Methods for estimating fetal fraction in cell-free DNA from maternal samples

Abstract

Digital amplification methods and kits provide the ability to estimate the fetal fraction of cell-free DNA (cfDNA) in maternal samples, e.g., plasma or serum, by analysis of differentially methylated target sites in fetal and maternal cfDNA.

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This patent application claims the benefit of U.S. Provisional Patent Application No. 63 / 406,194, filed September 13, 2022, and U.S. Provisional Patent Application No. 63 / 472,183, filed June 9, 2023, each of which is incorporated by reference for all purposes. [Background technology]

[0002] Accurate diagnosis of chromosomal aneuploidies, such as trisomies 13, 18, and 21, requires accurate estimation of the fetal fraction of cell-free DNA (cfDNA) in maternal plasma. Cell-free DNA is present at low levels (0–100 ng / mL) in maternal whole blood, and the median fetal fraction of cfDNA is approximately 11% in the first trimester but may be less than 4% in the early stages of pregnancy when noninvasive prenatal testing (NIPT) is recommended (Wang et al., Prenat. Diagn. 33:662–666, 2013). Furthermore, despite their high gene sequence similarity, fetal cfDNA must be distinguished from maternal cfDNA. [Prior art documents] [Non-patent literature]

[0003] [Non-Patent Document 1] Wang et al, Prenat. Diagn. 33:662-666, 2013 Summary of the Invention

[0004] In one aspect, the present disclosure provides a digital PCR (dPCR) method for quantifying the fetal fraction of cfDNA in maternal plasma using methylation-sensitive restriction enzyme (MSRE) digestion. In some embodiments, the present disclosure provides a method for estimating the proportion of fetal DNA in a cfDNA sample obtained from a blood sample of a pregnant human subject. In some embodiments, the method includes: (a) partitioning (e.g., distributing) an amplification reaction mixture containing cfDNA from the cfDNA sample, amplification reagents, and multiple amplification sets containing primer and probe sets into partitions, each amplification set containing primers and probes for multiplex amplification, each amplification set generating an amplification product (containing a distinct label, if present, that is distinguishable from the labels for each of the other amplification sets), and the multiple amplification sets include: (i) an amplification set targeting sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA; (ii) an amplification set targeting sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA; and, optionally, one or more of (iii), (iv), and (v), (iii) a target exhibiting aneuploidy. The digital amplification reaction method includes: (a) an amplification set targeting total cfDNA containing methylation-insensitive regions of chromosomes unlikely to exhibit aneuploidy; (b) an amplification set targeting hypermethylated sites in fetal DNA and maternal DNA; (c) an amplification set targeting hypomethylated sites in fetal DNA and maternal DNA; (d) incubating cfDNA with a methylation-sensitive restriction enzyme (MSRE) cocktail containing at least one methylation-sensitive restriction enzyme that cleaves unmethylated (e.g., hypomethylated) DNA; (e) amplifying target nucleic acid sequences, if present, within partitions to obtain amplification products; (f) detecting signals from each distinct label of the amplification products within the partitions; and (f) quantifying the signals from each distinct label. In some embodiments, the multiple amplification sets include (iii) an amplification set targeting total cfDNA containing methylation-insensitive regions of chromosomes unlikely to exhibit aneuploidy. In some embodiments, each of the amplification sets (i) and (ii) includes primers and probes for targeting at least three sites.In some embodiments, each of the amplification sets (i) to (v) includes primers and probes targeting at least three sites or six to ten sites. In some embodiments, the method further includes using an amplification set targeting a methylation-insensitive region of the Y chromosome. In additional embodiments, the method may further include using an amplification set targeting hypomethylated sites in both fetal and maternal cfDNA and / or an amplification set targeting hypermethylated sites in both fetal and maternal cfDNA. In some embodiments, the amplification reaction mixture further includes a control target fully methylated synthetic DNA sequence and / or a fully unmethylated version of the same synthetic DNA sequence. In some embodiments, the digital amplification reaction method is a digital PCR method, such as a droplet digital PCR method. In some embodiments, the amplification reaction mixture in the distributing step includes an MSRE cocktail, and incubating is performed after distributing and before amplifying. In some embodiments, step (b) is performed before distributing, and the digested cfDNA is added to the amplification reaction mixture. In some embodiments, the MSRE cocktail includes at least two, at least three, or at least four methylation-sensitive restriction enzymes, and / or the MSRE cocktail includes a restriction enzyme selected from HhaI, HpaII, AciI, HpyCH4IV, and BsaHI. In some embodiments, the MSRE cocktail includes at least two or three of the restriction enzymes HhaI, HpaII, AciI, HpyCH4IV, and BsaHI. In some embodiments, the MSRE cocktail includes at least HhaI and HpyCH4IV. In some embodiments, the cfDNA sample is obtained from plasma or serum. In some embodiments, each label is a fluorescent label. In some embodiments, the probe is a molecular beacon probe containing a fluorescent label. In some embodiments, each probe is an oligonucleotide that hybridizes to a complementary oligonucleotide containing a label that provides a detectable signal. In some embodiments, the cfDNA is incubated with the MSRE cocktail in a partition.In some embodiments, the incubation of the cfDNA with the MSRE cocktail is performed in bulk solution prior to distribution (a). In some embodiments, the method is carried out by, for example, determining the number of targets in the amplification set (N). i In some embodiments, the method further comprises determining a normalized copy number concentration of each target based on a corrected concentration of fetal cfDNA (Fet Corr ) and / or the corrected concentration of maternal cfDNA in the cfDNA sample (Mat Corr ), wherein determining the corrected concentration of fetal cfDNA includes calculating:

number

number

number

[0005] In some embodiments, the method further comprises calculating an estimated fetal fraction based at least in part on the fetal fraction in the cfDNA sample and the model. In some embodiments, the model is a generalized additive model (GAM), a linear model, or a quadratic polynomial model based at least in part on a dataset of clinical fetal fractions and a set of corresponding fetal fraction measurements using next-generation sequencing (NGS).

[0006] In some embodiments, the method further includes determining a fetal fraction of male fetuses in the cfDNA sample, wherein determining the fetal fraction of male fetuses comprises:

number

[0007] In some embodiments of the methods or kits described herein, the sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA are sites within one or more or all of the following ranges: TIFF2025531887000006.tif75128

[0008] In some embodiments of the methods or kits described herein, the sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA are sites within one or more or all of the following ranges: TIFF2025531887000007.tif67128

[0009] In some embodiments of the methods or kits described herein, the hypermethylated sites in the fetal DNA and maternal DNA are sites within one or more or all of the following ranges: TIFF2025531887000008.tif66128

[0010] In some embodiments of the methods or kits described herein, the sites that are hypomethylated in the fetal DNA and maternal DNA are sites within one or more or all of the following ranges: TIFF2025531887000009.tif58128

[0011] In some embodiments of the methods or kits described herein, the methylation-insensitive regions of a chromosome that are unlikely to exhibit aneuploidy are sites within one or more or all of the following: TIFF2025531887000010.tif84128

[0012] In some embodiments of the methods or kits described herein, the methylation-insensitive region of the Y chromosome is within one or more or all of the following: TIFF2025531887000011.tif93128

[0013] In a further aspect, the present disclosure provides a digital amplification kit for estimating the percentage of fetal DNA in a cfDNA sample obtained from a plasma or serum sample of a pregnant human subject, comprising: (a) an amplification reaction mixture comprising amplification reagents and a plurality of amplification sets comprising primer and probe sets, each amplification set comprising a distinct label distinguishable from the label for each of the other sets, each set comprising primers and probes for multiplex amplification, the plurality of amplification sets comprising: (i) an amplification set targeting sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA; (ii) an amplification set targeting sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA, and optionally one or more of (iii), (iv), and (v); (iii) an amplification set targeting total cfDNA containing methylation-insensitive regions of chromosomes unlikely to exhibit aneuploidy; (iv) an amplification set targeting sites hypermethylated in fetal and maternal DNA; (v) an amplification reaction mixture comprising an amplification set targeting hypomethylated sites in fetal DNA and maternal DNA. In some embodiments, each of the amplification sets (i) and (ii) comprises primers and probes for targeting at least three sites. In some embodiments, the kit further comprises an amplification set targeting a methylation-insensitive region of the Y chromosome. In some embodiments, the kit further comprises a fully methylated synthetic sequence of DNA and / or a fully unmethylated version of the same synthetic sequence. In some embodiments, each label is a fluorescent label. In some embodiments, the probe is a molecular beacon probe comprising a fluorescent label. In some embodiments, each probe is an oligonucleotide that hybridizes to a complementary oligonucleotide comprising a label that provides a detectable signal. In some embodiments, the kit further comprises a methylation-sensitive restriction enzyme (MSRE) cocktail comprising at least one MSRE that cleaves hypomethylated DNA. In some embodiments, the MSRE cocktail includes at least two, at least three, or at least four methylation-sensitive restriction enzymes, and / or the MSRE cocktail includes a restriction enzyme selected from HhaI, HpaII, AciI, HpyCH4IV, and BsaHI. In some embodiments, the kit includes a methylation-sensitive restriction enzyme (MSRE) cocktail including at least two, three, or more of the restriction enzymes HhaI, HpaII, AciI, HpyCH4IV, and BsaHI. In some embodiments, the MSRE cocktail includes at least HhaI and HpyCH4IV. In some embodiments, the kit further includes a methylation-sensitive restriction enzyme (MSRE) cocktail including at least two, three, or more of the restriction enzymes HhaI, HpaII, AciI, HpyCH4IV, and BsaHI.

[0014] In one embodiment, a data set of calculated fetal fraction is created based on the disclosed method and corresponding measured fetal fraction by using next generation sequencing (NGS) of various NIPT clinical samples from human subjects.Several models have been developed that can take into account not only calculated fetal fraction and corresponding NGS measurement values, but also parameters such as gestational age and Y chromosome calculation.

[0015] term Unless otherwise defined, all technical and scientific terms used herein generally have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Generally, the nomenclature used herein and the laboratory methods in cell culture, molecular genetics, organic chemistry, and nucleic acid chemistry and hybridization described below are well known and commonly employed in the art. Standard techniques are used for nucleic acid and peptide synthesis. Techniques and procedures are generally performed according to conventional methods in the art and various general references provided throughout this literature (generally, Sambrook et al. MOLECULAR CLONING: A LABORATORY MANUAL, 2nd ed. (1989) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, which is incorporated herein by reference). The nomenclature used herein and the laboratory methods in analytical chemistry and organic synthesis described below are well known and commonly employed in the art.

[0016] The term "cell-free DNA sample" or "cfDNA sample" refers to a nucleic acid sample containing extracellular DNA, and the nucleic acid sample is obtained from any cell-free biological fluid, such as whole blood that has been processed to remove cells, urine, saliva, or other biological fluids. In a typical embodiment, the cfDNA for analysis is obtained from whole blood that has been processed to remove cells, such as plasma or serum samples. Thus, as used herein, the term "cfDNA" refers to DNA that can be recovered from the non-cellular fraction of a bodily fluid, such as blood.

[0017] Methylation status refers to the presence of methyl groups in a specific DNA sequence. In some embodiments, DNA methylation refers to the presence or absence of methylcytosine at one or more CpG dinucleotides within a DNA sequence. The term "methylation state" or "methylation status" with respect to CpG dinucleotide methylation refers to the presence or absence of 5-methylcytosine ("5-mC" or "5-mCyt") at one or more CpG dinucleotides within a DNA sequence. Methylation status at one or more specific methylation sites within a DNA sequence includes "unmethylated," "fully methylated," and "semi-methylated." For purposes of this application, the term "hypermethylated" refers to a region in which the average frequency of methylation for a particular subset of samples, e.g., fetal DNA or maternal DNA, is greater than 80% as determined by methylation sequencing. "Hypomethylated" refers to a region in which the average frequency of methylation for a particular subset of samples, e.g., fetal DNA or maternal DNA, is less than 20% as determined by methylation sequencing.

[0018] Methylation-sensitive restriction enzymes (MSREs) refer to enzymes that cleave DNA at specific unmethylated cytosine residues, but do not cleave at their recognition sequences if the cytosine residues are methylated.

[0019] As used herein, "methylation-sensitive" genomic region refers to genomic DNA that can be methylated, for example, at CpG sequences, whereby the site is not cleavable by methylation-sensitive restriction enzymes in the methylated state, but is cleavable when methylation is absent.As used herein, "cleavable" means that when the recognition sequence is unmethylated, at least 50% of the DNA is digested by methylation-sensitive restriction enzymes compared to when the recognition sequence is methylated.Therefore, detection of an amplification product obtained from the amplification of cfDNA containing a methylation site after digestion with MSRE indicates that the cfDNA is methylated at that site.

[0020] The term "amplification reaction" refers to any in vitro method for linearly or exponentially doubling the copy number of a nucleic acid target sequence. Such methods include, but are not limited to, two-pair primer methods such as polymerase chain reaction (PCR); ligase methods such as DNA ligase chain reaction (U.S. Pat. Nos. 4,683,195 and 4,683,202; PCR Protocols: A Guide to Methods and Applications (Innis et al., eds., 1990)) (LCR); QBeta RNA replicase and RNA transcription-based amplification reactions (e.g., amplification involving T7, T3, or SP6-induced RNA polymerization), such as transcription amplification system (TAS), nucleic acid sequence-based amplification (NASBA), and self-sustained sequence replication (3SR); isothermal amplification reactions (e.g., single primer isothermal amplification (SPIA)); and others known to those skilled in the art.

[0021] "Amplification" refers to subjecting a solution to conditions sufficient to allow amplification of a polynucleotide when all components of the reaction are intact. Components of an amplification reaction include, for example, primers, a polynucleotide template, a polymerase, nucleotides, etc. In some embodiments, "amplification" refers to PCR amplification using first and second amplification primers.

[0022] A "primer" refers to a polynucleotide sequence that hybridizes to a target nucleic acid sequence and serves as a starting point for nucleic acid synthesis. Primers can be of various lengths, often less than 100 nucleotides, e.g., 18 to 55 nucleotides in length. The length and sequence of primers for use in amplification reactions such as PCR can be designed based on principles known to those skilled in the art. Primers can be DNA, RNA, or chimeras of DNA and RNA portions. In some cases, primers can contain one or more modified or non-natural nucleotide bases. In some cases, primers are labeled. In some cases, primers may also contain nucleic acid sequences that are not involved in hybridization to the target for amplification, such as sequences that hybridize to another labeled oligonucleotide, or a sequence that hybridizes to a capture oligonucleotide, or a tag sequence such as a barcode.

[0023] A nucleic acid or a portion thereof "hybridizes" to another nucleic acid in a physiological buffer at a specified temperature under conditions that minimize nonspecific hybridization. In some cases, a nucleic acid or a portion thereof hybridizes to a conserved sequence shared among a group of target nucleic acids. In some cases, a primer or a portion thereof can hybridize to a primer binding site when there are at least about 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 adjacent complementary nucleotides, including "universal" nucleotides that are complementary to two or more nucleotide partners. Alternatively, a primer or portion thereof can hybridize to a primer binding site with one or fewer than two complementary mismatches across at least about 12, 13, 14, 15, 16, 17, 18, 19, 20, 25, or 30 contiguous complementary nucleotides. In some embodiments, the specified temperature at which specific hybridization occurs is room temperature. In some embodiments, the specified temperature at which specific hybridization occurs is above room temperature. In some embodiments, the specified temperature at which specific hybridization occurs is at least about 37°C, 40°C, 42°C, 45°C, 50°C, 55°C, 60°C, 65°C, 70°C, 75°C, or 80°C, e.g., about 45°C to about 60°C, e.g., about 55°C to 59°C. In some embodiments, the specified temperature at which specific hybridization occurs is about 5°C below the calculated melting temperature of the primer.

[0024] As used herein, "nucleic acid" refers to DNA, RNA, single-stranded, double-stranded, or more highly aggregated hybridization motifs, and any chemical modifications thereof. Modifications include, but are not limited to, modifications that provide additional charge, polarizability, hydrogen bonding, electrostatic interactions, and chemical groups that incorporate attachment points and functionality to the nucleic acid ligand bases or the nucleic acid ligand as a whole. Such modifications include, but are not limited to, peptide nucleic acids (PNAs), modifications of phosphodiester groups (e.g., phosphorothioates, methylphosphonates), 2'-sugar modifications, 5-position pyrimidine modifications, 8-position purine modifications, modifications with exocyclic amines, 4-thiouridine substitutions, 5-bromo- or 5-iodo-uracil substitutions, backbone modifications, methylation, and unusual base pair combinations (such as isobases, isocytidines, and isoguanidines). Nucleic acids may also contain unnatural bases, such as nitroindoles. Modifications may also include 3' and 5' modifications, including, but not limited to, capping with fluorophores (e.g., quantum dots) or other moieties.

[0025] "Polymerase" refers to an enzyme that performs template-directed synthesis of polynucleotides, such as DNA. The term encompasses both full-length polypeptides and domains with polymerase activity. At least five families of DNA-dependent DNA polymerases are known, with most classified into families A, B, and C. DNA polymerases are well known to those skilled in the art. The DNA polymerase for use in the compositions and methods disclosed herein can be any polymerase capable of replicating DNA molecules. In some embodiments, the DNA polymerase is a thermostable polymerase. Thermostable polymerases are isolated from a wide variety of thermophilic bacteria, such as Thermus aquaticus (Taq), Pyrococcus furiosus (Pfu), Pyrococcus woesei (Pwo), Bacillus sterothermophilus (Bst), Sulfolobus acidocaldarius (Sac), Sulfolobus solfataricus (Sso), Pyrodictium occultum (Poc), Pyrodictium abyssi (Pab), and Methanobacterium thermoautotrophicum (Mth), as well as other species. DNA polymerases are known in the art and commercially available. In some embodiments, the DNA polymerase is Taq, Tbr, Tfl, Tru, Tth, Tli, Tac, Tne, Tma, Tih, Tfi, Pfu, Pwo, Kod, Bst, Sac, Sso, Poc, Pab, Mth, Pho, ES4, VENT™, DEEPVENT™, or an active mutant, variant, or derivative thereof. In some embodiments, the DNA polymerase is Taq DNA polymerase. In some embodiments, the DNA polymerase is a high-fidelity DNA polymerase (e.g., iProof™ High-Fidelity DNA Polymerase, Phusion® High-Fidelity DNA Polymerase, Q5® High-Fidelity DNA Polymerase, Platinum® Taq High-Fidelity DNA Polymerase, Accura® High-Fidelity Polymerase).In some embodiments, the DNA polymerase is a fast-start or hot-start polymerase (eg, FastStart™ Taq DNA polymerase, FastStart™ High-Fidelity DNA polymerase, or iTaq™ DNA polymerase).

[0026] As used herein, the terms "distribute" or "distributed" refer to the separation of a sample into multiple portions or "partitions." The partitions are generally physical, such that the sample in one partition does not mix, or is substantially unmixed, with the sample in an adjacent partition. The partitions can be solid or fluid. In some embodiments, the partitions are solid partitions, e.g., microchannels or microwells. In some embodiments, the partitions are fluid partitions, e.g., droplets. In some embodiments, the fluid partitions (e.g., droplets) are mixtures of immiscible fluids (e.g., water and oil). In some embodiments, the fluid partitions (e.g., droplets) are aqueous droplets surrounded by an immiscible carrier liquid (e.g., oil). Exemplary arrays of wells and descriptions of wells can be found, for example, in U.S. Pat. Nos. 9,103,754 and 10,391,493. Arrays of wells (nanowells, microwells, sets of wells) can function to capture solid supports at known, optionally addressable locations. In this manner, the array of wells can be configured to facilitate bead capture in a single solid support format or, optionally, on at least one of a small group of solid supports. Exemplary microwell arrays and methods for delivering beads to the microwells and analyzing them are described, for example, in PCT / US2021 / 034152.

[0027] An "oligonucleotide" is a polynucleotide. In many embodiments, an oligonucleotide has fewer than 250 nucleotides, and in some embodiments, between 4 and 200 nucleotides, for example, between 10 and 150 nucleotides.

[0028] As used herein, the terms "a," "an," or "the" include not only embodiments having one element, but also embodiments having two or more elements. For example, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, a reference to a "bead" includes a plurality of such beads, a reference to an "array" includes a reference to one or more sequences known to those of skill in the art, and so forth. [Brief explanation of the drawings]

[0029] [Figure 1] Figure 1 shows fetal fraction estimation by methylation-sensitive digestion ddPCR and Y chromosome ddPCR. This figure shows the correlation between methylation-sensitive and SRY-based fetal fraction estimation for the exemplary assays detailed herein, including in the Examples section. [Figure 2A] 2A-2D show the analysis of methylation patterns for maternal and fetal methylated sites and ubiquitously methylated or unmethylated sites. [Figure 2B] 2A-2D show the analysis of methylation patterns for maternal and fetal methylated sites and ubiquitously methylated or unmethylated sites. [Figure 2C] 2A-2D show the analysis of methylation patterns for maternal and fetal methylated sites and ubiquitously methylated or unmethylated sites. [Figure 2D] 2A-2D show the analysis of methylation patterns for maternal and fetal methylated sites and ubiquitously methylated or unmethylated sites. [Figure 3]Figure 3 shows how methylation-sensitive restriction enzymes (MSREs) enable fetal and maternal cfDNA quantification by ddPCR based on differences in fetal and maternal cfDNA methylation. As shown, MSRE digestion is performed within droplets and does not interfere with the ddPCR workflow. Fetal and maternal cfDNA are quantified simultaneously in the same ddPCR reaction. 1) Hypermethylated fetal cfDNA is quantified after MSRE digestion of hypomethylated maternal cfDNA. 2) Maternal cfDNA is quantified after MSRE digestion of hypomethylated fetal cfDNA. 3) Total cfDNA is quantified from the undigested region. [Figure 4A] Figures 4A-B show a multiplex ddPCR assay. Figure 4A shows the general ddPCR assay format. Figure 4B shows that multiple primer pairs targeting the same chromosome are combined in a single fluorescent channel using a unique universal probe. [Figure 4B] Figures 4A-B show a multiplex ddPCR assay. Figure 4A shows the general ddPCR assay format. Figure 4B shows that multiple primer pairs targeting the same chromosome are combined in a single fluorescent channel using a unique universal probe. [Figure 5A] Figures 5A-D show the estimation of fetal fraction using the linear model, polynomial model, and generalized additive model (GAM) developed herein, including in Example 2. Figure 5A shows the estimation of fetal fraction using the developed linear model ("LM") and the traditional method ("FF_calculated") compared to the estimation of fetal fraction using next-generation sequencing (NGS). Figure 5B shows the calculated fetal fraction relative to the variables fetal / high and maternal / high. The residual plot shows the nonlinear relationship of fetal / high (510) compared to maternal / high (512). Figure 5C shows the estimation of fetal fraction using a second-order polynomial model ("poly2LM") compared to the estimation of fetal fraction using NGS. Figure 5D shows the GAM for the estimation of fetal fraction compared to the estimation of fetal fraction using NGS. MPAE, mean absolute percent error. MSE, mean squared error. [Figure 5B] Figures 5A-D show the estimation of fetal fraction using the linear model, polynomial model, and generalized additive model (GAM) developed herein, including in Example 2. Figure 5A shows the estimation of fetal fraction using the developed linear model ("LM") and the traditional method ("FF_calculated") compared to the estimation of fetal fraction using next-generation sequencing (NGS). Figure 5B shows the calculated fetal fraction relative to the variables fetal / high and maternal / high. The residual plot shows the nonlinear relationship of fetal / high (510) compared to maternal / high (512). Figure 5C shows the estimation of fetal fraction using a second-order polynomial model ("poly2LM") compared to the estimation of fetal fraction using NGS. Figure 5D shows the GAM for the estimation of fetal fraction compared to the estimation of fetal fraction using NGS. MPAE, mean absolute percent error. MSE, mean squared error. [Figure 5C] Figures 5A-D show the estimation of fetal fraction using the linear model, polynomial model, and generalized additive model (GAM) developed herein, including in Example 2. Figure 5A shows the estimation of fetal fraction using the developed linear model ("LM") and the traditional method ("FF_calculated") compared to the estimation of fetal fraction using next-generation sequencing (NGS). Figure 5B shows the calculated fetal fraction relative to the variables fetal / high and maternal / high. The residual plot shows the nonlinear relationship of fetal / high (510) compared to maternal / high (512). Figure 5C shows the estimation of fetal fraction using a second-order polynomial model ("poly2LM") compared to the estimation of fetal fraction using NGS. Figure 5D shows the GAM for the estimation of fetal fraction compared to the estimation of fetal fraction using NGS. MPAE, mean absolute percent error. MSE, mean squared error. [Figure 5D]Figures 5A-D show the estimation of fetal fraction using the linear model, polynomial model, and generalized additive model (GAM) developed herein, including in Example 2. Figure 5A shows the estimation of fetal fraction using the developed linear model ("LM") and the traditional method ("FF_calculated") compared to the estimation of fetal fraction using next-generation sequencing (NGS). Figure 5B shows the calculated fetal fraction relative to the variables fetal / high and maternal / high. The residual plot shows the nonlinear relationship of fetal / high (510) compared to maternal / high (512). Figure 5C shows the estimation of fetal fraction using a second-order polynomial model ("poly2LM") compared to the estimation of fetal fraction using NGS. Figure 5D shows the GAM for the estimation of fetal fraction compared to the estimation of fetal fraction using NGS. MPAE, mean absolute percent error. MSE, mean squared error. DETAILED DESCRIPTION OF THE INVENTION

[0030] Introduction The present disclosure provides a method for estimating the proportion of fetal DNA in cfDNA samples obtained from maternal cell-free biological samples.As described herein, the method comprises a digital amplification reaction, comprising evaluating the cfDNA obtained from pregnant subjects, typically human subjects, and using methylation-sensitive restriction enzymes to determine the methylation status of the loci that are differentially methylated in fetal cfDNA versus maternal cfDNA.In particular, the method comprises: The method includes analyzing the methylation status of one or more sites in fetal and maternal cfDNA that are hypermethylated in fetal DNA and hypomethylated in maternal DNA, and detecting the methylation status of one or more sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA. In some embodiments, the method further includes amplifying target sites in total cfDNA, including methylation-insensitive genomic regions of chromosomes that are unlikely to exhibit aneuploidy, which provides the ability to quantify the total concentration of DNA in the sample (i.e., both fetal and maternal). In some embodiments, the method further includes detecting a methylation-insensitive region of the Y chromosome, as desired, for example, to determine the gender of the fetus. In some embodiments, the method further includes assessing hypomethylated sites in both fetal and maternal cfDNA and / or assessing hypermethylated sites in both fetal and maternal cfDNA for use as an internal standard for MSRE digestion. Thus, assessing the levels of these various chromosomal regions with different methylation profiles provides the ability to quantify the proportion of cfDNA from a maternal sample that originates from the fetus. Estimation of fetal fraction is important for quality check in noninvasive prenatal testing and for prediction of fetal aneuploidy.

[0031] In some embodiments, the fetal fraction calculated / determined according to the methods of the present disclosure may be used in methods for assessing fetal aneuploidy (e.g., trisomy such as chromosome 13, 18, or 21), chromosomal deletion (e.g., microdeletion such as chromosome 22), or other genomic alteration (e.g., genetic mutations associated with diseases such as alpha- or beta-thalassemia, cystic fibrosis, sickle cell anemia, or hemophilia A). For example, the fetal fraction calculated / determined according to the methods of the present disclosure may be used for quality control in such methods.

[0032] Components In the present disclosure, the cell-free DNA obtained from pregnant subject is evaluated to determine the amount of fetal cfDNA in maternal blood, that is, the proportion of cfDNA in maternal blood that is derived from fetus.The cfDNA sample from pregnant subject is digested with one or more methylation-sensitive restriction enzymes, and then multiple target loci that have different methylation profiles in fetal DNA and maternal DNA are amplified.The proportion of fetal cfDNA can be calculated based on the level of differentially methylated DNA.

[0033] CfDNA The cell-free DNA for use in the present invention is obtained from cell-free biological fluid sample, typically blood sample.Therefore, in typical embodiments, sample is plasma or serum sample.The isolation of cfDNA can be achieved by using many different methods, for example, by using column or magnetic beads, or other isolation methods.The kit for extracting cfDNA from sample is commercially available, for example, from Qiagen, Beckman (for example, Apostle™ kit) and ThermoFisher (for example, MagMax™ kit).

[0034] Methylation-sensitive restriction enzymes cfDNA is subjected to digestion with one or more methylation-sensitive restriction enzymes (MSREs). Such enzymes digest unmethylated regions of DNA but not methylated regions. Thus, amplification products from hypermethylated regions of DNA are more abundant than amplification products from hypomethylated regions of DNA.

[0035] In some embodiments, one restriction enzyme is used. In alternative embodiments, two MSREs are used. In other embodiments, at least three MSREs are used. In other embodiments, at least four MSREs are used. Exemplary restriction enzymes include AatII, AciI, AclI, AfeI, AgeI, AscI, BmgBI, BsaAI, BsaHI, BspDI, ClaI, EagI, FseI, PauI, HhaI, HpaII, HpyCH4IV, HinPII, MluI, NarI, NotI, NruI, PvuI, SacII, and SalI, and SmaI. In some embodiments, one or more of HhaI, HpaII, AciI, and HpyCH4IV are used in the analysis.

[0036] In some embodiments, as detailed below, before dividing reaction mixture into partitions, one or more MSRE digestion is carried out in bulk.However, in a preferred embodiment, cfDNA is added to dPCR reaction mixture together with PCR reagents and one or more restriction enzymes for target amplification site.Then, restriction enzyme digestion can be carried out in partitions but before amplification.

[0037] Additionally, targets in the cfDNA eluate may be pre-amplified after bulk MSRE digestion, e.g., to reduce the amount of assay multiplexing required to achieve sufficient sensitivity and precision.

[0038] Amplification Target Determining fetal fraction typically involves multiplex amplification of each of the hypomethylated or hypermethylated sites targeted in genome being evaluated.Therefore, in some embodiments, at least two sites, or at least three sites, or at least four sites, or more sites are targeted for each category of DNA that can be used in assay, namely, the hypermethylated site in fetal cfDNA, the hypomethylated site in fetal cfDNA, the hypermethylated site in maternal cfDNA, the hypomethylated site in maternal cfDNA, the site from the methylation-insensitive region of the chromosome that is unlikely to show aneuploidy, the site from the methylation-insensitive region of the Y chromosome, the hypomethylated site in both fetal and maternal cfDNA, and the hypermethylated site in both fetal and maternal cfDNA.

[0039] The differentially methylated sites in maternal compared to fetal cfDNA have been described (see, for example, Ionnides, Mol. Genet. Genomic Med. 8:e1094, 2020; Hatte et al., PLOS ONE DOI:10.1371 / journal.pone.012891, 2015; Bunce et al., Prenat. Diagn. 32:542-54, 2012; Xiang et al., Mol Hum Reprod 20:875-884, 2014). Also see Hatt et al., PLOS ONE, July 31, 2015, pages 1-12; DoI:10.1371 / journal.pone.0128918. Differentially methylated sites can also be determined. For example, methylation sequencing can be used to identify hypermethylated versus hypomethylated sequences in a panel of fetal, maternal, and non-pregnant samples. Exemplary target sites are listed in Table 6, and one, some, or all of the target sites listed in Table 6 can be used, optionally along with other target sites not listed in Table 6.

[0040] The site from a chromosome unlikely to exhibit aneuploidy may be from an autosome other than chromosome 21, chromosome 13, or chromosome 18. In some embodiments, the site is from chromosome 3. Methylation-insensitive regions refer to regions of a chromosome that do not contain CpG sites, and are therefore unlikely to be methylated in any cell, and lack a recognition sequence for the MSRE used in the method, meaning that these sites will not be cleaved by the MSRE even if they are not methylated.

[0041] Sites derived from the methylation-insensitive region of the Y chromosome refer to methylation-insensitive sequences unique to the Y chromosome, the detection of which indicates the presence of a Y chromosome, i.e., a male fetus.

[0042] For the purpose of this application, a differentially methylated site is a site where the methylation pattern between fetal DNA and maternal DNA is statistically different by two-sample Kolmogorov-Smirnov test (p<0.015).Furthermore, for selecting the methylated site in fetal cfDNA, the site with an average methylation frequency of more than 80% in fetal cfDNA (i.e., hypermethylated site) and the site with an average methylation frequency of less than 20% in maternal DNA (i.e., hypomethylated site) are selected.Similarly, for selecting the methylated site in maternal cfDNA, the site with an average methylation frequency of more than 80% in maternal cfDNA (i.e., hypermethylated site) and the site with an average methylation frequency of less than 20% in fetal DNA (i.e., hypomethylated site) are selected.

[0043] When selecting hypermethylated target sites as reference sites, sites with an average methylation frequency greater than 80% in both fetal and maternal cfDNA are selected.

[0044] When selecting hypomethylated target sites as reference sites, sites with an average methylation frequency of less than 20% in both fetal and maternal cfDNA are selected.

[0045] Illustratively, the results of such selection based on analysis of the difference between fetal and maternal methylation for the selection of fetal hypermethylated sites, maternal hypermethylated sites, reference hypermethylated sites, and reference hypomethylated sites are provided in Figures 2A-2D, respectively. In some embodiments, the target sites in Table 6 are assayed according to the methods described herein.

[0046] Primers and probes Primer and probe sequences for detecting amplification products of desired targets can be designed based on known principles. The amplification products are detected with detectable labels. Those skilled in the art will understand that there are many label configurations for detecting amplification products. In some embodiments, oligonucleotides are labeled with detectable agents such as fluorescent agents, phosphorescent agents, and chemiluminescent agents.

[0047] In some embodiments, the probe is labeled, for example, with a fluorescent label.In alternative embodiments, at least one of a pair of amplification primers is labeled with a detectable label, for example, a fluorescent label.In some embodiments, the complementary oligonucleotide that hybridizes with the non-target region of the primer or probe is labeled with a detectable label, for example, a fluorescent label.

[0048] In some embodiments, the probe is a TAQMAN™ probe, a SCORPION™ probe, an ECLIPSE™ probe, a molecular beacon probe, a double-stranded probe, a dual hybridization probe, or a double-quencher probe.

[0049] In some embodiments, the oligonucleotide, e.g., a primer or a probe, is labeled with a detectable label, e.g., a fluorescent label. In some embodiments, the agent is a fluorophore. Numerous fluorophores are available, including those from commercial vendors. Non-limiting examples of fluorophores include cyanines (e.g., Cy3, Cy5), indocarbocyanines (e.g., Quasar® 570, Quasar® 670, and Quasar® 705), fluoresceins (e.g., 5'-carboxyfluorescein (FAM), 6-carboxyfluorescein (6-FAM), 5- and 6-carboxyfluorescein (5,6-FAM), 2'-chloro-7'phenyl-1,4-dichloro- -6-carboxy-fluorescein (VIC), 6-carboxy-4'-,5'-dichloro-2'-,7'-dimethoxy-fluorescein (JOE), 4,7,2',4',5',7'-hexachloro-6-carboxy-fluorescein (HEX), 4,7,2',7'-tetrachloro-6-carboxy-fluorescein (TET), 2'-chloro-5'-fluoro-7',8'-benzo-1,4-dichloro-6-carboxyfluorescein (NED), Oregon Green, and Alexa 488), rhodamine (e.g., N,N,N',N'-tetramethyl-6-carboxyrhodamine (TAMRA) and 5- and 6-carboxy-X-rhodamine (ROX)), tetramethylrhodamine, and tetramethylrhodamine isothiocyanate (TRITC)), Atto dye, eosin, coumarin, pyrene, tetrapyrrole, arylmethine, and oxazine. In some embodiments, the dye is selected from Cy3, Cy5, Cy4.4, ROX, Atto, FAM, HEX, JOE, QUASAR, rhodamine, TAMRA, TET, Texas Red, TYE, and VIC.

[0050] In some embodiments, detection of the amplification product is carried out via a reporter-quencher pair. The reporter-quencher pair can be selected from xanthene dyes, including fluorescein and rhodamine dyes. Many suitable forms of these compounds are commercially available, with substituents on the phenyl group that can be used as sites for conjugation or as linking functionality for attachment to oligonucleotides. Another group of fluorescent compounds for use as reporters are naphthylamines, which have an amino group in the alpha or beta position. Among such naphthylamino compounds are 1-dimethylaminonaphthyl-5-sulfonate, 1-anilino-8-naphthalenesulfonate, and 2-p-toluidinyl-6-naphthalenesulfonate. Other dyes include 3-phenyl-7-isocyanatocoumarin; acridines such as 9-isothiocyanatoacridine; N-(p-(2-benzoxazolyl)phenyl)maleimide; benzoxadiazole; stilbene; pyrene; and the like.

[0051] Suitable examples of quenchers can be selected from 6-carboxy-tetramethyl-rhodamine, 4-(4-dimethylaminophenylazo)benzoic acid (DABYL), tetramethylrhodamine (TAMRA), BHQ-0™, BHQ-1™, BHQ-2™, and BHQ-3™ (each of which is available from Biosearch Technologies, Inc., Novato, Calif.), Qy7™, QSY-9™, QSY-21™, and QSY-35™ (each of which is available from Molecular Probes, Inc.), and ZEN™ and TAO™ double quencher probes (available from Integrated DNA Technologies). Fluorescent and dark quenchers from which exemplary reporter-quencher pairs can be selected and their associated optical properties are listed and described, for example, in R.W. Sabnis, HANDBOOK OF FLUORESCENT DYES AND PROBES, John Wiley and Sons, New Jersey, 2015.

[0052] Primers can be designed taking into account the recognition sequences of one or more MSREs used in the reaction. Thus, for example, the primers and target region to be amplified are selected to avoid the presence of one or more MSRE recognition sequences in the primers and / or the amplicons generated during the amplification reaction.

[0053] In some embodiments, multiple amplicons can be detected with a probe of the same color. For example, in some embodiments, 2 to 20, for example, 6 to 10, amplicons are detected with a single-color probe. In some embodiments, multiple color signals are used. For example, in some embodiments, one color signal is used to detect regions that are hypermethylated in fetal cfDNA compared to maternal cfDNA, a second color signal is used to detect regions that are hypermethylated in maternal cfDNA compared to fetal cfDNA, and optionally, a third color is used to detect Y chromosome-specific sequences (if present). Additional colors can be used to detect regions that are hypermethylated in both fetal and maternal cfDNA, regions that are hypomethylated in both fetal and maternal cfDNA, or controls to measure whether the assay is working.

[0054] Components for additional amplification reactions The reagent mixture can further include additional reagents, such as amplification reagents (e.g., including one or more of buffers, salts, nucleotides, stabilizers, primers, polymerase, or nuclease-free water). In some embodiments, additives, such as tetramethylammonium chloride (TMAC), DMSO, DTT, or betaine, can be used to improve amplification specificity or yield.

[0055] distribution The reaction mixture, for example, cfDNA, digital amplification reaction components, and methylation-sensitive restriction enzyme, can be distributed into partitions by any available method.In some embodiments, the method and composition for delivering reagents to one or more partitions comprises microfluidic methods using microwell plates, capillaries, oil emulsions, and arrays of small chambers for distribution.In some embodiments, distribution is performed by droplets.Methods for generating such droplets include droplet or microcapsule merging, coalescence, fusion, rupture, or decomposition (for example, those described in US Patent Application Publication No. 2015 / 0027,892, US Patent Application Publication No. 2014 / 0227,684, WO 2012 / 149,042, and WO 2014 / 028,537), droplet ejection methods (for example, those described in WO 2010 / 151,776), and combinations thereof. Thus, for example, in the method in which partitions are droplets, droplets can be formed as emulsions with immiscible fluids such as oil, so that bulk solution forms droplets containing reaction mixture reagents containing cfDNA template.Methods for emulsion formation are described, for example, in International Patent Publication Nos. 2011 / 109546 and 2012 / 061444.

[0056] In some embodiments, the amplification reaction is droplet digital PCR reaction.Methods for carrying out PCR in droplets are described in, for example, US Patent Application Publication Nos. 2014 / 0162266, 2014 / 0302503 and 2015 / 0031034, the contents of each of which are incorporated by reference.In some embodiments, QX600 Droplet Digital PCR (ddPCR) system (Bio-Rad) is used.

[0057] In some embodiments, the detection reagent or detectable label in the partition can be detected using any of a variety of detection devices. Exemplary detection methods include optical detection (e.g., fluorescence or chemiluminescence). As a non-limiting example, a fluorescent label can be detected using a detection device that includes a module for generating excitation light that can be absorbed by the fluorophore, as well as a module for detecting the light emitted by the fluorophore.

[0058] In some embodiments, the detector further includes a dispensed sample (e.g., droplet) handling capability, where individual dispensed samples enter the detector, undergo detection, and then exit the detector. In some embodiments, dispensed samples (e.g., droplets) can be continuously detected while the dispensed samples are flowing. In some embodiments, dispensed samples (e.g., droplets) are arrayed on a surface, and a detector moves relative to the surface to detect a signal(s) at each location containing a single partition. Examples of detectors are provided in International Publication No. 2010 / 036352, the contents of which are incorporated herein by reference. In some embodiments, detectable labels in dispensed samples can be continuously detected without flowing the dispensed samples (e.g., using a chamber slide).

[0059] After obtaining the fluorescence detection data, a general-purpose computer system (referred to herein as a "host computer") can be used to store and process the data. Computer-executable logic can be used to perform functions such as subtracting background signals, assigning target and / or reference sequences, and quantifying the data. The host computer can be useful for displaying, storing, retrieving, or calculating the fetal fraction in a sample, storing, retrieving, or calculating raw data from nucleic acid detection, or displaying, storing, retrieving, or calculating any sample or source information useful in the method.

[0060] Host computers can be configured with many different hardware components and can be produced in many sizes and styles (e.g., desktop PCs, laptops, tablet PCs, handheld computers, servers, workstations, mainframes). Standard components such as monitors, keyboards, disk drives, CD and / or DVD drives can be included. If the host computer is connected to a network, connectivity can be provided via any suitable transport medium (e.g., wired, optical, and / or wireless) and any suitable communications protocol (e.g., TCP / IP), and the host computer can include appropriate network hardware (e.g., modems, Ethernet cards, WiFi cards). The host computer can implement any of a variety of operating systems, including UNIX, Linux, Microsoft Windows, MacOS, or any other operating system.

[0061] Computer code for carrying out aspects of the present invention can be written in a variety of languages, including PERL, C, C++, Java, JavaScript, VBScript, AWK, or any other scripting or programming language that can be executed on a host computer or compiled to run on a host computer. Code can also be written or delivered in a lower-level language, such as assembler language or machine language.

[0062] Scripts or programs incorporating various features of the present invention may be encoded on various computer-readable media for storage and / or transmission. Examples of suitable media include magnetic disks or tapes, optical storage media such as compact disks (CDs) or digital versatile disks (DVDs), flash memory, and carrier signals suitable for transmission over wired, optical, and / or wireless networks conforming to various protocols, including the Internet.

[0063] Calculation of fetal fraction Data processing can be used to obtain the detected copy number concentrations of fetal-specific or maternal-specific target sites to calculate the fetal fraction and, optionally, determine fetal sex (if desired). Table 3 shows the various copy number concentrations provided by six exemplary assay collections across six exemplary fluorescent channels. As described herein, copy number concentrations from one or more fetal-specific or maternal-specific target sites, as well as various control target sites, can be determined within partitions from a sample. Thus, for example, the copy number concentration of fetal cfDNA can be determined from one or more target sites that are specifically hypermethylated in fetal cells, hypomethylated in maternal cells, cleaved by one or more MSREs, and detected, for example, by probes in a digital assay. The signals of one or more target sites can be multiplexed, so that, for example, probes detecting different fetal hypermethylated target sites have probes of the same color, and the sum of the partitions with signals of that color divided by the number of targets indicates the fetal copy number concentration.

[0064] In some embodiments, prior to any downstream calculations, each copy number concentration (e.g., each of the six copy number concentrations) is normalized by dividing by the number of assays in the associated multiplex. The number of partitions positive for a particular signal (e.g., detectable wavelength, or "color") may represent amplicons from multiple targets (each detected with a probe of the same color). In these embodiments, N i represents the number of multiple targets detected in the amplification set (and, for example, detected with probes having the same color label). In an exemplary embodiment, the average corrected concentrations of fetal and maternal cfDNA can be determined by interpolation within the reference corrections for hypermethylation and hypomethylation, as shown in Equations 1 and 2.

number

[0065] Furthermore, several relationships exist between independently measured metrics. Therefore, adherence to the relationships expressed in Equations 3-5 can be used as a measure of data quality.

number

[0066] A male fetus can be determined by verifying that the copy number concentration of the Y chromosome is not zero and satisfies the relationship shown in Equation 5. Furthermore, in the case of Y chromosome sex aneuploidy, the coefficient will be 1 instead of 2.

[0067] For either female or male fetuses, the fetal fraction can be calculated by four methods, such as those shown in Equations 6-9, which may be averaged to obtain more accurate values. Furthermore, failure of the fetal fraction calculation to converge to the average value indicates problems with data quality.

number

[0068] For male fetuses, the inclusion of data on Y chromosome copy number concentration allows the calculation of two additional fetal fractions (Equations 10 and 11).

number

[0069] Model for estimating fetal fraction To improve the accuracy of the calculated fetal fraction, a model has been developed that takes the calculated fetal fraction as input and outputs an adjusted fetal fraction based on the model. The model is developed with the goal of minimizing the mean error of the calculated fetal fraction from clinical samples using next-generation sequencing (NGS) measurements of the fetal fraction of the same clinical samples. In one embodiment, the model is designed to reduce the mean absolute percentage error or mean squared error.

[0070] For example, consider that x1, x2, x3, ..., xN represent calculated fetal fractions, and z1, z2, z3, ..., zN are corresponding NGS measurements, e.g., z1 corresponds to x1, z2 corresponds to x2, and so on. Furthermore, consider that the estimation model is represented as a function f( ) that incorporates the calculated values ​​x1, x2, ..., xN, and calculates, for each calculated fraction, an estimated fetal fraction y1, y2, y3, ..., yN (e.g., y1 = f(x1), y2 = f(x2), ..., yN = f(xN)). The error of each estimate can be defined as e1 = (y1 - z1) = (f(x1) - z1), e2 = (y2 - z2) = (f(x2) - z2), ..., eN = (yN - zN) = (f(xN) - zN). In defining the estimation or mapping function f(), several criteria may be considered. For example, in minimizing the mean squared error (MSE), the function f() may be determined by minimizing the MSE, e.g., (e1 2 +e2 2 +...+eN 2) / N is minimized. In another embodiment, f() may be designed to minimize the mean absolute percentage error (MAPE), e.g., (|e1| / z1 + |e2| / z2 + ... + |eN| / zN) / N, where |.| is the absolute value operation.

[0071] In one example, once a model is designed using clinical or training data, the model can be used to estimate fetal fraction based on calculated fetal fractions, hi another example, the model may be dynamic and can be updated using additional training data.

[0072] In one example, the function f(·) is a linear function, e.g., f(x) = ax + b. In another example, randomized noise can be included in the model (e.g., f(x) = ax + b + n, where n is a random variable with a specified distribution).

[0073] Figure 5A shows the impact of applying a linear model to the calculated fetal fraction and its effect on the MSE and MAPE. 501 shows the calculated fetal fraction relative to the corresponding NGS fetal fraction. The calculated fetal fraction data showed a MAPE of 28% and a MAPE of 6.9 × 10. -4 In 503, the same data as in 501 are presented using a box plot. In 502, a linear model is applied to the calculated fetal fraction to obtain the estimated fetal fraction shown. As the figure shows, the estimates are more aligned with the corresponding NGS values, which is also confirmed by the reduced MSE and MAPE values. Using a linear model, the MAPE is 19% and the MSE is 2.8 × 10 -4 In 504, the same data as in 502 is presented using a box plot. Comparing the box plots in 504 and 503 shows that the data has less variability and the estimated fetal fraction is closer to the measured NGS value compared to the calculated fetal fraction.

[0074] In one embodiment, the function f() can be a polynomial, for example, f(x) = a0 + a1x + a2x 2 + a3x 3 + … + aMx M where M determines the degree of the polynomial. For example, for a quadratic polynomial, M=2, and f(x) = a0 + a1x + a2x 2 The coefficients a0, a1, and a2 can be calculated by applying the model to clinical data using an optimization criterion (e.g., minimizing the MSE or MAPE).

[0075] FIG. 5B shows a residual plot. At 510, the calculated fetal fraction is shown against the variable fetal / hypermethylated. At 512, the calculated fetal fraction is shown against the variable maternal / hypermethylated compared to the residual plot of fetal / hypermethylated. This plot does not suggest a linear relationship between fetal / hypermethylated and the calculated fetal fraction.

[0076] Figure 5C shows the impact of applying a linear model to the calculated fetal fraction and its effect on the MSE and MAPE. Using the same calculated fetal fraction data as in 501, a second-order polynomial model is developed. A second-order polynomial model (poly2LM) is applied to the calculated fetal fraction to obtain the estimated fetal fraction shown. Figure 5C shows that the estimates are more aligned with the corresponding NGS values ​​along the line, which is also confirmed by the reduced MSE and MAPE values. Using the second-order polynomial model, the MAPE is reduced to 17% and the MSE is reduced to 2.5 x 10 -4 decreases to.

[0077] In one example, a generalized additive model (GAM) can be used to estimate the fetal fraction based on the calculated fetal fraction. A GAM can be used to model nonlinearities using an additive model. For example, at various NGS values, there may be different relationships between the calculated fetal fraction and the corresponding NGS value. Piecewise models defining different estimation functions may be developed. For example, the NGS range can be divided into K regions (e.g., K = 20, 30, or 40), and each region is estimated using a spline, e.g., a 2-, 3-, 4-, or different-order spline. Furthermore, a GAM can incorporate other parameters, such as gestational age, maternal weight, etc., as model variables.

[0078] Figure 5D shows the effect of applying GAM to the calculated fetal fraction, as well as its effect on MSE and MAPE. The same calculated fetal fraction data as in 501 is used. GAM is applied to the calculated fetal fraction to obtain the estimated fetal fraction shown. Figure 5D shows that the estimated values ​​are more aligned with the corresponding NGS values ​​along the line, which is also confirmed by the reduced MSE and MAPE values. Using GAM, the MAPE is reduced to 13% and the MSE is reduced to 1.5 x 10 -4 decreases to.

[0079] kit In a further aspect, the present disclosure provides a digital amplification kit for estimating the proportion of fetal DNA in a cfDNA sample obtained from a blood sample (such as a plasma or serum sample) from a pregnant subject, for example, a human. The kit may include any of the components described herein with respect to the method. In some embodiments, the kit includes an amplification reaction mixture including an amplification reagent and a plurality of amplification sets including primer and probe sets, wherein each amplification set includes a distinct label that is distinguishable from the labels of each of the other sets, and each set includes primers and probes for multiplex amplification, and the plurality of amplification sets include (i) an amplification set that targets sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA, and / or (ii) an amplification set that targets sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA. In some embodiments, the kit further includes an amplification set that targets total cfDNA including methylation-insensitive regions of chromosomes that are unlikely to exhibit aneuploidy. In some embodiments, the kit includes an amplification set targeting hypermethylated sites in fetal and maternal DNA, and / or an amplification set targeting hypomethylated sites in fetal and maternal DNA. In some embodiments, the kit further includes an amplification set targeting a methylation-insensitive region of the Y chromosome. In some embodiments, the one or more sites are selected from Table 6 described elsewhere herein. In some embodiments, the kit described in this paragraph includes a fully methylated synthetic sequence of DNA and / or a fully unmethylated version of the same synthetic sequence. In some embodiments, the probe used for detection is a molecular beacon probe, e.g., a fluorescently labeled molecular beacon probe.

[0080] In some embodiments, each of the (i) and (ii) amplification sets comprises primers and probes for targeting at least two sites, at least three sites, or more, e.g., 1 to 20 sites, e.g., or 6 to 10 sites.

[0081] In some embodiments, each label of the amplification set is a fluorescent label. In some embodiments, one or more primers or probes contain a region that does not hybridize to the target amplification site but is complementary to an oligonucleotide that provides a detectable signal.

[0082] In some embodiments, a kit comprising one or more amplification sets described in the preceding paragraph further comprises at least one methylation-sensitive restriction enzyme that cleaves hypomethylated DNA. In some embodiments, the kit comprises one or more of the restriction enzymes HhaI, HpaII, AciI, HpyCH4IV, or BsaHI. In some embodiments, a methylation-sensitive restriction enzyme (MSRE) cocktail comprises at least one, two, at least three, or at least four methylation-sensitive restriction enzymes selected from HhaI, HpaII, AciI, HpyCH4IV, and BsaHI. In some embodiments, the kit comprises a methylation-sensitive restriction enzyme (MSRE) cocktail comprising at least two, three, or more of the restriction enzymes HhaI, HpaII, AciI, HpyCH4IV, and BsaHI. In some embodiments, the MSRE cocktail comprises at least HhaI and HpyCH4IV.

[0083] In some embodiments, the reaction mixture is lyophilized.

[0084] In some embodiments, the kit includes a standard ddPCR kit, a set of primers and probes for a fetal fraction assay described herein, and at least one MSRE. In some embodiments, such kits further include at least one PCR enhancer, such as TMAC and / or a salt.

[0085] In some embodiments, the kit includes a standard ddPCR kit, a set of primers and probes for the fetal fraction assay described herein, at least one MSRE, and a stabilizer (e.g., trehalose, potassium glutamate, ammonium sulfate) (lyophilized together in one mixture).

[0086] In some embodiments, the kit includes a standard ddPCR kit, a set of primers and probes for the fetal fraction assay described herein, a stabilizing reagent (lyophilized together), and at least one MSRE (which may or may not be lyophilized). [Example]

[0087] The following examples further illustrate aspects and embodiments of the methods of the present disclosure.

[0088] Example 1: Methylation-based digital PCR assay for quantification of fetal cell-free DNA in NIPT samples. First, cfDNA is isolated from maternal plasma using a commercially available kit, such as a magnetic bead kit designed for the preferential capture and elution of cfDNA (e.g., Apostle Magnetic Bead-Based Extraction Kit). The fragment length distribution of the cfDNA eluate can be confirmed by commercially available methods, such as a high-sensitivity DNA bioanalyzer kit (Agilent).

[0089] CfDNA isolated from the maternal sample is digested with a methylation-sensitive restriction enzyme (MSRE) cocktail, including enzymes such as, but not limited to, HhaI, HpaII, AciI, or HpyCH4IV. These enzymes cleave DNA at sites where multiple bases are present in a specific sequence, but only if the site is unmethylated (Table 1). For differentially methylated sites (DMS) that are hypermethylated in fetal cfDNA and hypomethylated in maternal cfDNA, only the DMS in fetal cfDNA remains to be quantified. Conversely, for DMS that are hypermethylated in maternal cfDNA and hypomethylated in fetal cfDNA, only the DMS in maternal cfDNA remains to be quantified. While in some embodiments, digestion of cfDNA can be performed in bulk, in this exemplary embodiment, digestion is performed after distribution of the sample and PCR reagents but before PCR amplification.

[0090] cfDNA (e.g., eluate from a sample) is added to a dPCR reaction mixture that includes a dPCR supermix (containing all reagents necessary for both PCR and partitioning), an MSRE cocktail, and, in this example, a 6-channel assay set (Table 1). This intrapartition digestion technique allows for a more efficient procedure and a single thermocycling run (Table 2).

[0091] The methylation-sensitive digestion strategy described here is further exemplified below using a four-channel droplet digital PCR (ddPCR) instrument (QX ONE).

[0092] Twenty-two maternal plasma samples (CerbaXpert, France) with fetal fractions ranging from 10 to 25% as measured by VeriSeq NIPT were selected, along with two non-pregnant plasma samples (Stanford Blood Bank). cfDNA was extracted from 24 samples using the Apostle MiniMax High Efficiency cfDNA Isolation Kit, performed automatically on a KingFisher Flex (ThermoFisher). Prior to use in ddPCR, the cfDNA was characterized on a Bioanalyzer platform (Agilent) and found to contain 62% ± 6% mononucleosomal cfDNA, with the remaining nucleic acid fraction consisting of high-molecular-weight genomic DNA.

[0093] For each sample, 5.5 μL of the extracted eluate was used in six 22 μL ddPCR reactions (Table 4). All reactions contained a fetal DMS triplex (FAM channel) targeting sites hypermethylated in fetal cfDNA and hypomethylated in maternal cfDNA, a maternal DMS triplex (HEX channel) targeting sites hypermethylated in maternal cfDNA and hypomethylated in fetal cfDNA, and a Y chromosome target (SRY, Cy5.5 channel). In the remaining Cy5 channel, three reactions contained an X chromosome target (SPIN4), and three reactions contained a chromosome 10 target (RPP30). In addition, four reactions contained an MSRE cocktail (HhaI and HpyCH4IV, each at 10 U per ddPCR reaction), while two reactions did not.

[0094] Despite using uncorrected values, fetal and maternal triplexes were used to calculate the fetal fraction as shown in Equations 6-11. Copy number concentrations of SRY, SPIN4, and RPP30 were used to calculate orthogonal ddPCR fetal fraction estimates that do not rely on a methylation-sensitive digestion scheme.

[0095] result Figure 1 shows the correlation between methylation-sensitive and SRY-based fetal fraction estimates for 24 samples. When both SRY / SPIN4 and SRY / RPP30 calculations were included, the R 2 The R of the correlation between the methylation-sensitive fetal fraction estimate and the VeriSeq noninvasive prenatal testing (NIPT) determination was 0.88. 2 is 0.84 (Table 5).

[0096] Use with other fetal diagnostic assays With sufficient multiplexing, this fetal fraction quantification method can be used simultaneously with other fetal cfDNA tests, including trisomy and microdeletion detection. In some embodiments, because the method uses only the cfDNA portion obtained from the maternal sample, the assay can be performed in parallel with these tests in separate wells using the same sample. In some embodiments, when multi-well testing is desired, performing MSRE digestion in droplets for fetal fraction determination simplifies the workflow. In this case, cfDNA eluate is added to each test well, and the presence of MSRE in the fetal fraction test reaction mixture digests only the cfDNA used to determine the fetal fraction. In some embodiments, the assay can be performed in parallel with diagnostic tests for monogenic diseases, such as, but not limited to, alpha- or beta-thalassemia, cystic fibrosis, or hemophilia A.

[0097] Example 2: Highly multiplexed methylation-based ddPCR assay and machine learning method for determination of fetal cell-free DNA fraction in NIPT samples. In this example, a highly multiplexed methylation-based ddPCR assay was utilized in conjunction with a machine learning-based method to analyze fetal cfDNA fraction in NIPT samples.

[0098] Fetal fraction (FF) refers to the fetal cfDNA fraction in the blood of a pregnant mother. Estimating fetal fraction can be used, for example, to assess fetal aneuploidy noninvasively. Currently, next-generation sequencing (NGS) serves as the gold standard method for estimating fetal fraction. Approaches include profiling single-nucleotide polymorphisms to analyze differential genotypes between the fetus and the mother, measuring the proportion of Y-chromosome cfDNA reads in male fetuses, and examining differences in methylation profiles. However, sequencing-based approaches are not cost-effective. Digital droplet PCR (ddPCR), which requires distributing individual PCR reactions into droplets, allows for high levels of sensitivity and accuracy as well as reduced costs compared to NGS.

[0099] In Example 1, a methylation-based ddPCR assay was designed based on the differential methylation patterns between maternal and fetal cfDNA in NIPT samples to calculate the fetal fraction. This example describes a highly multiplexed methylation-based ddPCR assay for simultaneously quantifying fetal and maternal cfDNA in the same ddPCR reaction, which improves the accuracy of quantifying the fetal fraction in NIPT samples.

[0100] ddPCR assay design and machine learning methods The ddPCR assay used six fluorescent channels (QX600 instrument, Bio-Rad Laboratories) with 6–10 targets per channel targeting the following: (i) sites that are hypermethylated in fetal cfDNA and hypomethylated in maternal cfDNA ("fetal") for targeting fetal cfDNA after digestion with a methylation-sensitive restriction enzyme; (ii) sites that are hypermethylated in maternal cfDNA and hypomethylated in fetal cfDNA ("maternal"), for targeting maternal cfDNA after digestion with a methylation-sensitive restriction enzyme; (iii) sites that are hypomethylated in fetal cfDNA and maternal cfDNA ("hypomethylation") to target the digestion efficiency of methylation-sensitive enzymes in conjunction with "hypermethylation" as described in (iv); (iv) sites that are hypermethylated in fetal and maternal cfDNA ("hypermethylation") to target for digestion by methylation-sensitive enzymes along with "hypomethylation"; (v) sites on the Y chromosome ("Chr Y") that do not contain CpG sites (i.e., methylation-insensitive regions) in male fetuses; and (vi) A control site containing no CpG sites ("Control") for assay performance control.

[0101] Estimation of fetal cfDNA fractions was performed using fetal / hypermethylated, maternal / hypermethylated, hypomethylated / hypermethylated, and chrY / control. Fetal / hypermethylated, maternal / hypermethylated, hypomethylated / hypermethylated, and chrY / control were calculated based on the lambda ratio of "fetal" divided by "hypermethylated," the lambda ratio of "maternal" divided by "hypermethylated," and the lambda ratio of "chrY" divided by "control."

[0102] An overview of the assay design is provided in Figures 3 and 4A-4B. Table 6 provides exemplary target sites that can be used to quantify the fetal fraction in this ddPCR assay.

[0103] To further improve the assessment of fetal fraction, we developed machine learning linear regression, polynomial, and generalized additive models trained on clinical NIPT samples. Due to limited clinical NIPT samples, we trained three models using over 70 samples. The fetal / hypermethylated, maternal / hypermethylated, hypomethylated / hypermethylated, and chrY / control associations calculated from the highly multiplexed fetal fraction assay, as described in Table 6, served as input variables for the three models. Other metadata from clinical NIPT samples, such as gestational age, was also tested to improve model performance.

[0104] result As shown in Figure 5A-5D and Table 7, the mean absolute percent error (MAPE) and mean squared error (MSE) of ddPCR-based fetal fraction estimates relative to NGS-based calculations dramatically decreased when using linear regression, polynomial, and generalized additive models compared to traditional analysis methods.

[0105] For example, estimation of fetal fraction using a linear regression model ("LM") trained with clinical samples had smaller errors compared to calculation of fetal fraction using Equations 1-11 ("FF_calculated"), including smaller MAPE and MSE (Figure 5A and Table 7). Figure 5B illustrates the relationship between the fetal / hypermethylated and maternal / hypermethylated variables when fetal fraction was obtained using NGS and a linear model. To further improve the model, residual plots were utilized to decipher the relationship between fetal / hypermethylated, maternal / hypermethylated, and fetal fraction obtained using NGS. The plot in Figure 5B shows the nonlinear relationship between fetal fraction and fetal / hypermethylated compared to the maternal / hypermethylated variable. Using a quadratic term for fetal / hypermethylated in a polynomial model may improve performance measures, such as MAPE and MSE. For example, Figure 5C shows the estimated fetal fraction using a second-order polynomial model ("poly2LM"). The second-order polynomial model had smaller errors compared to the conventional method, including smaller MAPE and MSE (Figures 5A, 5C, and Table 7).

[0106] The generalized additive model (GAM) is a generalized linear model that can model nonlinear data with interpretability. The GAM had smaller errors compared to conventional methods, including smaller MAPE and MSE (see Figure 5D and Table 7).

[0107] The ddPCR assay and fetal fraction calculation method described in this example hold great promise for improving the accuracy and reliability of ddPCR in estimating fetal fraction, which will contribute to aneuploidy NIPT testing by ddPCR, making it a useful tool for prenatal screening and diagnosis.

[0108] It should be understood that the examples and embodiments described herein are for illustrative purposes only and that various modifications or changes will be suggested to those skilled in the art in light thereof and are to be included within the spirit and scope of this application and the appended claims. All publications, patents, and patent applications cited herein are hereby incorporated by reference in their entirety for all purposes. [Table 1] [Table 2] [Table 3] [Table 4] [Table 5] [Table 6] TIFF2025531887000022.tif21789 [Table 7]

Claims

1. 1. A method for estimating the percentage of fetal DNA in a cfDNA sample obtained from a blood sample of a pregnant human subject, comprising: (a) distributing into partitions an amplification reaction mixture comprising cfDNA from the cfDNA sample, amplification reagents, and a plurality of amplification sets comprising primer and probe sets, wherein each amplification set comprises primers and probes for multiplex amplification, and each amplification set produces an amplification product (containing a distinct label distinguishable from labels for each of the other amplification sets, if target is present), and wherein the plurality of amplification sets: (i) an amplification set targeting sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA; (ii) a set of amplifications targeting sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA, and optionally one or more of (iii), (iv), and (v): (iii) an amplification set targeting total cfDNA containing methylation-insensitive regions of chromosomes unlikely to exhibit aneuploidy; (iv) an amplification set targeting sites hypermethylated in fetal and maternal DNA; (v) an amplification set that targets hypomethylated sites in fetal DNA and maternal DNA; (b) incubating the cfDNA with a methylation-sensitive restriction enzyme (MSRE) cocktail comprising at least one methylation-sensitive restriction enzyme that cleaves unmethylated DNA; (c) amplifying a target nucleic acid sequence, if present, within said partition to obtain an amplification product; (d) detecting within the partition a signal from each distinct label of the amplification product; (e) quantifying the signal of each distinct label.

2. 2. The method of claim 1, wherein the plurality of amplification sets comprises (iii) an amplification set targeting total cfDNA comprising a methylation-insensitive region of a chromosome that is unlikely to exhibit aneuploidy.

3. 3. The method of claim 1 or 2, wherein each of the amplification sets of (i) and (ii) comprises primers and probes for targeting at least three sites.

4. 3. The method of claim 1 or 2, wherein each of the amplification sets (i) to (v) comprises primers and probes for targeting at least three sites or 6 to 10 sites.

5. The method of any one of claims 1 to 4, further comprising an amplification set targeting a methylation-insensitive region of the Y chromosome.

6. The method of any one of claims 1 to 5, further comprising an amplification set targeting hypomethylated sites in both fetal and maternal cfDNA, and / or an amplification set targeting hypermethylated sites in both fetal and maternal cfDNA.

7. The method of any one of claims 1 to 6, wherein the amplification reaction mixture further comprises a control target fully methylated synthetic DNA sequence and / or a fully unmethylated version of the same synthetic DNA sequence.

8. The method according to any one of claims 1 to 7, wherein the digital amplification reaction method is a digital PCR method.

9. 9. The method of claim 8, wherein the digital PCR method is a droplet digital PCR method.

10. 10. The method of claim 1, wherein the amplification reaction mixture comprises the MSRE cocktail and the incubating step occurs after the dispensing step and before the amplifying step.

11. 10. The method of claim 1, wherein step (b) is performed before the distributing and the digested cfDNA is added to the amplification reaction mixture.

12. 12. The method of claim 1, wherein the MSRE cocktail comprises at least two, at least three, or at least four methylation-sensitive restriction enzymes, and / or the MSRE cocktail comprises restriction enzymes selected from HhaI, HpaII, AciI, HpyCH4IV, and BsaHI.

13. 13. The method of claim 12, wherein the MSRE cocktail comprises at least two, three, or four of the restriction enzymes HhaI, HpaII, AciI, HpyCH4IV, and BsaHI.

14. The method of claim 12 or claim 13, wherein the MSRE cocktail comprises at least HhaI and HpyCH4IV.

15. The method of any one of claims 1 to 14, wherein the cfDNA sample is obtained from plasma or serum.

16. The method of any one of claims 1 to 15, wherein each label is a fluorescent label.

17. 17. The method of claim 16, wherein the probe is a molecular beacon probe comprising a fluorescent label.

18. The method of any one of claims 1 to 15, wherein each probe is an oligonucleotide that hybridizes to a complementary oligonucleotide that contains a label that provides a detectable signal.

19. Number of targets in the amplification set (N i 10. The method of claim 1, further comprising determining a normalized copy number concentration of each of the targets based on:

20. The corrected concentration of fetal cfDNA (Fet Corr ) and / or the corrected concentration of maternal cfDNA in the cfDNA sample (Mat Corr ), wherein determining the corrected concentration of fetal cfDNA comprises calculating: [Equation 1] where [Total] is the total cfDNA copy number concentration based on signals in partitions from the amplification set targeting total cfDNA that contain methylation-insensitive regions of chromosomes unlikely to exhibit aneuploidy, [Hyper] is a hypermethylated reference copy number concentration based on signals in partitions from the amplification set targeting hypermethylated sites in fetal DNA and maternal DNA, [Hypo] is a hypomethylated reference copy number concentration based on signals in partitions from the amplification set targeting hypomethylated sites in fetal DNA and maternal DNA, and [Fet] is a fetal cfDNA copy number concentration based on signals in partitions from the amplification set targeting sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA, Determining the corrected concentration of maternal cfDNA comprises calculating: [Equation 2] wherein [Mat] is the maternal cfDNA copy number concentration based on the signal in the partition from the amplification set targeting sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA.

21. 10. The method of any one of the preceding claims, further comprising determining a fetal fraction (FF) in the cfDNA sample, wherein determining the fetal fraction comprises at least one of the following calculations (a) to (d): [Equation 3] 。

22. 22. The method of claim 21, further comprising calculating an estimated fetal fraction based at least in part on the fetal fraction in the cfDNA sample and a model.

23. 23. The method of claim 22, wherein the model is a generalized additive model (GAM), a linear model, or a second-order polynomial model based at least in part on a clinical fetal fraction dataset and a corresponding set of fetal fraction measurements using next-generation sequencing (NGS).

24. further comprising determining a fetal fraction of male fetuses in the cfDNA sample, wherein determining the fetal fraction of male fetuses comprises: [Equation 4] wherein [YChr] is the concentration of Y chromosome-specific sequences based on signals in partitions from the amplification set targeting methylation-insensitive regions of the Y chromosome.

25. 25. The method of any one of claims 1 to 24, wherein the sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA are sites within one or more or all of the following ranges: 。

26. 26. The method of any one of claims 1 to 25, wherein the sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA are sites within one or more or all of the following ranges: 。

27. 27. The method of any one of claims 1 to 26, wherein the sites that are hypermethylated in fetal and maternal DNA are sites within one or more or all of the following ranges: 。

28. 28. The method of any one of claims 1 to 27, wherein the sites that are hypomethylated in fetal and maternal DNA are sites within one or more or all of the following ranges: 。

29. The method of any one of claims 1 to 28, wherein the methylation-insensitive region of a chromosome unlikely to exhibit aneuploidy is a site within one or more or all of the following: 。

30. 6. The method of claim 5, wherein the methylation-insensitive region of the Y chromosome is within one or more or all of the following: 。

31. 1. A digital amplification kit for estimating the percentage of fetal DNA in a cfDNA sample obtained from a plasma or serum sample of a pregnant human subject, comprising: (a) an amplification reaction mixture comprising amplification reagents and a plurality of amplification sets comprising primer and probe sets, each amplification set comprising a distinct label distinguishable from the label for each of the other sets, each set comprising primers and probes for multiplex amplification, said plurality of amplification sets comprising: (i) an amplification set targeting sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA; (ii) an amplification set targeting sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA, and optionally one or more of (iii), (iv), and (v); (iii) an amplification set targeting total cfDNA containing methylation-insensitive regions of chromosomes unlikely to exhibit aneuploidy; (iv) an amplification set targeting sites hypermethylated in fetal and maternal DNA; (v) a kit comprising an amplification reaction mixture comprising an amplification set targeting hypomethylated sites in fetal DNA and maternal DNA.

32. 32. The kit of claim 31 , wherein each of the amplification sets of (i) and (ii) comprises primers and probes for targeting at least three sites.

33. The kit of any one of claims 31 to 32, further comprising an amplification set targeting a methylation-insensitive region of the Y chromosome.

34. 34. The kit of any one of claims 31 to 33, further comprising a fully methylated synthetic sequence of DNA and / or a fully unmethylated version of the same synthetic sequence.

35. The kit of any one of claims 31 to 34, wherein each label is a fluorescent label.

36. 36. The kit of claim 35, wherein the probe is a molecular beacon probe comprising a fluorescent label.

37. 35. The kit of any one of claims 31 to 34, wherein each probe is an oligonucleotide that hybridizes to a complementary oligonucleotide that includes a label that provides a detectable signal.

38. 38. The kit of any one of claims 31 to 37, further comprising a methylation-sensitive restriction enzyme (MSRE) cocktail comprising at least one MSRE that cleaves unmethylated DNA.

39. The kit of claim 38, wherein the MSRE cocktail comprises at least two, at least three, or at least four methylation-sensitive restriction enzymes, and / or the MSRE cocktail comprises restriction enzymes selected from HhaI, HpaII, AciI, HpyCH4IV, and BsaHI.

40. 38. The kit of any one of claims 31 to 37, wherein the MSRE cocktail comprises at least two, three, or more of the restriction enzymes HhaI, HpaII, AciI, HpyCH4IV, and BsaHI.

41. The kit of claim 39 or 40, wherein the MSRE cocktail comprises at least HhaI and HpyCH4IV.

42. The kit of any one of claims 31 to 40, wherein the sites that are hypermethylated in fetal DNA and hypomethylated in maternal DNA are sites within one or more or all of the following ranges: 。

43. The kit of any one of claims 31 to 40, wherein the sites that are hypermethylated in maternal DNA and hypomethylated in fetal DNA are sites within one or more or all of the following ranges: 。

44. The kit of any one of claims 31 to 40, wherein the sites hypermethylated in fetal and maternal DNA are sites within one or more or all of the following ranges: 。

45. The kit of any one of claims 31 to 40, wherein the sites that are hypomethylated in fetal DNA and maternal DNA are sites within one or more or all of the following ranges: 。

46. The kit according to any one of claims 31 to 40, wherein the methylation-insensitive region of a chromosome unlikely to exhibit aneuploidy is a site within one or more or all of the following ranges: 。

47. 34. The kit of claim 33, wherein the methylation-insensitive region of the Y chromosome is within one or more or all of the following: 。

Images