Methods and systems for improving accurate identification of fetal genetic conditions in maternal blood

Abstract

A non-invasive method for genotyping a fetus is disclosed, the method comprising analyzing sequencing data of maternal free DNA (cfDNA) and genomic DNA (gDNA) of parents (maternal and optionally parent) from a pair of parents rearing the fetus. Accurate genotype prediction is obtained using a variation detection method and evaluation of cfDNA data collected at several points in time.

Description

Technical Field

[0001] The present disclosure relates to the field of prenatal genetic analysis.

[0002] References:

[0003] Rabinowitz et al. (2019) Genome Res. 29(3), pp. 428-438

[0004] Scotchman et al. (2020) Clin. Chem. 66(1), pp. 53-60

[0005] Zhang et al. (2019) Nat. Med. 25(3), p. 439 Background Art

[0006] Non-invasive prenatal testing (NIPT) is a procedure used to assess the health of an unborn child by determining the risk of the fetus being born with a harmful genetic abnormality. NIPT relies on the presence of cell-free fetal DNA (cffDNA), a small fraction of total cell-free DNA (cfDNA) that circulates in maternal plasma from the first weeks of pregnancy until birth.

[0007] In the NIPT procedure, blood is drawn from the mother, cfDNA is extracted and sequenced, and then used to obtain genetic information about the fetus. Current NIPT procedures are offered in clinics around the world and can detect large genetic aberrations across the entire chromosome range, or very large, specific copy number variations.

[0008] Therefore, NIPT is used to screen for chromosomal abnormalities (e.g., trisomies, subchromosomal deletions, and duplications), but also for single-gene disorders caused by point mutations. Commercially available next-generation sequencing (NGS) panels consist of only up to 30 genes (Zhang et al. (2019)). Furthermore, false-negative results can occur in customized tests and panels (Scotchman et al. (2020)).

[0009] Rabinowitz et al. (2019) describe a different approach to whole-genome NIPT for monogenic conditions that frames this problem as a unique case of variant calling, termed non-invasive prenatal variant calling. Thus, a Bayesian genotyping algorithm exploits information from every read, encompassing every candidate variant, and a machine learning-based fine-tuning step then integrates information from previous validation results. By considering every read, the authors are able to exploit features that distinguish fetal from maternal DNA, such as fragment length. The algorithm was implemented as Hoobari, the first non-invasive fetal variant caller capable of genotyping all fetal locations, including parental loci and indels. However, performance in parental loci and indels is lower than in locations where only one parent is heterozygous (WO2021 / 0340601). Summary of the Invention

[0010] In one of its aspects, the present invention provides a method of genotyping a fetus, comprising:

[0011] a. receiving sequencing data of maternal cell-free DNA (cfDNA) from a parent carrying a fetus, the sequencing data comprising at least two sets of sequencing reads collected at different pregnancy-related time points, wherein one set of sequencing reads is from time point A during pregnancy, and at least another set of sequencing reads is from time point B, wherein time point B is different from time point A and is before or during pregnancy;

[0012] b. receiving sequencing data of genomic DNA (gDNA), the sequencing data comprising a third set of sequencing reads of maternal gDNA and optionally a fourth set of sequencing reads of paternal gDNA of the paired parent of the fetus;

[0013] c. analyzing the first set and the second set to generate a background reference set;

[0014] d. analyzing the second set, the third set, and optionally the fourth set to identify sequence reads comprising: (i) a first set of sites where both parents have two copies of the same allele at a given locus, and (ii) a second set of sites where at least one of the parents has a variant allele;

[0015] e. for each site in the first group, determining a probability that the analyzed cfDNA sequence read originated from the fetus, wherein the determining comprises introducing at least the background reference set into at least one statistical model to obtain a probability value; and

[0016] f. Using the probability values, classify each locus in the second set as having fetal or maternal origin, thereby genotyping the fetus.

[0017] In one embodiment, one or both of the gDNA sequencing data and the cfDNA sequencing data are obtained by a method selected from the group consisting of whole genome sequencing (WGS), whole exome sequencing (WES), next generation sequencing (NGS), targeted sequencing, panel sequencing, gene sequencing, long read genome sequencing, paired-end sequencing, single-end sequencing, and amplicon sequencing.

[0018] In one embodiment, the WGS or WES data is obtained by deep sequencing.

[0019] In one embodiment, determining said probability value is based on at least one sequence alignment map (SAM) parameter.

[0020] In one embodiment, determining the probability value is based on additional data parameters.

[0021] In some embodiments, additional parameters include length, GC content, genetic linkage, and haplotype of the reads.

[0022] In one embodiment, said step (e) as defined above further comprises calculating the total fetal fraction.

[0023] In one embodiment, the method further comprises constructing a fetal size distribution and a maternal size distribution, wherein determining the probability value comprises binning the fetal size distribution and calculating a fetal fraction for each fragment size bin, and calculating, for at least one size at the at least one site and at least one fragment, a probability that the fragment is fetal based on the fetal fraction of the corresponding fragment size bin to which the fragment belongs.

[0024] In one embodiment, determining the probability value comprises applying a statistical model, such as a Bayesian procedure.

[0025] In one embodiment, the method further comprises recalibrating the output of the Bayesian procedure using machine learning.

[0026] In one embodiment, the background reference set comprises one or both of: (i) maternal cfDNA sequence reads prior to pregnancy, and (ii) maternal cfDNA sequence reads sampled at one or more time points during the pregnancy.

[0027] In one embodiment, the background reference set comprises a time-dependent modulator.

[0028] In some embodiments, the time-dependent modulator is based on data parameters including length, GC content, genetic linkage, and haplotype of the reads.

[0029] In one embodiment, the probability value is determined using the Hoobari algorithm.

[0030] In one embodiment, the maternal cfDNA is sampled at least at two different time points such that cfDNA sequencing data reads are received from at least one time point before the pregnancy and one or more time points during the pregnancy with the fetus.

[0031] In one embodiment, maternal cfDNA is sampled at two or more different time points during said pregnancy.

[0032] In one embodiment, the two or more samples are at time points that are at least one day apart from each other.

[0033] In one embodiment, the maternal cfDNA is sampled early in the pregnancy.

[0034] In one embodiment, the maternal cfDNA is sampled during the first trimester of the pregnancy.

[0035] In another embodiment, said maternal cfDNA is sampled during the second trimester and / or the third trimester.

[0036] In another aspect, the present invention provides a computer software product comprising a computer-readable medium having program instructions stored therein, which, when read by a data processor, configure the data processor to: (1) receive sequencing data reads of (i) maternal cell-free DNA (cfDNA) comprising at least two sets of sequencing reads collected at different pregnancy-related time points, and (ii) maternal and optionally paternal genomic DNA (gDNA) from a pair of parents raising a fetus, and (2) perform a method of the present invention.

[0037] In another aspect, the present invention provides a system for genotyping a fetus, comprising: an input utility for receiving (i) sequencing data reads of maternal cell-free DNA (cfDNA) comprising at least two sets of sequencing reads collected at different pregnancy-related time points, and (ii) sequencing data reads of maternal and optionally paternal genomic DNA (gDNA) from a pair of parents raising a fetus; and a data processor configured to analyze the data to perform the method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to better understand the subject matter disclosed herein and to illustrate how it may be carried out in practice, embodiments will now be described, by way of non-limiting example only, with reference to the accompanying drawings, in which:

[0039] Figure 1 is a flow chart of a method suitable for fetal genotyping according to various exemplary embodiments of the present invention. DETAILED DESCRIPTION

[0040] The present invention relates to a non-invasive method for genotyping a fetus based on analysis of cfDNA in maternal plasma. The invention is based on the surprising discovery that sequence analysis of cfDNA in maternal plasma at several time points before and / or during pregnancy generates a genetic "fingerprint" or profile that facilitates determination of the source of the cfDNA, i.e., whether it originated from the mother or the fetus, with significantly improved accuracy.

[0041] According to the present invention, at least one of the maternal plasma samples is collected during pregnancy (preferably in the early stages of pregnancy), and at least one more maternal plasma sample is collected before or at different time points during pregnancy. Sequence analysis is performed using deep whole genome sequencing. Multiplex analysis of maternal cfDNA improves the accuracy of genotype prediction and can identify various genetic variations in the fetal genome, including single nucleotide mutations.

[0042] Thus, the present invention is directed to improving the accuracy of genetic predictions by providing more substantial information about the characteristics of cfDNA in maternal plasma, thereby enabling improved separation of the fetal cfDNA fraction from total maternal cfDNA.

[0043] In an embodiment, the method of the present invention includes obtaining maternal cfDNA sequencing data at two or more time points before and / or during pregnancy. Such sequencing data can be obtained from maternal blood samples (including samples collected in the past) as a reference for maternal cfDNA characteristics before and during pregnancy. An important step in classifying identified genetic variations (i.e., mutations) as fetal or maternal is the process of distinguishing between maternal and fetal sequencing data.

[0044] According to the present invention, cfDNA sequencing data obtained at different time points allows for the generation of a more accurate maternal cfDNA profile, thereby increasing the strength of the maternal:fetal cfDNA differential signal in the statistical model, enabling better probabilistic estimation of the origin of deep sequencing reads, and therefore, achieving higher accuracy in predicting fetal genotyping.

[0045] According to the present invention, integration of profile information of maternal cfDNA prior to pregnancy and / or changes that occur in cfDNA during pregnancy enhances the ability to determine whether a particular read is of maternal or fetal origin, thereby confirming fetal genotype predictions.

[0046] cfDNA sequencing data obtained during pregnancy can be obtained early in pregnancy (e.g., in the first trimester) so that intervention procedures can be performed if necessary. However, in some cases, for example, to diagnose suspected brain morphological abnormalities, cfDNA sequencing data can also be obtained later in pregnancy, during the second trimester, and / or the third trimester. Therefore, according to the present invention, cfDNA sequencing data can be obtained at one or more time points at any stage during pregnancy.

[0047] In addition to sequencing maternal cfDNA, paternal and maternal genomic DNA data (gDNA) were obtained by sequencing DNA from cell-containing samples using a whole-genome sequencing (WGS) approach to assign a priori probabilities to plasma sequencing reads as to their origin (fetal / maternal).

[0048] Typically, the method includes the following steps:

[0049] 1. WGS data collection: gDNA sequencing data from the mother and optionally the father, and cfDNA sequencing data obtained from the mother at at least two time points (at least one time point before pregnancy and at least one time point during pregnancy, or two or more time points during pregnancy);

[0050] 2. Variation analysis.

[0051] A probability value is assigned to each identified variant, thereby predicting whether the variant is of fetal origin. This probability value is assigned, for example, using the algorithm Hoobari described in Rabinowitz et al., 2019 and WO2021 / 340601, and is being calculated for each cfDNA dataset received at different time points to predict whether the variant is indeed inherited from the fetus at each genomic position. Figure 1 A flowchart describing this process is shown in .

[0052] Thus, in one aspect, the present invention provides a method for genotyping a fetus, comprising:

[0053] a. receiving sequencing data of maternal cell-free DNA (cfDNA) from a parent carrying a fetus, the sequencing data comprising at least two sets of sequencing reads collected at different pregnancy-related time points, wherein one set of sequencing reads is from time point A during pregnancy, and at least another set of sequencing reads is from time point B, wherein time point B is different from time point A and is before or during pregnancy;

[0054] b. receiving sequencing data of genomic DNA (gDNA), the sequencing data comprising a third set of sequencing reads of maternal gDNA and optionally a fourth set of sequencing reads of paternal gDNA of the paired parent of the fetus;

[0055] c. analyzing the first set and the second set to generate a background reference set;

[0056] d. analyzing the second set, the third set, and optionally the fourth set to identify sequence reads comprising: (i) a first set of sites where both parents have two copies of the same allele at a given locus, and (ii) a second set of sites where at least one of the parents has a variant allele;

[0057] e. for each site in the first group, determining a probability that the analyzed cfDNA sequence read originated from the fetus, wherein the determining comprises introducing at least the background reference set into at least one statistical model to obtain a probability value; and

[0058] f. Using the probability values, classify each locus in the second set as having fetal or maternal origin, thereby genotyping the fetus.

[0059] "Blood sample" herein refers to whole blood samples that have not been fractionated or separated into their component parts, as well as fractionated blood samples. Whole blood is typically used in conjunction with an anticoagulant (such as EDTA or ACD) during the collection process, but is typically untreated.

[0060] "Blood fractionation" is the process of fractionating or separating whole blood into its component parts. This is typically done by centrifuging the blood. The resulting components are: (a) a clear solution of plasma in the upper phase (which can be separated into its own fractions), (b) the buffy coat, which is a thin layer of white blood cells (leukocytes) mixed with platelets in the middle, and (c) red blood cells (erythrocytes) in the hematocrit fraction at the bottom of the centrifuge tube.

[0061] "Blood plasma" or "plasma" is the liquid component of blood (the component of blood that does not include cells). It makes up approximately 55% of the total volume of blood. It is primarily water (93% by volume) and contains dissolved proteins (including albumin, immunoglobulins, and fibrinogen), as well as glucose, clotting factors, electrolytes, hormones, carbon dioxide, and free DNA.

[0062] Plasma can be prepared by centrifuging a tube of whole blood in the presence of an anticoagulant until the blood cells separate and pull to the bottom of the tube. The plasma is then poured off or aspirated.

[0063] "Cell-free DNA" (cfDNA), also known as "circulating free DNA," is DNA fragments that exist outside of cells and circulate in the body in body fluids such as plasma. The length of cfDNA fragments is typically between about 150 and 200 base pairs (bp), and the average is about 170 bp, which may be related to the length of the DNA stretches wrapped around nucleosomes. During pregnancy, free fetal DNA can be found circulating in maternal plasma. Therefore, cfDNA in maternal plasma is a mixture of both maternal DNA and fetal DNA; the amount of cfDNA and the fraction of fetal DNA in it increase throughout pregnancy.

[0064] The term cfDNA also refers to fragments of DNA obtained from extracellular sources in vivo and isolated, separated, or otherwise processed in vitro. cfDNA can be obtained by extracting DNA from plasma after removing intact cells. Methods for extracting cfDNA are well known in the art, for example, as shown in the examples below.

[0065] The term "genomic DNA" or "gDNA" as used herein refers to DNA present in cells in vivo and containing the complete genome of a cell or organism. The term also refers to DNA obtained from cells in vivo and isolated, separated, or otherwise processed in vitro. Typically, cells are isolated before being lysed to produce in vitro cellular DNA. The term gDNA as used herein does not include cfDNA.

[0066] The term "sample" as used herein refers to a sample, typically derived from a biological fluid, cell, tissue, organ, or organism, comprising nucleic acid or a mixture of nucleic acids comprising at least one nucleic acid sequence to be analyzed for the presence of a genetic variation. Such samples include, but are not limited to, blood or blood fractions (e.g., peripheral blood mononuclear cells) obtained from a whole blood sample.

[0067] The sample is preferably obtained from a human subject.

[0068] The sample can be used directly, such as from a biological source, or after a pretreatment to alter the properties of the sample. The sample can be used fresh or frozen and then thawed.

[0069] DNA is extracted using standard protocols, eg, as described in the Examples below.

[0070] Maternal gDNA data, maternal cfDNA data, and optionally paternal gDNA data are obtained by sequencing methods including, but not limited to, deep whole-genome sequencing (WGS), whole-exome sequencing (WES), next-generation sequencing (NGS), targeted sequencing, panel sequencing, gene sequencing, long-read genome sequencing, paired-end sequencing, single-end sequencing, and amplicon sequencing.

[0071] The term "next generation sequencing" (NGS) herein refers to sequencing methods that allow for massively parallel sequencing of clonally amplified molecules and single nucleic acid molecules. Non-limiting examples of NGS include sequencing by synthesis using reversible dye terminators, and sequencing by ligation.

[0072] Deep sequencing involves sequencing a genomic region many times, sometimes hundreds or even thousands of times. This NGS approach allows researchers to detect rare genetic variants.

[0073] As used herein, the term "deep whole genome sequencing" refers to deep sequencing of the entire genome. In the context of the present invention, the cfDNA circulating in maternal plasma before and / or during pregnancy is subjected to deep whole genome sequencing. Sequencing is repeated many times, such as but not limited to 5 times (5X) and 2000 times (2000X), such as 5 times (5X), 10 times (10X), 20 times (20X), 30 times (30X), 50 times (50X), 100 times (100X), 200 times (200X), 300 times (300X), 500 times (500X), 1000 times (1000X), 2000 times (2000X). In a non-limiting example, the cfDNA in maternal plasma is sequenced 300 times (300X), for example, as described in the examples below. In addition, maternal and paternal gDNA are also affected by WGS. Such genomic DNA can be obtained from any cell type (e.g., from blood cells (e.g., leukocytes)). In an embodiment, whole genome sequencing of paternal and maternal gDNA is performed to a target depth of between about 20-fold and 40-fold (e.g., 30-fold).

[0074] Whole genome sequencing can be performed using any method known in the art (e.g., HiSeq X Ten system (Illumina), HiSeq 4000 (Illumina), nanopore WGS sequencing using the MinION device (Oxford Nanopore Technologies), and WGS by Ultima Genomics).

[0075] Sequencing generates "reads," which are sequences of DNA fragments of varying lengths. Typically, though not necessarily, a read represents a short sequence of consecutive bases in a sample. A read can be symbolically represented by a base pair sequence (in ATCG). It can be stored in a storage device and processed as needed. Reads can be obtained directly from a sequencing device or indirectly from stored sequence information.

[0076] Sequencing input can be long reads (e.g., from about 1 KBP (kilo base pairs) to about 100 KBP or more) or short reads (e.g., from between about 50 base pairs and 400 base pairs), or a combination of long and short reads.

[0077] After sequencing, the regions of overlap between reads are used to assemble and align the reads to a reference genome to reconstruct the complete DNA sequence.

[0078] As used herein, the terms "align," "alignment," or "aligning" refer to the process of comparing a read to a reference sequence and, therefore, determining whether the read is contained in the reference sequence. If the reference sequence contains the read, the read can be mapped to a specific position in the reference sequence. In some cases, alignment simply indicates whether a read is present in the reference sequence.

[0079] Optionally, additional information about one or both parents is also received (also referred to herein as "metadata"). The received metadata optionally and preferably includes at least one of the following characteristics, more preferably more than one: age, country of origin ( / ancestry), and known genetic conditions.

[0080] Sequence alignment technologies that can be used according to some embodiments of the present invention include, but are not limited to, BurrowsWheeler Aligner (BWA), ABA, ALE, AMAP, anon, BAli-Phy, Base-By-Base, BHAOS / DIALIGN, Bowtie, Bowtie 2, ClustalW, CodonCode Aligner, Comass, DECIPHER, DIALIGN-TX, DIALIGN-T, DNA Aligner, DNA Baser Sequencer Assembler, EDNA, FSA, Geneious, Kalign, MAFFT, MARNA, MAVID, MSA, MSAProbs, MULTALIN, Multi-LAGEN, MUSCLE, Opal, Pecan, Phylo, Praline, PicXAA, POA, Probal ign, ProbCons, PROMALS3D, PRRN / PRRP, PSAlign, RevTrans, SAGA, SAM, Se-AI, STAR, STAR-Fusion, StatAlign, Stemloc, T-Coffee, UGENE, VectorFriends and GLProbs.

[0081] Exemplary variant callers suitable for use in this embodiment include, but are not limited to, the Genome Analysis Toolkit (GATK) and Freebayes. For example, Freebayes can include alignments based on text sequences of reads aligned to a specific target, rather than their exact alignment. GATK can include: (i) preprocessing; (ii) variant discovery; and (iii) call set optimization. Preprocessing can include starting with raw sequence data (e.g., in FASTQ or uBAM format) and generating a BAM file for which analysis has been completed; processing can include alignment to a reference genome and data cleaning operations to correct for technical biases and make the data suitable for analysis; variant discovery can include starting with a BAM file for which analysis has been completed and generating a call set in VCF format; processing can involve identifying sites at which one or more individuals show possible genomic variation and applying filtering methods appropriate to the experimental design; call set refinement can include starting and ending with a VCF call set; processing can involve using metadata to assess and improve the accuracy of genotyping, appending additional information, and evaluating the overall quality of the call set.

[0082] Also contemplated are variant callers such as, but not limited to, Platypus, VarScan, Bowtie analysis, MuTect, and / or SAMtools. For example, Bowtie analysis can include implementing a Burrows-Wheeler transform for alignment. MuTect can include: (i) preprocessing; (ii) statistical analysis; and (iii) postprocessing. Preprocessing can include initial alignment of sequencing reads; statistical analysis can include using two Bayesian classifiers, one classifier that can detect whether a single nucleotide polymorphism (SNP) is a non-reference site at a given site, and for those sites that are found to be non-reference sites, another classifier that can ensure that normal sites do not carry the SNP; postprocessing can include removal of sequencing artifacts, short read alignment, and hybrid capture. SAMtools can include storing, manipulating, and aligning sequencing reads stored as SAM files.

[0083] In various exemplary embodiments of the invention, the method includes determining a probability that each variant site in the first set is of fetal origin.

[0084] As used herein, the term "variation" or "genetic variation" refers to a change in a DNA sequence (also known as a mutation). The methods of the present invention are suitable for detecting various types of genetic variation, including but not limited to single nucleotide polymorphisms (SNPs), copy number variations (CNVs), including insertions / deletions (Indels) and aneuploidies, STR (short tandem repeat) variations, amplification mutations, and de novo mutations.

[0085] The term "single nucleotide polymorphism" or "SNP" as used herein refers to a genomic variation at a single base position in DNA.

[0086] The term "copy number variation" (CNV) as used herein refers to any structural genomic variation in which the amount of a genomic sequence changes (increases or decreases). Thus, CNV encompasses deletions, insertions, duplications, proliferations, and translocations. CNV also encompasses chromosomal aneuploidies and partial aneuploidies.

[0087] The term "aneuploidy" as used herein refers to an imbalance in genetic material caused by the loss or gain of an entire chromosome or a portion of a chromosome.

[0088] The term "short tandem repeat variation" or "STR variation" (also known as microsatellite) as used herein refers to variation in short tandem repeat (STR) DNA sequences. These STRs involve repeating units of 1 to 6 base pairs and form a series of repeats up to 100 nucleotides in length.

[0089] The term "amplification mutation" herein refers to an increase in the number of copies of a repeat unit, typically a dinucleotide or trinucleotide.

[0090] The term "de novo mutation" as used herein refers to a germline mutation that occurs newly within a generation. That is, a new genetic variation arises in a fetus that neither parent carries.

[0091] In an embodiment, determining the probability that a variant is of fetal origin comprises constructing a fetal size distribution and a maternal size distribution, binning the fetal size distribution and calculating a fetal fraction for each fragment size bin, and calculating, for at least one size at the at least one site and at least one fragment, a probability that the fragment is fetal based on the fetal fraction of the corresponding fragment size bin to which the fragment belongs.

[0092] As used herein, the term "fetal fraction" or "FF" refers to the fraction of fetal cfDNA within the total amount of cfDNA in maternal blood. The fraction of fetal cfDNA in maternal blood (fetal fraction) varies throughout pregnancy and between individuals.

[0093] In an embodiment, said determining a probability comprises applying a Bayesian procedure. Optionally, said Bayesian procedure comprises a prior probability calculated using sequencing data of at least one of said parents.

[0094] In an embodiment, the process further comprises recalibrating the output of said Bayesian process using machine learning.

[0095] In a specific embodiment, determining the probability of each variant site being of fetal origin is performed as described in Rabinowitz et al., 2019 and WO2021 / 0340601.

[0096] As used herein, the term "heterozygous" refers to different versions (alleles) of a genomic locus. The term "homozygous" refers to the presence of the same version (allele) of a genomic locus.

[0097] The term "locus" is used to refer to a specific location of a nucleic acid sequence or variation on a reference chromosome.

[0098] In addition, cfDNA is analyzed to create a maternal cfDNA "fingerprint" or profile, which identifies unique characteristics of maternal cfDNA. These characteristics include, but are not limited to, read length, GC content, genetic linkage, haplotype, and any other measurable characteristics of cfDNA.

[0099] Next, genotype predictions received at earlier time points were used as additional prior distribution parameters and input into the Bayesian estimation process to further improve the accuracy level of maternal-fetal cfDNA separation.

[0100] As used herein, the term "about" means that the value may deviate by up to 1%, more specifically 5%, more specifically 10%, more specifically 15%, and in some cases up to 20% higher or lower than the value mentioned, and the deviation range includes integer values ​​and (if applicable) non-integer values ​​and constitutes a continuous range. Disclosed and described, it is to be understood that the invention is not limited to the specific examples, method steps and compositions disclosed herein, as the steps and compositions of such methods may vary. It is also to be understood that the terminology used herein is for the purpose of describing specific embodiments only and is not intended to be limiting, as the scope of the present invention is limited only by the appended claims and their equivalents.

[0101] It must be noted that, as used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise.

[0102] Throughout this specification and the accompanying examples and claims, unless the context requires otherwise, the word "comprise" and variations such as "comprises" and "comprising" will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

[0103] Examples

[0104] Materials and Methods

[0105] Sample collection and DNA extraction

[0106] With informed consent, samples were collected from each family during the 11th week of pregnancy. DNA from chorionic villus sampling (CVS) was extracted using the DNA tissue protocol of the MagNA Pure Compact Nucleic Acid Isolation Kit I – Large Capacity (Roche Life Science). Peripheral maternal blood was collected using 2 to 4 ethylenediaminetetraacetic acid (EDTA) tubes. Plasma was separated from total blood by centrifugation at 1600 x g for 10 minutes at 4°C. The plasma was then centrifuged again at 16,000 x g for 10 minutes at room temperature to remove any residual cells. cfDNA was extracted using the QIAamp Circulating Nucleic Acid Kit (Qiagen). Excess salts generated by cfDNA purification were removed using Agencourt AMPure XP beads (Beckman Coulter, Inc.) at a ratio of 2 times the volume of cfDNA. Pure maternal DNA was extracted from leukocytes in maternal buffy coats using a protocol that included (i) buffy coat isolation and (ii) DNA purification using the Gentra Puregene blood kit (Qiagen) according to the manufacturer's instructions. Pure paternal DNA was collected and purified similarly.

[0107] Library preparation and sequencing

[0108] Libraries were prepared for WGS samples using the TruSeq DNA PCR-free library preparation kit (Illumina) according to the manufacturer's instructions, and 151-bp paired-end reads were sequenced using the HiSeq X Ten system (Illumina).

[0109] For samples undergoing whole-exome sequencing (WES), library preparation was performed using the SureSelect V5 Exome Kit (Agilent) according to the manufacturer's instructions. Enrichment was achieved by hybridizing the prepared genomic DNA with complementary RNA probes. 101-bp paired-end reads were then sequenced using a HiSeq 4000 (Illumina).

[0110] Cell-free DNA samples were not fragmented during library preparation and sequenced using HiSeq 4000 (Illumina) with 151-bp paired-end reads to the required 300× coverage.

[0111] Compare with genome

[0112] Reads were aligned to the Genome Reference Consortium Human Build 38 (GRCh38 / hg38) using Burrows-Wheeler v0.7.834 with default parameters. Duplicate reads resulting from PCR clonality or optical duplications, as well as reads mapping to multiple positions, were excluded from downstream analysis.

[0113] Variant calling from pure genome sequencing data

[0114] Single-nucleotide substitutions and small insertions and deletions were identified using GATK HaplotypeCaller software v4.2.4.0 with default parameters. HaplotypeCaller was first run on the aligned sequencing data of the two parents and then on the aligned data of the CVS sample using the variant sites identified in the parental genomes. Reported variants were not filtered, so all reported SNPs and indels were retained for downstream analysis.

[0115] Preprocessing of cell-free DNA data

[0116] HaplotypeCaller was run on cfDNA samples only at variant sites identified in the parental genomes. Using Hoobari, the observed allele for each read was stored in a separate database along with the read insert size.

[0117] Non-invasive fetal mutation detection

[0118] Hoobari is run using a database of parental variants and pre-processed cfDNA results as input. The output is a standard variant call format (VCF) file. The results are analyzed using several VCF-specific software packages, such as vcflib and vcftools.

[0119] Bayesian non-invasive genotyping

[0120] At each locus of interest, a Bayesian calculation is applied. For each possible fetal genotype:

[0121]

[0122] where G is the fetal genotype, and G i is the i-th possible fetal genotype out of n possibilities. For biallelic variants, the reference allele (AA) is homozygous, (Aa) is heterozygous, or the alternative allele (aa) is homozygous. P(G) is the prior probability of each genotype and is calculated based on Mendelian inheritance. The data variable represents the reads covering the site, and P(data|G) represents the likelihood function, which is defined in this example as the product of the likelihoods of each read:

[0123]

[0124] Read segment r j The probability of β-catenin depends on the fetal genotype and is calculated using the maternal genotype and fetal fraction. j |fet)P(r j |fet) and P(r j mat) are the probability of observing a read that supports a certain allele, considering that the read is fetal or maternal. This depends on the fetal genotype tested G i , maternal genotype G M and observed alleles. P(fet), P(fet), and P(mat) are the probabilities of observing a fetal or maternal read based solely on the fetal fraction, regardless of the allele it supports. To exploit the size difference between fetal and maternal fragments, the fetal fraction for each read is calculated only based on reads with the same fragment size. For reads that are not correctly paired or have a fragment size > 500, the total fetal fraction is used.

Claims

1. A method for genotyping a fetus, comprising: a. receiving sequencing data of maternal cell-free DNA (cfDNA) from a parent carrying a fetus, the sequencing data comprising at least two sets of sequencing reads collected at different pregnancy-related time points, wherein one set of sequencing reads is from time point A during pregnancy and at least another set of sequencing reads is from time point B, wherein Time point B is different from time point A and occurs before or during pregnancy; b. receiving sequencing data of genomic DNA (gDNA), the sequencing data comprising a third set of sequencing reads of maternal gDNA and optionally a fourth set of sequencing reads of paternal gDNA of the paired parent of the fetus; c. analyzing the first set and the second set to generate a background reference set; d. analyzing the second set, the third set, and optionally the fourth set to identify sequence reads comprising: (i) a first set of sites where both parents have two copies of the same allele at a given locus, and (ii) a second set of sites where at least one of the parents has a variant allele; e. for each site in the first group, determining a probability that the analyzed cfDNA sequence reads originated from the fetus, wherein the determining comprises subjecting at least the background reference set to at least one statistical model to obtain a probability value; as well as f. Using the probability values, classify each locus in the second set as having fetal or maternal origin, thereby genotyping the fetus.

2. The method of claim 1 , wherein one or both of the gDNA sequencing data and the cfDNA sequencing data are obtained by a method selected from the group consisting of whole genome sequencing (WGS), whole exome sequencing (WES), next generation sequencing (NGS), targeted sequencing, panel sequencing, gene sequencing, long read genome sequencing, paired-end sequencing, single-end sequencing, and amplicon sequencing.

3. The method according to claim 2, wherein the WGS or WES data are obtained by deep sequencing.

4. The method according to any one of the preceding claims, wherein determining the probability value is based on at least one sequence alignment map (SAM) parameter.

5. Method according to any of the preceding claims, wherein determining the probability value is based on additional data parameters.

6. The method of claim 5, wherein the additional parameters include read length, GC content, genetic linkage and haplotype.

7. The method according to any one of the preceding claims, wherein said step (e) in claim 1 further comprises calculating a total fetal fraction.

8. The method of claim 7, further comprising constructing a fetal size distribution and a maternal size distribution, wherein determining the probability value comprises: The fetal size distribution is binned and a fetal fraction is calculated for each fragment size bin, and for at least one size at the at least one site and at least one fragment, a probability that the fragment is fetal is calculated based on the fetal fraction of the corresponding fragment size bin to which the fragment belongs.

9. The method according to any of the preceding claims, wherein determining the probability value comprises applying a statistical model, such as a Bayesian procedure.

10. The method of claim 9, further comprising recalibrating the output of the Bayesian program using machine learning.

11. A method according to any of the preceding claims, wherein the background reference set comprises one or both of the following: (i) maternal cfDNA sequence reads before pregnancy, and (ii) maternal cfDNA sequence reads sampled at one or more time points during the pregnancy. The method of claim 11 , wherein the background reference set comprises a time-dependent modulator.

13. The method of claim 12, wherein the time-dependent modulator is based on data parameters including length, GC content, genetic linkage and haplotype of the reads.

14. A method according to any one of the preceding claims, wherein the probability value is determined using the Hoobari algorithm.

15. A method according to any of the preceding claims, wherein the maternal cfDNA is sampled at least at two different time points, such that cfDNA sequencing data reads are received from at least one time point before the pregnancy and one or more time points during the pregnancy with the fetus.

16. The method of claim 15, wherein the maternal cfDNA is sampled at two or more different time points during the pregnancy.

17. The method of claim 16, wherein the two or more samples are at time points at least one day apart from each other.

18. The method of any of the preceding claims, wherein the maternal cfDNA is sampled early in the pregnancy.

19. The method of claim 18, wherein the maternal cfDNA is sampled during the first trimester of the pregnancy.

20. The method of any one of claims 1 to 17, wherein the maternal cfDNA is sampled during the second trimester and / or the third trimester.

21. A computer software product comprising a computer-readable medium having program instructions stored therein, which, when read by a data processor, configure the data processor to: (1) receive (i) sequencing data reads of maternal cell-free DNA (cfDNA) comprising at least two sets of sequencing reads collected at different pregnancy-related time points, and (ii) sequencing data reads of maternal and optionally paternal genomic DNA (gDNA) from a pair of parents raising a fetus, and (2) perform the method according to any one of claims 1 to 20.

22. A system for genotyping a fetus, comprising: An input utility for receiving (i) sequencing data reads of maternal cell-free DNA (cfDNA) comprising at least two sets of sequencing reads collected at different pregnancy-related time points, and (ii) sequencing data reads of maternal and optionally paternal genomic DNA (gDNA) from a pair of parents raising a fetus; and a data processor configured to analyze the data to perform the method according to any one of claims 1 to 20.

Images