Methods, apparatus, devices, and media of whole genome reconstitution of embryos

Abstract

The application provides a method, device, equipment and medium for reconstructing a whole genome of an embryo. The method comprises correcting and filling a whole genome genotype of the embryo. The method comprises embryo sample data processing, parent sample data processing, effective site selection, haplotype framework construction, reconstruction and filling of the whole genome of the embryo. The method can correct and fill free DNA in embryo culture solution. The method is based on genotype information of a to-be-tested sample at a single nucleotide polymorphism site, genotype and haplotype information of parents of the sample, and determines a haplotype of the embryo, and then corrects and fills the whole genome genotype of the embryo by using allele information in the haplotype of the parents of the sample.

Description

Technical Field

[0001] This invention belongs to the field of genetic testing, specifically relating to a method, apparatus, device, and medium for reconstructing the whole genome of an embryo capable of correcting and filling free DNA in embryo culture medium. Background Technology

[0002] Assisted reproductive technology has made significant theoretical and technological progress in overcoming human infertility. Babies born through assisted reproductive procedures account for about 3%-4% of the world's total births, but the practical implementation of the concept of "healthy babies" still faces challenges.

[0003] Humans are diploid organisms, meaning their DNA comes half from their father and half from their mother. Current whole-genome sequencing methods are based on the assumption that the sample is diploid (i.e., containing genetic material from both parents), and the culture medium contains cell-free DNA (cfDNA), including mitochondrial DNA (mtDNA) and genomic DNA (gDNA). Using next-generation high-throughput sequencing to detect cfDNA in the culture medium can determine whether the embryo has chromosomal abnormalities and, combined with the grading assessment of the embryo's implantation potential, guide clinical practice and improve the success rate of assisted reproductive technology. However, for populations with polygenic familial genetic diseases, whole-genome sequencing is needed during assisted reproduction to score the embryo and predict its relative risk of developing the disease in adulthood. Large fragments are lost in the culture medium sample, leading to genotype errors. For preimplantation genetic testing (PGT) specifically PGT-P (PGT for polygenic disorders), the accuracy of the genotypes obtained using existing sequencing methods is insufficient to meet the requirements of PGT. However, research has found that cell-free DNA in the culture medium exists in a subgenomic form, meaning that some genomic regions contain only DNA from the father, while others contain only DNA from the mother. This results in large, random segments of genetic information lost from one parent, making it impossible to obtain the true genotype of the embryo through sequencing of cell-free DNA in the culture medium. To achieve non-invasive preimplantation genetic testing (PGT) based on cell-free DNA in the culture medium, the detected genotype must be corrected and filled in. Summary of the Invention

[0004] To address the aforementioned technical problems, the present invention aims to provide a method, apparatus, device, and medium for reconstructing the whole genome of an embryo. This method can correct and fill cell-free DNA in embryo culture medium. Based on the genotype information of the sample to be tested at SNP (Single Nucleotide Polymorphism) sites and the genotype and haplotype information of the sample's parents, the method first determines the haplotype of the embryo, and then uses the allele information from the haplotypes of the sample's parents to correct and fill the whole genome genotype of the embryo.

[0005] To achieve the above objectives, the present invention provides a method for reconstructing the whole genome of an embryo, wherein the reconstruction method includes correcting and filling in the whole genome genotype of the embryo; the reconstruction method includes the following steps: Embryo sample data processing: The whole genome amplification products of the embryo culture medium were extracted and next-generation sequencing was performed. The raw sequencing data were preprocessed to obtain the preprocessed sequencing data of the embryo culture medium samples. Kinship sample data processing: Next-generation sequencing of genomic DNA from peripheral blood samples and / or family pedigree samples of embryonic parents to determine the haplotypes of embryonic parents; Effective locus selection: Based on the haplotypes of the embryo parents, effective locus selection was performed on the alleles of the sequencing data after preprocessing the embryo culture medium samples to obtain the effective locus selection results for each locus; Haplotype framework construction: Based on the effective site selection results, the initial paternal and maternal haplotype frameworks of the embryo are obtained; Reconstruction and filling of the whole genome of the embryo: Based on the hidden Markov model, the haplotype continuity of the initial paternal and maternal haplotype framework of the embryo is inferred to obtain the inferred paternal and maternal haplotypes; based on the paternal and maternal haplotypes, the complete diploid genotype of the embryo is reconstructed at each locus, and the genotype, its reliability score and polygenic risk score of each locus are output.

[0006] In this invention, the sequencing platform is a second-generation sequencing platform, including one or more combinations of Illumina, ThermoFisher Scientific, BGI, and Genemax.

[0007] According to a specific embodiment of the present invention, the whole genome amplification method is transposon linear amplification technology (LIANTI). The method for preprocessing the raw sequencing data includes: aligning the raw data to the human reference genome to obtain alignment results; performing deduplication and quality control on the alignment results; and performing single nucleotide polymorphism detection to obtain preprocessed sequencing data of the embryo culture medium sample, which is a VCF file. Preferably, the version of the human reference genome is hg19, and the alignment method is BWA (Burrows-Wheeler Aligner). Preferably, the tool for detecting single nucleotide polymorphisms is the Genome Analysis Toolkit (GATK). Preferably, the sequencing data after pretreatment of the embryo culture medium sample includes the sequencing error rate.

[0008] According to a specific embodiment of the present invention, the family pedigree sample includes one or more of the following: proband sample, discarded embryo sample, and amniotic fluid sample; The method for determining the haplotypes of the embryonic parents is as follows: Bayesian linkage analysis or population-based haplotype inference method is used to determine the haplotypes of the embryonic parents. Preferably, the population-based haplotype inference method includes SHAPEIT2.

[0009] According to a specific embodiment of the present invention, the parental effective sites include paternal effective sites, maternal effective sites, and dual-source effective sites. Sites that do not belong to the paternal effective sites, maternal effective sites, and dual-source effective sites do not participate in the subsequent construction. Preferably, the definitions of the paternal effective site, maternal effective site, and dual-source effective site are as follows: Paternal effective locus: The allele detected in the embryo culture medium is identical only to the paternal haplotype, and the mother is homozygous at this locus; Maternally effective locus: Alleles detected in embryo culture medium are identical only to the mother's haplotype, and the father is homozygous at this locus; Dual-source effective loci: Alleles from both parents were detected in the embryo culture medium, and both were consistent with the haplotypes of the parents.

[0010] According to a specific embodiment of the present invention, the missing regions in the initial paternal and maternal haplotype framework of the embryo are marked as locations where haplotypes could not be determined (i.e., "contamination or noise").

[0011] According to a specific embodiment of the present invention, the method for inferring the continuity of the monomer type is as follows: Record the state space: estimate the maternal cell contamination rate, allele loss rate and haploid detection status using a Bayesian linkage analysis model. The haploid detection status consists of the possible haplotype source status for each single nucleotide polymorphism site, including paternal, maternal or bi-source haplotype source status. Record observations: Record the allele sequencing depth of the embryo culture medium sample at this locus; Obtaining the transfer probability: The transfer probability is calculated based on the genetic distance (cM) and the known recombination rate; Obtaining the emission probability: Construct a binomial distribution model based on the parameters to obtain the emission probability; Obtaining continuous paternal and maternal haplotypes: The most likely haplotype path is inferred through the Viterbi algorithm or the forward-backward algorithm, thereby extending the continuous paternal and maternal haplotypes of the embryo. Preferably, the genetic distance and the known recombination rate are obtained from the DECODE database and / or HapMap.

[0012] Preferably, in the step of obtaining the emission probability, the parameters include one or a combination of two or more of the following: maternal cell contamination rate, allele loss rate, and sequencing error rate.

[0013] Preferably, the method for calculating the transfer probability based on genetic distance (cM) and the known recombination rate is as follows: (i) Provide s datasets for the analysis, where s is a positive integer ≥3, wherein the datasets are genomic information-related datasets, and the datasets include: a first dataset from the object, a second dataset from the object's father, and a third dataset from the object's mother. (ii) Select Y1 target sites, where Y1 is a positive integer ≥ 1; (iii) For each target site selected in the previous step, molecular markers are analyzed and detected in their upstream and downstream regions, thereby identifying at least one molecular marker for each target site in its upstream and downstream regions. (iv) In each dataset, the molecular markers determined in step (iii) are labeled to obtain the corresponding labeled dataset 1, dataset 2, and dataset 3; (v) For each molecular marker site upstream and downstream of each target site in each dataset, construct a binary genetic vector of (0, 1). The n datasets constitute a vector Vi of 2n numbers, where i represents the site and Vi is the hidden Markov chain state; where n is s or sj, where s is as defined above, and j is the number of ancestral individuals without parents at the top level (i.e. individuals without parents in the family). (vi) For each target site, the maximum likelihood estimate L is determined using a hidden Markov model according to the following formula:

[0014] in, m represents the number of molecular markers upstream and downstream of each target site; P(V1) represents the prior value of the genetic vector; P(V i |V i-1 ) represents the haplotype state transition probability between two adjacent sites; G i This represents the genotype observation value at the i-th locus; P(G i |V i ) represents the emission probability in a single-unit state; (vii) Using the Viterbi dynamic programming algorithm, V1, V2, ... V mThe most likely composition is estimated to determine the haplotype of the object. According to a specific embodiment of the present invention, the location where the haplotype could not be determined is filled using a population reference panel combined with Minimac4 software for genotype imputation based on a hidden Markov model. Preferably, the group reference panel includes 1000 Genomes.

[0015] On the other hand, the present invention also provides an apparatus for reconstructing the whole genome of an embryo, the apparatus comprising: Embryo sample data processing module: Extract whole genome amplification products from embryo culture medium and perform next-generation sequencing, preprocess the raw sequencing data to obtain preprocessed sequencing data of embryo culture medium samples; Kinship sample data processing module: performs high-depth sequencing of gDNA from peripheral blood samples and / or family pedigree samples of embryonic parents, and determines the haplotypes of embryonic parents; Effective locus selection module: Based on the haplotypes of the embryo parents, the module selects effective loci from the parental sources in the alleles of the sequencing data after preprocessing the embryo culture medium samples, and obtains the effective locus selection results for each locus; Haplotype framework construction module: Based on the effective site selection results, the initial paternal and maternal haplotype frameworks of the embryo are obtained; Embryo whole genome reconstruction and filling module: Based on the hidden Markov model, the haplotype continuity of the initial paternal and maternal haplotype framework of the embryo is inferred to obtain the inferred paternal and maternal haplotypes; based on the paternal and maternal haplotypes, the complete diploid genotype of the embryo is reconstructed at each locus, and the genotype, reliability score and polygenic risk score of each locus are output.

[0016] On the other hand, the present invention also provides an electronic device comprising: First processor; and memory for storing executable instructions of the first processor; The first processor is configured to execute the above-described method for reconstructing the whole genome of an embryo by executing the executable instructions.

[0017] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein, When the computer program is executed by the second processor, it implements the above-described method for reconstructing the whole genome of the embryo.

[0018] The method of this invention corrects genotype errors caused by large fragment loss, achieving whole-genome genotype filling in culture medium samples for the first time, laying the foundation for the application of non-invasive PGT-P technology. Specifically, it includes the following key technical points: 1. The reconstruction method of the present invention includes the selection rules for parental effective sites and maternal effective sites.

[0019] 2. The reconstruction method of the present invention is based on a hidden Markov model and combines genetic distance to determine the paternal haplotype and maternal haplotype of the embryo.

[0020] 3. The reconstruction method of the present invention reconstructs the whole genome genotype of the embryo based on the paternal and maternal haplotypes of the embryo.

[0021] 4. The reconstruction method of this invention is based on a haploid hypothesis-based genotype filling strategy for whole-genome genotype filling. Traditional whole-genome filling methods rely on the diploid hypothesis, which assumes that each locus contains alleles from both parents. However, cell-free DNA in embryo culture medium exhibits significant "haploid region" characteristics, meaning that only the paternal or maternal allele can be detected in certain genomic regions, resulting in random large fragment loss. The method of this invention proposes for the first time a genotype filling strategy based on the haploid hypothesis, which avoids filling errors caused by incorrect assumptions by pre-determining the haploid detection status (paternal, maternal, or biploid) of each SNP locus.

[0022] 5. This invention integrates Bayesian linkage analysis (used to determine the chromosomes inherited from parents in an embryo) and Hidden Markov Models (HMMs) (used to infer the genotype of undetected SNP loci based on genetic distance) to construct a two-stage joint imputation framework, as follows: Phase 1: Estimate maternal cell contamination rate, allele loss rate, and haploid detection status using a Bayesian linkage analysis model; Phase 2: Based on parental haplotypes and known genetic recombination patterns, use HMM to infer paternal and maternal haplotypes of the embryo across the entire genome.

[0023] 6. Multi-level Reliability Classification Mechanism: Addressing the issue of inconsistent data quality in culture medium samples, this invention provides a reliability assessment for each locus (high reliability, moderate reliability, probable reliability, undetermined) along with the output of the filled genotype, and ultimately outputs the overall diagnostic reliability level of the embryo (Highly Confident, Moderately Confident, Possibly Confident, Undetermined). This mechanism significantly improves the reliability of clinical applications. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the analysis process of the method of the present invention.

[0025] Figure 2 Circos diagram showing the results of whole-genome genotyping reconstruction of embryo culture medium samples. Detailed Implementation

[0026] In order to provide a clearer understanding of the technical features, objectives and beneficial effects of the present invention, the technical solution of the present invention will now be described in detail below, but it should not be construed as limiting the scope of implementation of the present invention.

[0027] It should be noted that, unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0028] Unless otherwise specified, all raw materials, reagents, instruments and equipment used in this invention can be purchased from the market or prepared by existing methods.

[0029] It should be understood that the terms “comprising,” “including,” and / or “containing” as used herein specify the presence of the stated features, integers, steps, components, or combinations thereof, but do not exclude the presence or addition of one or more other features, integers, steps, components, or combinations thereof.

[0030] The endpoints and any values ​​of the ranges disclosed in this invention are not limited to the precise ranges or values, and these ranges or values ​​should be understood to include values ​​close to these ranges or values. For numerical ranges, the endpoint values ​​of the various ranges, the endpoint values ​​of the various ranges and individual point values, and individual point values ​​can be combined with each other to obtain one or more new numerical ranges, which should be considered as specifically disclosed in this invention.

[0031] Example 1

[0032] This embodiment provides a method for reconstructing the whole genome of an embryo capable of correcting and filling free DNA in the embryo culture medium. The specific steps are as follows: Samples: A total of 14 families (autosomal dominant, recessive and X-linked genetic diseases) were included, and 126 embryo culture medium samples were collected. Each culture medium was matched with at least one of the following as the gold standard: trophectoderm biopsy, discarded embryo or amniotic fluid sample.

[0033] Refactoring process: A schematic diagram of the analysis process of the method of the present invention is shown below. Figure 1 As shown, the specific process is as follows: (1) Data preprocessing 1. Extract gDNA from peripheral blood samples of the embryonic parents and whole-genome amplification products from embryo culture medium; wherein, the modified LIANTI method was used to amplify the whole genome of the culture medium samples to improve the amplification efficiency and uniformity of short DNA fragments; 3. Perform NGS sequencing on the whole genome amplification products of the embryo culture medium to obtain raw data in fastq format; 4. Use BWA to align the raw data in fastq format to the hg19 human reference genome to obtain the alignment results; 5. After deduplication and quality control of the results, SNP detection was performed using GATK to obtain the downstream input vcf file containing information such as genotype; 6. Next-generation sequencing was performed on parental and family samples (trophic ectodermal biopsy, discarded embryos, or amniotic fluid samples), and the haplotypes of the parents were determined using Bayesian linkage analysis or population-based haplotype inference methods (such as SHAPEIT2).

[0034] (2) Valid site selection: Valid sites are selected for each site in the VCF file based on the haplotype of the parents, according to the following rules: Paternal effective locus: The allele detected in the culture medium sample is identical only to the paternal haplotype, and the mother is homozygous at this locus; Maternally effective locus: Alleles detected in the culture medium sample are identical only to the mother's haplotype, and the father is homozygous at this locus; Dual-source effective sites: Alleles from both parents detected in the culture medium sample, and both are identical to the haplotypes of the parents; If an allele at a certain locus cannot match either parent, it is marked as "contamination or noise" and will not participate in subsequent construction.

[0035] (3) Construction of a single-unit framework

[0036] Based on the aforementioned valid sites, the initial paternal and maternal haplotype framework of the embryo was determined. For missing regions, vacancies were temporarily reserved.

[0037] (4) Haplotype continuity inference based on hidden Markov model: Bayesian linkage analysis was used to determine the chromosomes inherited from the parents in the embryo, and the genotypes of undetected SNP loci were inferred from the genetic distance using a hidden Markov model, thus constructing a two-stage joint filling framework: Phase 1: Estimate maternal cell contamination rate, allele loss rate, and haploid detection status using a Bayesian linkage analysis model; Phase 2: Based on parental haplotypes and known genetic recombination patterns, use HMM to infer paternal and maternal haplotypes of the embryo across the entire genome.

[0038] The following combined results were obtained after two phases of analysis: State space: the possible haplotype origin states (paternal, maternal, and bi-origin) for each SNP site. Observation: Allele sequencing depth at this locus in embryo culture medium samples; Translocation probability: The translocation probability is calculated based on genetic distance (cM) and the known recombination rate (obtained from the DECODE database or HapMap); the calculation method is as follows: (i) Provide s datasets for the analysis, where s is a positive integer ≥3, wherein the datasets are genomic information-related datasets, and the datasets include: a first dataset from the object, a second dataset from the object's father, and a third dataset from the object's mother. (ii) Select Y1 target sites, where Y1 is a positive integer ≥ 1; (iii) For each target site selected in the previous step, molecular markers are analyzed and detected in their upstream and downstream regions, thereby identifying at least one molecular marker for each target site in its upstream and downstream regions. (iv) In each dataset, the molecular markers determined in step (iii) are labeled to obtain the corresponding labeled dataset 1, dataset 2, and dataset 3; (v) For each molecular marker site upstream and downstream of each target site in each dataset, construct a binary genetic vector of (0, 1). The n datasets constitute a vector Vi of 2n numbers, where i represents the site and Vi is the hidden Markov chain state; where n is s or sj, where s is as defined above, and j is the number of ancestral individuals without parents at the top level (i.e. individuals without parents in the family). (vi) For each target site, the maximum likelihood estimate L is determined using a hidden Markov model according to the following formula:

[0039] in, m represents the number of molecular markers upstream and downstream of each target site; P(V1) represents the prior value of the genetic vector; P(V i |V i-1 ) represents the haplotype state transition probability between two adjacent sites; G i This represents the genotype observation value at the i-th locus; P(G i |V i ) represents the emission probability in a single-unit state; (vii) Using the Viterbi dynamic programming algorithm, V1, V2, ... V m The maximum possible composition is estimated to determine the object's monomer type.

[0040] Emission probability: A binomial distribution model was constructed based on parameters such as maternal cell contamination rate, allele loss rate, and sequencing error rate (from preprocessed sequencing data); Inferring paternal and maternal haplotypes: The most likely haplotype path is inferred using the Viterbi algorithm or the forward-backward algorithm, thereby extending the initial paternal and maternal haplotype framework of the embryo to the continuous paternal and maternal haplotypes of the embryo, and obtaining the inferred paternal and maternal haplotypes.

[0041] (5) Reconstruction and filling of the whole genome

[0042] Based on the inferred paternal and maternal haplotypes, the complete diploid genome genotype of the embryo is synthesized at each locus; Population imputation: For haplotypes whose locations could not be inferred, HMM-based genotype imputation was performed using a population reference panel (e.g., 1000 Genomes) in conjunction with Minimac4 software. The final output is the genotype and its reliability score for each locus.

[0043] (6) Based on the genotype of the filled embryo whole genome, perform polygenic risk scoring (PRS) to assess the feasibility of using it to predict the risk of type 2 diabetes (T2D) with multiple genes.

[0044] Data Results: (1) Results of whole genome reconstruction The original SNP detection density was 222 ± 127 SNPs / Mb, with a genome-wide coverage of 6.4%. After reconstruction using the method described in the examples, the SNP density increased to 1479 ± 554 SNPs / Mb, with a coverage rate of 93.5% (Circos diagram of whole-genome genotyping reconstruction results for embryo culture medium samples is shown in the figure). Figure 2 As shown, this illustrates a comparison of the original and filled SNP densities. In samples with low maternal cell contamination rates (<2%), the paternal haplotype concordance was 99.8%, and the maternal haplotype concordance was 98.6%. The overall consistency between the reconstructed genotype and the gold standard averaged 96.9%, reaching as high as 99.9% in some samples.

[0045] (2) Application example of polygenic risk score (PRS) (taking type 2 diabetes T2D as an example)

[0046] A total of 114 T2D-related risk loci were included; Directly detected sites accounted for 28.7%; Loci obtained through the method (family reconstruction) of this embodiment accounted for 52.5%; 18.8% of the sites were obtained through population imputation; Relevance R of reconstructed genotype to gold standard PRS 2 = 0.79, PRS percentile correlation R 2 = 0.68.

[0047] like Figure 2 As shown, the raw SNPs directly detected in the culture medium sample are very sparse, covering only a small number of type 2 diabetes risk sites (blue triangles), and cannot be used to predict the polygenic risk of type 2 diabetes; while the number of SNPs after whole-genome filling increases significantly, covering the vast majority of type 2 diabetes risk sites (orange triangles), and can be used to predict the polygenic risk of type 2 diabetes.

[0048] Technical effects: This invention's reconstruction method achieves, for the first time, high-precision whole-genome genotyping reconstruction based on cell-free DNA in embryo culture medium, overcoming application bottlenecks caused by traditional methods such as DNA fragment loss, maternal contamination, and low coverage. This method is not only applicable to non-invasive preimplantation genetic testing (niPGT-M) for monogenic diseases but also provides a feasible pathway for risk assessment of polygenic diseases (niPGT-P). Validation in 126 clinical samples showed that the reconstructed genotypes were highly consistent with the gold standard (100% accuracy, 74.6% reporting rate), and the output reliability was graded, demonstrating clear clinical translational potential.

[0049] The above embodiments illustrate and describe the main features and advantages of the present invention in detail. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for reconstructing the whole genome of an embryo, wherein, The reconstruction method includes correcting and filling the whole genome genotype of the embryo; the reconstruction method includes the following steps: Embryo sample data processing: The whole genome amplification products of the embryo culture medium were extracted and next-generation sequencing was performed. The raw sequencing data were preprocessed to obtain the preprocessed sequencing data of the embryo culture medium samples. Kinship sample data processing: Next-generation sequencing of genomic DNA from peripheral blood samples and / or family pedigree samples of embryonic parents to determine the haplotypes of embryonic parents; Effective locus selection: Based on the haplotypes of the embryo parents, effective locus selection was performed on the alleles of the sequencing data after preprocessing the embryo culture medium samples to obtain the effective locus selection results for each locus; Haplotype framework construction: Based on the effective site selection results, the initial paternal and maternal haplotype frameworks of the embryo are obtained; Reconstruction and filling of the whole genome of the embryo: Based on the hidden Markov model, the haplotype continuity of the initial paternal and maternal haplotype framework of the embryo is inferred to obtain the inferred paternal and maternal haplotypes; based on the paternal and maternal haplotypes, the complete diploid genotype of the embryo is reconstructed at each locus, and the genotype, its reliability score and polygenic risk score of each locus are output.

2. The reconstruction method according to claim 1, wherein, The method for whole-genome amplification is transposon linear amplification technology; The method for preprocessing the raw sequencing data includes: aligning the raw data to the human reference genome to obtain alignment results; performing deduplication and quality control on the alignment results; and performing single nucleotide polymorphism detection to obtain preprocessed sequencing data of the embryo culture medium sample, which is a VCF file. Preferably, the version of the human reference genome is hg19, and the alignment method is BWA (Burrows-WheelerAligner). Preferably, the tool for detecting single nucleotide polymorphisms is the Genome Analysis Toolkit (GATK). Preferably, the sequencing data after pretreatment of the embryo culture medium sample includes the sequencing error rate.

3. The reconstruction method according to claim 1, wherein, The family pedigree sample includes one or more of the following: proband sample, discarded embryo sample, and amniotic fluid sample; The method for determining the haplotypes of the embryonic parents is as follows: Bayesian linkage analysis or population-based haplotype inference method is used to determine the haplotypes of the embryonic parents. Preferably, the population-based haplotype inference method includes SHAPEIT2.

4. The reconstruction method according to claim 1, wherein, The parental effective loci include paternal effective loci, maternal effective loci, and dual-source effective loci. Loci that do not belong to paternal effective loci, maternal effective loci, or dual-source effective loci will not participate in the subsequent construction. Preferably, the definitions of the paternal effective site, maternal effective site, and dual-source effective site are as follows: Paternal effective locus: The allele detected in the embryo culture medium is identical only to the paternal haplotype, and the mother is homozygous at this locus; Maternally effective locus: Alleles detected in embryo culture medium are identical only to the mother's haplotype, and the father is homozygous at this locus; Dual-source effective loci: Alleles from both parents were detected in the embryo culture medium, and both were consistent with the haplotypes of the parents.

5. The reconstruction method according to claim 1, wherein, Missing regions in the initial paternal and maternal haplotype frameworks of the embryo were marked as locations where haplotypes could not be determined.

6. The reconstruction method according to claim 1, wherein, The method for inferring the continuity of the monomer type is as follows: Record the state space: estimate the maternal cell contamination rate, allele loss rate and haploid detection status using a Bayesian linkage analysis model. The haploid detection status consists of the possible haplotype source status for each single nucleotide polymorphism site, including paternal, maternal or bi-source haplotype source status. Record observations: Record the allele sequencing depth of the embryo culture medium sample at this locus; Obtaining the transfer probability: The transfer probability is calculated based on the genetic distance and the known recombination rate; Obtaining the emission probability: Construct a binomial distribution model based on the parameters to obtain the emission probability; Obtaining continuous paternal and maternal haplotypes: The most likely haplotype path is inferred through the Viterbi algorithm or the forward-backward algorithm, thereby extending the continuous paternal and maternal haplotypes of the embryo. Preferably, the genetic distance and the known recombination rate are derived from a DECODE database and / or HapMap; Preferably, in the step of obtaining the emission probability, the parameters include one or a combination of two or more of the following: maternal cell contamination rate, allele loss rate, and sequencing error rate.

7. The reconstruction method according to claim 5, wherein, The location of the undetermined haplotype was filled by genotype imputation based on a hidden Markov model using a population reference panel and Minimac4 software. Preferably, the group reference panel includes 1000 Genomes.

8. A device for reconstructing the whole genome of an embryo, wherein, The device includes: Embryo sample data processing module: Extract whole genome amplification products from embryo culture medium and perform next-generation sequencing, preprocess the raw sequencing data to obtain preprocessed sequencing data of embryo culture medium samples; Kinship sample data processing module: performs high-depth sequencing of gDNA from peripheral blood samples and / or family pedigree samples of embryonic parents, and determines the haplotypes of embryonic parents; Effective locus selection module: Based on the haplotypes of the embryo parents, the module selects effective loci from the parental sources in the alleles of the sequencing data after preprocessing the embryo culture medium samples, and obtains the effective locus selection results for each locus; Haplotype framework construction module: Based on the effective site selection results, the initial paternal and maternal haplotype frameworks of the embryo are obtained; Embryo whole genome reconstruction and filling module: Based on the hidden Markov model, the haplotype continuity of the initial paternal and maternal haplotype framework of the embryo is inferred to obtain the inferred paternal and maternal haplotypes; based on the paternal and maternal haplotypes, the complete diploid genotype of the embryo is reconstructed at each locus, and the genotype, reliability score and polygenic risk score of each locus are output.

9. An electronic device comprising: First processor; and memory for storing executable instructions of the first processor; The first processor is configured to execute the method for reconstructing the whole genome of an embryo according to any one of claims 1-7 by executing the executable instructions.

10. A computer-readable storage medium having a computer program stored thereon, wherein, When the computer program is executed by the second processor, it implements the method for reconstructing the whole genome of an embryo as described in any one of claims 1-7.

Images