0103] A total of 499 peripheral blood samples from 407 pregnant women with a gestational age of 12 weeks or more were included in the present invention. Single or twin pregnancies were confirmed by prenatal ultrasound. The present study was approved by the Institutional Review Board of China International Peace Maternity and Child Health Hospital (GKLW2019-72 and GKLW2019-52). Informed consent was obtained from all participants.

[0104] Probe design, cell-free DNA extraction, and sequencing

[0105] To detect disease-associated chromosomal copy number variations, >2,600 probes were designed for loci associated with frequent chromosomal abnormalities. In addition, >200 probes were designed to target SNPs without common (≥1% population frequency) copy number variation, reported in the Database of Genomic Variants (http: / / dgv.tcag.ca / dgv / app / home, last accessed on August 5, 2019). The above SNPs have a minor allele frequency of 30-70% in East Asia, Africa, and non-Finnish Europe to ensure sufficient polymorphism in different populations. To reduce probe hybridization bias between the reference allele and the alternative allele, one of the four possible nucleotides (A, C, G, and T) was selected at the locus corresponding to the target SNP to have a minimum melting temperature difference for their pairing with the reference allele and the alternative allele. More than 99.0% of the targeted loci are required to have ≥200× read depth to meet the coverage quality control requirements in the NIPS assay.

[0106] For NIPS assay, 5-10 ml of peripheral blood was collected from each participant. For twin pregnancies undergoing double reduction surgery, peripheral blood was collected at different time points before and after double reduction surgery. Peripheral blood was centrifuged twice at 4°C and then cfDNA was extracted from maternal plasma using a magnetic serum / plasma circulating DNA MaxiKit kit (Cat. No. YDP710, TIANGEN, China).

[0107] In order to construct the NGS library, the extracted cfDNA was end-repaired and connected with adapters according to the manufacturer's protocol (Cat.No.1002212, Nanodigbio, China). The sample barcode was introduced by PCR, and the PCR product was quantified with Qubit 1X dsDNA HS assay kit (Cat.No.Q33231, Thermo Fisher Scientific, Invitrogen, USA). After target enrichment (Cat.No.1072281, Integrated DNA Technologies, USA) with a probe (Cat.No.HS2008, Heristar, USA), the recovered DNA was washed and purified with Dynamag-270 magnetic beads (Cat.No.65306, Thermo Fisher SCIENTIFIC, Invitrogen, USA). In order to generate a sequencing library, PCR (Cat.No.1002212, Nanodigbio, China) was performed with 12 amplification cycles. Next, the concentration of the sequencing library was quantified using the Qubit ssDNA assay kit (Cat. No. Q10212, Thermo Fisher SCIENTIFIC, Invitrogen, USA). Finally, the library was sequenced on MGISEQ-2000 in 2 × 100 paired-end mode (Cat. No. 1000012554, MGI, Shenzhen, China). A previous study reported the detailed laboratory procedures.

[0108] The present invention includes 499 samples (170 for method development and 329 for validation). The sample size was chosen to enable the inventors to determine FF with a 10% error margin and 90% confidence in both data sets. Based on empirical data, the estimated population standard deviation was taken as 0.25 of the mean.

[0109] Example 1. Detection of zygosity of twins

[0110] 1.1 Development of a twin zygosity detection method based on fetal-specific SNPs and their allele fractions

[0111] cfDNA extracted from maternal plasma is a mixture of maternal and fetal genomic DNA. For any biallelic loci for which the mother is homozygous, fetal heterozygosity can be used to determine whether the twins are identical or fraternal.

[0112] The implications of twin zygosity based on informative loci are twofold: In DZ twins, the number of fetal-specific SNPs was approximately 25% greater than in singleton or MZ twins (Table 1).

[0113] Furthermore, the overall BAF (B allele fraction) changes of fetal-specific SNPs were different in DZ twins compared with MZ twins caused by two sets of FF ( Figure 5 ).

[0114] Table 1 Probability of twins with different genotypes and B allele fractions

[0115]

[0116] p: overall frequency of a specific SNP; BAF: B allele fraction; FF: fetal fraction; MZ: monozygotic; DZ: dizygotic.

[0117] Assume that the SNP frequency of the target population for a specific SNP is p, the total number of SNPs to be queried is N, the number of SNPs for which the mother is homozygous is N0, the number of heterozygous SNPs in singletons or monozygotic (MZ) twins among the SNPs for which the mother is homozygous is N1, and the number of heterozygous SNPs in dizygotic (DZ) twins is N2. The number of informative SNPs in DZ or MZ twins can be expressed as follows (lowercase letters are observed values, uppercase letters are derived values, and italic uppercase letters are used only for symbols).

[0118] N0=[(p 2 +(1-p) 2 )]*N

[0119] N1=[p 2 (1-p)+p(1-p) 2 ]*N=p(1-p)*N

[0120]

[0121]

[0122]

[0123] Thus, the number of fetal-specific SNPs was 25% greater than in singletons or MZ twins.

[0124] In singleton or MZ twin pregnancies, the B allele fraction (BAF) of informative SNPs at the loci in the fetus is expected to be FF / 2 or (1-FF / 2), while the mother is homozygous for the reference or alternative allele, respectively ( Figure 5 B and Table 1).

[0125] In a DZ twin pregnancy, assuming that FF1 and FF2 are the FFs of fetus 1 and fetus 2, respectively, the BAFs of informative fetal heterozygous SNPs are centered around FF1 / 2, FF2 / 2, (FF1FF2) / 2, 1-FF1 / 2, and 1-ff2 / 2, 1-(FF1FF2) / 2 in a DZ twin pregnancy ( Figure 5 C-5F and Table 1). BAF values ​​can then be used to infer twin zygosity ( Figure 5 ).

[0126] The maternal homozygous SNPs can be divided into two groups according to the maternal genotype (BB or AA), where the median BAF of these SNPs is M1 and M2. Then the standard deviation of the BAF of each group is calculated, that is, σ1 and σ2. The weighted average σ1 and σ2 is taken ave , which is weighted by the number of SNPs included in the statistics for each group. The median of the heterozygous SNP depth is denoted as M dep Then calculate the corrected BAF change σ cor , has a hypergeometric distribution.

[0127]

[0128] Therefore, the present invention establishes a binary algorithm that utilizes the number of informative fetal-specific SNPs and BAF or its relative fluctuation to efficiently and accurately determine twin zygosity, such as Figure 1 and Figure 2 As shown in A.

[0129] Based on the judgment results of the present invention, cell-free DNA samples collected from 11 confirmed DZ twin pregnancies and 57 singleton pregnancies were used for method validation, including 17 confirmed DZ twin pregnancies and 86 singleton pregnancies.

[0130] The validation results showed that all DZ twins were correctly detected, resulting in 100% (95% confidence interval, 77.1-100%) sensitivity and 100% (95% confidence interval, 94.7-100%) specificity. This shows that the method of the present invention has extremely high sensitivity and extremely high specificity.

[0131] 1.2 Verification of the effectiveness of the detection method

[0132] The method of the present invention was tested on a total of 323 samples from 215 singleton, 90 confirmed DZ and 18 confirmed MZ twin pregnancies.

[0133] All 90 DZ twin pregnancies in the test set were successfully identified with a sensitivity of 100% (95% CI, 94.9-100%) and a specificity of 100% (95% CI, 98.0-100%). Of note, twins with different karyotypes could be identified by different BAF patterns composed of different combinations of autosomal and X-chromosome SNPs ( Figure 1 ).

[0134] Furthermore, this new method can distinguish DZ twins from twin pregnancies with egg donor or assay artifacts (e.g., sample contamination) Figure 2 B and 2C; Figure 6 ).

[0135] In addition, twin pregnancies with complete molar fetuses and healthy fetuses were successfully identified ( Figure 2 D) To date, this is the first twin pregnancy detected by NIPS with healthy fetuses and intact molar pregnancy.

[0136] At any biallelic locus where the mother is homozygous and the fetus is heterozygous, the fetal-specific allele (reference allele or variant allele) is of reference significance for detecting twin zygotes and determining FF. Confirmed fraternal (DZ) twin pregnancies used for method development and validation are pregnancies where ultrasound imaging is consistent with twins of different sexes. The MZ twin pregnancy used in the present invention is monochorionic twins identified by ultrasound imaging, which shows a cfDNA SNP spectrum that is completely consistent with the number of fetal-specific SNPs and their allele score patterns described above.

[0137] Example 2. Calculation of fetal fraction

[0138] 2.1 Calculation of fetal fractions for fraternal twins of different sexes

[0139] In twins with different sexes, the X and Y chromosome reads can be used to calculate their respective FFs. Let FF1 and FF2 be the FFs for female fetus 1 and male fetus 2, respectively. Let the total haploid genome copies of fetus 1, fetus 2, and mother be R1, R2, and M, respectively, present in the sample.

[0140]

[0141]

[0142] For a given autosomal SNP i , whose reading depth is rd ai Similarly, for X and Y chromosome SNPs j , the reading depths are rd xj andrdyk For informative X chromosome SNPs f , let the read depth of the paternally inherited allele be rd xff Now, let the median read depth of SNPs for autosomes, chromosome X, and chromosome Y be RD a , R.D. x and RD y Finally, let RD xf is the median paternal genetic read depth for informative SNPs on chromosome X in fetus 1 (female twin). a , R.D. x , R.D. y and RD xf can be calculated as the observed rd ai , rd xj , rd yk andrd xff These values ​​are also a function of the number of maternal and fetal genome copies present in twins of different sexes, denoted as K a , K x and K y .

[0143] RD a =K a *2*(R1+R2+M) (3)

[0144]

[0145] RD y =K y *R2 (5)

[0146] RD xf =K x *R1 (6)

[0147] K a , K x and K y Can be calculated in euploid male samples. a , r x and r y The values ​​are the median read depths of autosomal, X- and Y-chromosome SNPs for such non-pregnant training samples, respectively.

[0148]

[0149]

[0150] According to equations 7 and 8, K x and K y It can be expressed as:

[0151] Kx =C1*K a (9)

[0152] K y =C2*K a (10)

[0153] Finally, FF1 and FF2 can be calculated as follows:

[0154]

[0155]

[0156] 2.2 Calculation of fetal fraction for same-sex fraternal twins

[0157] Unlike twin pregnancies of different sexes, sequencing data from chromosomes X and Y cannot discriminate the FF of same-sex twins. Therefore, BAFs of autosomal SNPs are used for FF calculation. Informative SNPs are loci that are homozygous in the mother and heterozygous in at least one twin. BAFs of these informative SNPs can be calculated based on different combinations of maternal and twin genotypes (Table 2).

[0158] Table 2 Expected B allele fraction levels in fraternal twins

[0159]

[0160] BAF: B allele fraction; R1: haploid genome copy number from fetus 1; R2: haploid genome copy number from fetus 2; M: haploid genome copy number from the mother.

[0161] The BAF of informative autosomal SNPs in DZ twins was centered on six theoretical levels when their respective FF levels were discernible ( Figure 5 C-5F). Since each twin can be heterozygous at both maternal homozygous BB and AA loci, the BAF is determined by the twin-specific genotypes and their read depth. Since the six BAF clusters are symmetrically distributed, for loci with BAF>0.5, their values ​​can be replaced by 1-BAF. Now, the BAF values ​​of the six clusters are reduced to three, which are expected to be centered around BAF1, BAF2, and BAF3 in ascending order. For DZ twins, assuming R1≤R2, the BAF can be expressed as follows. In order to calculate twin-specific FF, it is important to obtain accurate values ​​of BAF1, BAF2, and BAF3, especially when their levels are close to each other.

[0162] To this end, the BAF values ​​of all informative loci were grouped using a Bayesian Gaussian mixture model, which is a variant of the Gaussian mixture model. After all BAF values ​​of fetal-specific SNPs were assigned to different groups, the number of SNPs and the median BAF were calculated for each SNP cluster. Clusters with less than 5 SNPs were omitted to suppress outliers. When three clusters were left, the median BAF values ​​of these three reduced clusters, bafm1, and bafm3, were taken.

[0163]

[0164]

[0165]

[0166] Where BAF is the B allele fraction; R1 is the haploid genome copy number from fetus 1; R2 is the haploid genome copy number from fetus 2; and M is the haploid genome copy number from the mother.

[0167] Based on equations 13, 14, and 15, R1 and R2 can be expressed as:

[0168]

[0169]

[0170] Finally, the FF for each twin can be calculated from equations 1, 2, 18, and 19:

[0171]

[0172]

[0173] 2.3 Calculation of fetal fraction in twin pregnancy

[0174] After the NIPS method for detecting twin zygosity was established, its utility in twin-specific FF calculations was investigated. For singleton pregnancies, autosomal SNPs were first used to determine FF. In a trial of 338 male singleton pregnancies, the autosomal SNP and sex chromosome methods produced highly concordant FF calculations (concordance correlation coefficient or CCC = 0.953, 95% CI: 94.2-96.2%, Figure 3 A).

[0175] Next, an FF algorithm for DZ twin pregnancies was developed. At informative loci, the BAFs of twin-specific autosomal SNPs were used to calculate their respective FFs. In addition, NGS read depth data from chromosomes X and Y were used to determine twin-specific FFs. In 9 samples collected from pregnancies with twins of different sexes, FFs for each twin were calculated by both autosomal and sex chromosome methods (Table 3). The FFs calculated by both methods were strongly correlated (fetus 1: CCC = 0.970, 95% CI: 90.0-99.1%, Figure 3 B; Fetus 2: CCC = 0.915, 95% CI: 68.5-97.9%, Figure 3 C).

[0176] To verify the accuracy of the twin-specific FF calculation method, a total of 118 DZ twins were tested, of which 57 DZ twins had different sexes (Table 4). The FF calculated for these twins by the autosomal and sex chromosome methods again produced consistent results (fetus 1: CCC = 0.908, 95% CI: 84.9-94.4%; fetus 2: CCC = 0.812, 95% CI: 70.3-88.3%).

[0177] These results indicate that the present autosomal method is accurate for calculating twin-specific FFs without the need for sex-specific loci.

[0178] Table 3 Calculation of fetal fraction for each twin pair using autosomal and sex chromosome methods

[0179]

[0180] FF: Fetal Fraction

[0181] Table 4 Calculation of fetal fraction of fraternal twins using autosomal and sex chromosome methods

[0182]

[0183]

[0184]

[0185] FF1: fetal fraction of fetus 1; FF2: fetal fraction of fetus 2.

[0186] 2.5 Changes in fetal fraction after twin reduction

[0187] Following the development of the FF calculation, this method was used to measure changes in FF in pregnancies following twin reduction. FF levels were measured in nine twin pregnancies before and after one of the twins was reduced ( Figure 4FF decreased in approximately 80% (7 / 9) of the non-developing twins as pregnancy progressed until 7 weeks after twin reduction ( Figure 4 A). Interestingly, in all nine 44% (4 / 9) cases, FF decreased significantly within two days after the decrease, although it increased in two subjects and remained unchanged in the rest of the pregnancies ( Figure 4 A). For the surviving twin ( Figure 4 B), 67% (6 / 9) of fetuses had an immediate increase in FF within 2 days followed by a reduction in FF, while 44% (4 / 9) had an overall decrease in FF within 7 weeks.

[0188] FF variation demonstrated highly variable cfDNA production dynamics in both non-viable and surviving twins ( Figure 4 and Table 5). Therefore, NIPS involving cfDNA testing should be used with caution for pregnancies that are at least 7 weeks after twin reduction.

[0189] Table 5 Fetal fractions before and after twin reduction

[0190]

[0191] FF1: fetal fraction of the non-viable twin; FF2: fetal fraction of the viable twin;

[0192] Fetal fraction calculated by the sex chromosome method; GA: gestational age.

[0193] discuss

[0194] The major risk factor associated with adverse pregnancy outcomes in twin pregnancies is chorionicity. Monochorionic twins are susceptible to twin-to-twin transfusion syndrome and twin anemia-polycythemia syndrome. Chorionicity can be identified with appropriate antenatal management in the first trimester by ultrasonography. However, this test is not available to all pregnant women and the accuracy of the ultrasound method has been questioned. DZ twin pregnancies are predominantly dichorionic, whereas approximately 75% of MZ twins are monochorionic and the remaining 25% are dichorionic. Therefore, establishing zygosity helps determine chorionicity, which in turn facilitates antenatal management of high-risk twin pregnancies.

[0195] In conventional NIPS assays using targeted SNP assays, detection of DZ twins deteriorates when twins inherit parental chromosomes with an increase in the size of identical descent segments, thereby reducing the number of genetically unique DZ twin loci.

[0196] Compared with traditional biochemical methods and ultrasound screening, NIPS is more accurate in detecting fetal aneuploidy by analyzing cfDNA in fetal plasma. However, the performance of NIPS is challenged by residual cfDNA from an unrecognized twin or a nondeveloping fetus caused by twin death or twin reduction, which can result in approximately 15-33% false-positive NIPS results. The presence of cfDNA from the placenta of the nondeveloping twin interferes with the analysis of the surviving twin. Therefore, it is important to determine the half-life of cfDNA from the nondeveloping fetus to better use NIPS in twin pregnancies.

[0197] The inventors have developed a new SNP-based NIPS method that detects twin heterozygosity based on the allele fractions of fetal-specific SNPs and the number of such SNPs. Since the SNP allele fractions of each twin are a function of their respective FFs, the new method takes into account two independent parameters related to twin zygosity, which makes the determination of twin zygosity more accurate. In addition to true DZ twins, this method can also be used for samples complicated by egg donors, sample contamination, and twins with complete hydatidiform moles ( Figure 2 B-2D and Figure 6 ). Overall, the NIPS method of the present invention has very high accuracy and versatility for determining twin zygosity.

[0198] In the present invention, a method based on autosomes was established. Taking a validation experiment based on actual samples as an example, the method of the present invention can calculate the FF ( Figure 3 B and 3C). FF decreased significantly in 44% (4 / 9) of the nondeveloping twins within two days of reduction, whereas it increased in two subjects and remained unchanged in the other subjects ( Figure 4 A). For the surviving twin ( Figure 4 B), 67% (6 / 9) of fetuses had a rapid increase in FF within two days after twin reduction, whereas 44% (4 / 9) had an overall decreased FF after twin reduction at 7 weeks.

[0199] The changes in FF after twin reduction suggest that cfDNA production is highly dynamic in both the nondeveloped and surviving twins. A previous study demonstrated that unpredictable FF patterns after twin reduction do not support cfDNA testing at any given time interval after reduction. Another study found a peak FF at 7–9 weeks after twin reduction, followed by a gradual decrease with increasing gestational age. This may be due to changes in the maternal intrauterine physiological environment after twin reduction. The unclear rate of placental tissue autolysis after twin reduction may explain the complex changes in FF after twin reduction. Therefore, NIPS must be used with caution, as the FF of the nondeveloped twin remains high seven weeks after twin reduction.

[0200] In summary, the present invention provides a novel SNP-based NIPS method for detecting twin pregnancy zygosity and calculating FF using the number and allele fraction of fetal-specific SNPs. This method can also be used to monitor FF changes after twin reduction to manage high-risk twin pregnancies.

[0201] All documents mentioned in the present invention are cited as references in this application, just as each document is cited as reference individually. In addition, it should be understood that after reading the above teachings of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to this application.

Claims

1. A detection device for determining fetal zygosity, characterized in that: The device comprises: (a) a data input module, the data input module is used to input nucleic acid data of a sample to be analyzed, the nucleic acid data including sequence data based on fetal free deoxyribonucleic acid in the peripheral blood of a pregnant woman; (b) an analysis and calculation module, the analysis and calculation module being configured to analyze the nucleic acid data of the sample to be analyzed, thereby determining (i) the relative heterozygous abundance of fetal-specific alleles and (ii) the fetal-specific allele fraction or its relative fluctuation; (c) an evaluation module, the evaluation module being configured to evaluate the relative heterozygous abundance of the fetal-specific alleles and the fetal-specific allele fraction or the relative fluctuation thereof, thereby obtaining a detection result of fetal zygosity; and (d) an output module, which is used to output the detection result of the fetal zygosity.

2. The device according to claim 1, characterized in that Assume that the SNP frequency of the target population of a specific single nucleotide polymorphism (SNP) is p, the total number of queried SNPs is N, the number of SNPs that are homozygous in the mother is N0, the number of heterozygous SNPs in singletons or monozygotic (MZ) twins among the SNPs that are homozygous in the mother is N1, and the number of heterozygous SNPs in fraternal (DZ) twins is N2; The number of informative SNPs in DZ or MZ twin pairs was calculated by the following formula, where lowercase letters are observed values, uppercase letters are derived values, and italic uppercase letters are used only for symbols: N0=[(p 2 +(1-p) 2 )]*N N1=[p 2 (1-p)+p(1-p) 2 ]*N=p(1-p)*N 3. The device according to claim 1, characterized in that In the analysis and calculation module, the B allele fraction (BAF) value is calculated.

4. The device according to claim 3, characterized in that The calculation of the B allele fraction (BAF) value includes: For maternally homozygous SNPs, they were divided into two groups according to the maternal genotype (BB or AA), where the median BAFs of these SNPs were M1 and M2; Calculate the standard deviation of BAF for each group, i.e., σ1 and σ2; Take the weighted average σ of σ1 and σ2 ave , which is the weighted number of SNPs included in the statistics for each group, and the median of the heterozygous SNP depth is expressed as M dep ; The corrected BAF change σ is calculated as follows: cor 5. The device according to claim 1, characterized in that The analysis and calculation module further includes a fetal score calculation submodule, wherein the fetal score calculation submodule is configured to analyze the nucleic acid data of the sample to be analyzed, thereby determining the fetal score; And the assessment module also includes a twin pregnancy risk assessment submodule, and the twin pregnancy risk assessment submodule is configured to evaluate the fetal score to obtain a twin pregnancy risk value.

6. The device according to claim 1, characterized in that The calculation of the fetal fraction includes: Based on equations 1, 2, 18 and 19, the FF of each twin is calculated: In the formula, the total haploid genome copies of fetus 1, fetus 2, and mother are R1, R2, and M present in the sample, respectively; Among them, bafm1, bafm2, and bafm3 are calculated according to formulas 13-15 respectively; Where BAF is the B allele fraction; R1 is the haploid genome copy number from fetus 1; R2 is the haploid genome copy number from fetus 2; and M is the haploid genome copy number from the mother.

7. The device according to claim 1, characterized in that The fetal fraction calculation also includes equations 13, 14 and 15 to obtain R1 and R2 values: wherein bafm1, bafm2, and bafm3, R1, R2, and M are as defined above.

8. The device according to claim 1, characterized in that The sample to be analyzed is a blood sample.

9. The device according to claim 1, characterized in that The data input module is selected from the following group: a sequencer, a scanner, a keyboard, a tablet computer (PAD), a smart phone, or a combination thereof.

10. A method for data analysis of fetal zygosity, characterized in that: The method comprises the following steps: (a) providing data, wherein the data includes nucleic acid data of a sample to be analyzed, wherein the nucleic acid data includes sequence data based on fetal free deoxyribonucleic acid in the peripheral blood of a pregnant woman; (b) analyzing and calculating the nucleic acid data of the sample to be analyzed to determine (i) the relative heterozygous abundance of fetal-specific alleles and (ii) the fetal-specific allele fraction or its relative fluctuation; and (c) evaluating the results, evaluating the relative heterozygous abundance of the fetal-specific alleles and the fetal-specific allele fraction or their relative fluctuations, thereby obtaining a detection result of the fetal zygosity.

Images