Advertisement

Analysis of parental contribution for aneuploidy detection (APCAD): a novel method to detect aneuploidy and mosaicism in preimplantation embryos

Open AccessPublished:November 09, 2021DOI:https://doi.org/10.1016/j.rbmo.2021.10.023

      Abstract

      Research question

      Can (mosaic) aneuploidy be reliably detected in preimplantation embryos after multiple displacement amplification and single nucleotide polymorphism detection, independent of haplotyping and copy number detection, with a new method 'analysis of parental contribution for aneuploidy detection' or 'APCAD'?

      Design

      This method is based on the maternal contribution, a parameter that reflects the proportion of DNA that is of maternal origin for a given chromosome or chromosome segment. A maternal contribution deviating from 50% for autosomes is strongly indicative of a (mosaic) chromosomal anomaly. The method was optimized using cell mixtures with varying ratios of euploid and aneuploid (47,XY,+21) lymphocytes. Next, the maternal contribution was retrospectively measured for all chromosomes from 349 Karyomapping samples.

      Results

      Retrospective analysis showed a skewed maternal contribution (<36.4 or >63.6%) in 57 out of 59 autosome meiotic trisomies and all autosome monosomies (n = 57), with values close to theoretical expectation. Thirty-two out of 7436 chromosomes, for which no anomalies had been observed with Karyomapping, showed a similarly skewed maternal contribution.

      Conclusions

      APCAD was used to measure the maternal contribution, which is an intuitive parameter independent of copy number detection. This method is useful for detecting copy number neutral anomalies and can confirm diagnosis of (mosaic) aneuploidy detected based on copy number. Mosaic and complete aneuploidy can be distinguished and the parent of origin for (mosaic) chromosome anomalies can be determined. Because of these benefits, the APCAD method has the potential to improve aneuploidy detection carried out by comprehensive preimplantation genetic testing methods.

      Introduction

      In recent years, comprehensive preimplantation genetic testing (PGT) methods have been developed (
      • Zamani Esteki M.
      • Dimitriadou E.
      • Mateiu L.
      • Melotte C.
      • Van der Aa N.
      • Kumar P.
      • Das R.
      • Theunis K.
      • Cheng J.
      • Legius E.
      • Moreau Y.
      • Debrock S.
      • D'Hooghe T.
      • Verdyck P.
      • De Rycke M.
      • Sermon K.
      • Vermeesch J.R.
      • Voet T.
      Concurrent whole-genome haplotyping and copy-number profiling of single cells.
      ;
      • Backenroth D.
      • Zahdeh F.
      • Kling Y.
      • Peretz A.
      • Rosen T.
      • Kort D.
      • Zeligson S.
      • Dror T.
      • Kirshberg S.
      • Burak E.
      • Segel R.
      • Levy-Lahad E.
      • Zangen D.
      • Altarescu G.
      • Carmi S.
      • Zeevi D.A.
      Haploseek: a 24-hour all-in-one method for preimplantation genetic diagnosis (PGD) of monogenic disease and aneuploidy.
      ;
      • Masset H.
      • Zamani Esteki M.
      • Dimitriadou E.
      • Dreesen J.
      • Debrock S.
      • Derhaag J.
      • Derks K.
      • Destouni A.
      • Drüsedau M.
      • Meekels J.
      • Melotte C.
      • Peeraer K.
      • Tšuiko O.
      • van Uum C.
      • Allemeersch J.
      • Devogelaere B.
      • François K.O.
      • Happe S.
      • Lorson D.
      • Richards R.L.
      • Theuns J.
      • Brunner H.
      • de Die-Smulders C.
      • Voet T.
      • Paulussen A.
      • Coonen E.
      • Vermeesch J.R.
      Multi-centre evaluation of a comprehensive preimplantation genetic test through haplotyping-by-sequencing.
      ;
      • Treff N.R.
      • Zimmerman R.
      • Bechor E.
      • Hsu J.
      • Rana B.
      • Jensen J.
      • Li J.
      • Samoilenko A.
      • Mowrey W.
      • Van Alstine J.
      • Leondires M.
      • Miller K.
      • Paganetti E.
      • Lello L.
      • Avery S.
      • Hsu S.
      • Melchior Tellier L.C.A.
      Validation of concurrent preimplantation genetic testing for polygenic and monogenic disorders, structural rearrangements, and whole and segmental chromosome aneuploidy with a single universal platform.
      ;
      • Chen S.
      • Yin X.
      • Zhang S.
      • Xia J.
      • Liu P.
      • Xie P.
      • Yan H.
      • Liang X.
      • Zhang J.
      • Chen Y.
      • Fei H.
      • Zhang L.
      • Hu Y.
      • Jiang H.
      • Lin G.
      • Chen F.
      • Xu C.
      Comprehensive preimplantation genetic testing by massively parallel sequencing.
      ;
      • Zeevi D.A.
      • Backenroth D.
      • Hakam-Spector E.
      • Renbaum P.
      • Mann T.
      • Zahdeh F.
      • Segel R.
      • Zeligson S.
      • Eldar-Geva T.
      • Ben-Ami I.
      • Ben-Yehuda A.
      • Carmi S.
      • Altarescu G.
      Expanded clinical validation of Haploseek for comprehensive preimplantation genetic testing.
      ). These methods can combine PGT for monogenic disorders (PGT-M) with PGT for structural rearrangements (PGT-SR) and PGT for aneuploidy (PGT-A). This is achieved by combining copy number detection with interrogation of genetic markers, such as single nucleotide polymorphisms (SNP) for haplotyping.
      In principle, PGT-A using comprehensive PGT (cPGT) methods are expected to be more accurate compared with dedicated PGT-A methods that only analyse copy number, such as low coverage massive parallel sequencing. The haplotype information obtained with the cPGT methods allows detection of haploidy, triploidy and uniparental disomy (UPD), which are not detectable using copy number profiles only. In addition, cPGT methods can identify copy number gains as meiotic if both haplotypes from one parent are detected. These trisomies and duplications are collectively called 'both parental homologue’ (BPH) anomalies, whereas 'single parental homologue' (SPH) anomalies cannot be observed by haplotyping and are considered mainly of mitotic origin (
      • Rabinowitz M.
      • Ryan A.
      • Gemelos G.
      • Hill M.
      • Baner J.
      • Cinnioglu C.
      • Banjevic M.
      • Potter D.
      • Petrov D.A.
      • Demko Z.
      Origins and rates of aneuploidy in human blastomeres.
      ;
      • Kubicek D.
      • Hornak M.
      • Horak J.
      • Navratil R.
      • Tauwinklova G.
      • Rubes J.
      • Vesela K.
      Incidence and origin of meiotic whole and segmental chromosomal aneuploidies detected by karyomapping.
      ). The distinction between SPH and BPH anomalies is important as the transfer of embryos with meiotic anomalies should be avoided whereas presumed mitotic anomalies, e.g. mosaic SPH trisomy, could be considered for transfer after counselling, depending on the involved chromosome or chromosome segment (
      • Greco E.
      • Minasi M.G.
      • Fiorentino F.
      Healthy Babies after Intrauterine Transfer of Mosaic Aneuploid Blastocysts.
      ;
      • Fragouli E.
      • Alfarawati S.
      • Spath K.
      • Babariya D.
      • Tarozzi N.
      • Borini A.
      • Wells D.
      Analysis of implantation and ongoing pregnancy rates following the transfer of mosaic diploid-aneuploid blastocysts.
      ;
      • Lledó B.
      • Morales R.
      • Ortiz J.A.
      • Blanca H.
      • Ten J.
      • Llácer J.
      • Bernabeu R.
      Implantation potential of mosaic embryos.
      ;
      • Munne S.
      • Blazek J.
      • Large M.
      • Martinez-Ortiz P.A.
      • Nisson H.
      • Liu E.
      • Tarozzi N.
      • Borini A.
      • Becker A.
      • Zhang J.
      • Maxwell S.
      • Grifo J.
      • Babariya D.
      • Wells D.
      • Fragouli E.
      Detailed investigation into the cytogenetic constitution and pregnancy outcome of replacing mosaic blastocysts detected with the use of high-resolution next-generation sequencing.
      ;
      • Spinella F.
      • Fiorentino F.
      • Biricik A.
      • Bono S.
      • Ruberti A.
      • Cotroneo E.
      • Baldi M.
      • Cursio E.
      • Minasi M.G.
      • Greco E.
      Extent of chromosomal mosaicism influences the clinical outcome of in vitro fertilization treatments.
      ;
      • Victor A.R.
      • Tyndall J.C.
      • Brake A.J.
      • Lepkowsky L.T.
      • Murphy A.E.
      • Griffin D.K.
      • McCoy R.C.
      • Barnes F.L.
      • Zouves C.G.
      • Viotti M.
      One hundred mosaic embryos transferred prospectively in a single clinic: exploring when and why they result in healthy pregnancies.
      ;
      • Zore T.
      • Kroener L.L.
      • Wang C.
      • Liu L.
      • Buyalos R.
      • Hubert G.
      • Shamonki M.
      Transfer of embryos with segmental mosaicism is associated with a significant reduction in live-birth rate.
      ;
      • Munné S.
      • Spinella F.
      • Grifo J.
      • Zhang J.
      • Beltran M.P.
      • Fragouli E.
      • Fiorentino F.
      Clinical outcomes after the transfer of blastocysts characterized as mosaic by high resolution Next Generation Sequencing- further insights.
      ;
      • Viotti M.
      Preimplantation Genetic Testing for Chromosomal Abnormalities: Aneuploidy, Mosaicism, and Structural Rearrangements.
      ;
      • Zhang Y.X.
      • Chen J.J.
      • Nabu S.
      • Yeung Q.S.Y.
      • Li Y.
      • Tan J.H.
      • Suksalak W.
      • Chanchamroen S.
      • Quangkananurug W.
      • Wong P.S.
      • Chung J.P.W.
      • Choy K.W.
      The Pregnancy Outcome of Mosaic Embryo Transfer: A Prospective Multicenter Study and Meta-Analysis.
      ).
      In practice, however, PGT-A using cPGT methods is more challenging compared with dedicated copy number based PGT-A. First, the haplotyping requires the analysis of genomic DNA of both partners and often DNA samples from one or more relatives. Second, a different type of whole genome amplification (WGA) is optimal for copy number detection compared with SNP interrogation. The commonly used hybrid methods combining an isothermal and polymerase chain reaction, e.g. Sureplex and Picoplex, are optimal for copy number detection but cause considerably more noise in the B-allele frequencies (BAF) of the analysed SNPs, which is problematic for haplotyping. On the other hand, isothermal multiple displacement amplification (MDA), e.g. Repli-G, is the WGA method of choice for haplotyping but causes noisier copy number profiles (
      • Gawad C.
      • Koh W.
      • Quake S.R.
      Single-cell genome sequencing: current state of the science.
      ;
      • Deleye L.
      • Gansemans Y.
      • De Coninck D.
      • Van Nieuwerburgh F.
      • Deforce D.
      Massively parallel sequencing of micro-manipulated cells targeting a comprehensive panel of disease-causing genes: A comparative evaluation of upstream whole-genome amplification methods.
      ). Because of this trade-off, cPGT methods strongly depend on well-designed bioinformatics tools to minimize noise, and optimally integrate available allele frequency, copy number and haplotype data for accurate and unambiguous detection of all types of chromosomal anomalies.
      Karyomapping, a clinically valid method for PGT-M (
      • Natesan S.A.
      • Bladon A.J.
      • Coskun S.
      • Qubbaj W.
      • Prates R.
      • Munne S.
      • Coonen E.
      • Dreesen J.C.
      • Stevens S.J.
      • Paulussen A.D.
      • Stock-Myer S.E.
      • Wilton L.J.
      • Jaroudi S.
      • Wells D.
      • Brown A.P.
      • Handyside A.H.
      Genome-wide karyomapping accurately identifies the inheritance of single-gene defects in human preimplantation embryos in vitro.
      ;
      • Natesan S.A.
      • Handyside A.H.
      • Thornhill A.R.
      • Ottolini C.S.
      • Sage K.
      • Summers M.C.
      • Konstantinidis M.
      • Wells D.
      • Griffin D.K.
      Live birth after PGD with confirmation by a comprehensive approach (karyomapping) for simultaneous detection of monogenic and chromosomal disorders.
      ), can arguably be considered as a cPGT method. Even though Karyomapping was initially not intended nor validated for PGT-A, aneuploidy can be detected based on the genome-wide copy number and BAF profiles, together with the generated SNP haplotypes (
      • Thornhill A.R.
      • Handyside A.H.
      • Ottolini C.
      • Natesan S.A.
      • Taylor J.
      • Sage K.
      • Harton G.
      • Cliffe K.
      • Affara N.
      • Konstantinidis M.
      • Wells D.
      • Griffin D.K.
      Karyomapping-a comprehensive means of simultaneous monogenic and cytogenetic PGD: comparison with standard approaches in real time for Marfan syndrome.
      ;
      • Ben-Nagi J.
      • Jones B.
      • Naja R.
      • Amer A.
      • Sunkara S.
      • SenGupta S.
      • Serhal P.
      Live birth rate is associated with oocyte yield and number of biopsied and suitable blastocysts to transfer in preimplantation genetic testing (PGT) cycles for monogenic disorders and chromosomal structural rearrangements.
      ;
      • Kubicek D.
      • Hornak M.
      • Horak J.
      • Navratil R.
      • Tauwinklova G.
      • Rubes J.
      • Vesela K.
      Incidence and origin of meiotic whole and segmental chromosomal aneuploidies detected by karyomapping.
      , unpublished results). Some anomalies, such as BPH trisomies and monosomies, are well detectable, but detection of others, such as SPH duplications and mosaic SPH trisomies, can be challenging as this is based on the visual interpretation of BAF, Log2R profiles only, or both. This is especially cumbersome for samples of a lower quality. To overcome these shortcomings, a new method named 'analysis of parental contribution for aneuploidy detection' or 'APCAD' was developed. This method makes use of a new variable, the maternal contribution (%Mat), which is calculated using the raw SNP data. It is a measure for the proportion of maternal alleles and hence the proportion of chromosome copies of maternal origin in the sample. Interestingly, this parameter provides a second dimension to the one-dimensional copy number or Log2R used by traditional methods (Figure 1). Note that each type of aneuploidy has a distinct position on this figure. Therefore, the %Mat can be used to confirm diagnosis based on copy number or Log2R and vice versa, resulting in a more reliable and intuitive diagnosis. In addition, the %Mat can be used to identify and quantify mosaicism in samples independent of Log2R.
      Figure 1
      Figure 1The relationship between copy number and maternal contribution (%Mat) for all relevant types of aneuploidy. Blue numbers (left) in each circle indicate the number of paternal chromosome copies, whereas the pink number (right) indicates the number of maternal copies per cell. Note that each type of aneuploidy has a distinct position on the chart. Mosaic disomy/trisomy and mosaic monosomy/disomy caused by a single mitotic event that occurred in a normal (disomic) embryo can be placed on the grey trajectories. The position depends on the percentage of aneuploid cells. With Karyomapping, copy number profiles are noisy and an exact copy number cannot be detected. In this situation, the measurement of %Mat can make the difference to reach a diagnosis. UPD, uniparental disomy.

      Materials and methods

      Hormonal stimulation, oocyte retrieval, intracytoplasmic sperm injection, embryo culture, biopsy and cryopreservation were as previously described (
      • De Rycke M.
      • De Vos A.
      • Belva F.
      • Berckmoes V.
      • Bonduelle M.
      • Buysse A.
      • Keymolen K.
      • Liebaers I.
      • Nekkebroeck J.
      • Verdyck P.
      • Verpoest W.
      Preimplantation genetic testing with HLA matching: from counseling to birth and beyond.
      ).

       Peripheral lymphocytes, genomic DNA and embryos

      The local Ethics Committee approved obtaining DNA and cell lines for the cell mixture experiment under number BUN 143201630618 on 22 February 2017. After informed consent from the parents was obtained, blood was drawn from two siblings and their parents. One sibling was affected with Down's syndrome (47,XY,+21; trisomy 21) and the other one was euploid (46,XY). Blood was collected in sodium heparin tubes for collection of lymphocytes and in EDTA tubes for DNA extraction. Fresh peripheral lymphocytes were isolated from whole blood and washed in phosphate buffered saline. For the cell mixture experiment, the fresh lymphocytes from the two siblings were mixed in different proportions, with a total of eight cells per tube. Four replicate series containing nine possible combinations were generated (0 + 8, 1 + 7, 2 + 6, 3 + 5, 4 + 4, 5 + 3, 6 + 2, 7 + 1, 8 + 0) (Table 1).
      TABLE 1COMBINATIONS OF TUBED CELLS WITH CALCULATED PERCENTAGE OF MATERNAL CONTRIBUTION (%MAT) FOR CHROMOSOME 21
      Cells 47,XY,+21 (mat), nCells46,XY, nCells with trisomy, %Maternal chr21 copies, nPaternal chr21 copies, nMaternal contribution (%Mat), n %
      080888/16 (50)
      1712.5989/17 (52.9)
      262510810/18 (55.6)
      3537.511811/19 (57.9)
      445012812/20 (60.0)
      5362.513813/21 (61.9)
      627514814/22 (63.6)
      7187.515815/23 (65.2)
      8010016816/24 (66.7)
      For retrospective analysis, 359 embryos were selected with trophectoderm biopsy between 1 September 2015 and 31 December 2017, which were diagnosed with Karyomapping as being not genetically transferable, for which the parents gave written informed consent and for which the SNP data had passed the quality-control cut-offs (call rate >85%, ≤1% miscall rate). Embryos that were tested for a monogenic disorder (PGT-M, majority), a structural chromosomal rearrangement (PGT-SR, minority), or both, were included. The available data from these embryos were re-analysed. For aneuploidy detection, the analysis of conventional Karyomapping data (raw BAF and Log2R profiles and haplotype information) was extended with the analysis of the BAF profiles of selected SNPs. Five embryos that were not diploid (haploid or triploid) and five embryos with low-level genome-wide maternal contamination were excluded, leaving 349 embryos from 98 couples for further analysis. As the SNP data were available after preimplantation genetic testing with Karyomapping, no additional intervention on the patient or embryo was required for this study. The local Ethics Committee also approved this study under number BUN 143201731745 on 22 March 2017.

       Whole genome amplification and single nucleotide polymorphisms array analysis

      Karyomapping workflow was used for WGA and SNP array analysis according to the manufacturer's recommendations (Vitrolife, Göteborg, Sweden).

       Statistical analysis and curve fitting

      R was used for statistical analysis. Generalized linear models were generated using the R Stats package.

      Results

       Principle of the analysis of parental contribution for aneuploidy detection method

      The selected SNPs were expected to be heterozygous in a euploid female sample: SNPs for which both biological parents show a homozygous SNP call, albeit for a different allele. These SNPs were called APCAD SNPs and were classified in category C1 or category C2 (Table 2) based on the observed parental genotypes. The SNPs on the Y chromosome were also added (Table 2).
      TABLE 2CATEGORIES OF SINGLE NUCLEOTIDE POLYMORPHISMS USED FOR MEASURING PARENTAL CONTRIBUTION
      Paternal genotype call
      Single nucleotide polymorphism genotype calls are written as AA, AB, BB or NC. ‘NC’ indicates no call or very low signal intensity (log2R <–4). Hemizygous SNPs A/– and B/–, are written as AA or BB, respectively, as hemizygosity and homozygosity cannot be distinguished.
      Maternal genotype call
      Single nucleotide polymorphism genotype calls are written as AA, AB, BB or NC. ‘NC’ indicates no call or very low signal intensity (log2R <–4). Hemizygous SNPs A/– and B/–, are written as AA or BB, respectively, as hemizygosity and homozygosity cannot be distinguished.
      Expected genotype sample
      Single nucleotide polymorphism genotype calls are written as AA, AB, BB or NC. ‘NC’ indicates no call or very low signal intensity (log2R <–4). Hemizygous SNPs A/– and B/–, are written as AA or BB, respectively, as hemizygosity and homozygosity cannot be distinguished.
      Allele frequency selected fromReflects contribution from
      Category C1
      Autosomes and X chromosomeBBAAABB-allele (BAF)Paternal
      Y chromosomeBBNCNC or BBB-allele (BAF)Paternal
      Category C2
      Autosomes and X chromosomeAABBABB-allele (BAF)Maternal
      Y chromosomeAANCNC or AAB-allele (BAF)Maternal
      a Single nucleotide polymorphism genotype calls are written as AA, AB, BB or NC. ‘NC’ indicates no call or very low signal intensity (log2R <–4). Hemizygous SNPs A/– and B/–, are written as AA or BB, respectively, as hemizygosity and homozygosity cannot be distinguished.
      For the category C1 SNPs, the B-allele is always of paternal origin. Hence, measuring BAF of C1 SNPs (BAFC1) can potentially be used to determine the percentage of paternal copies of a chromosome or chromosome segment (%Pat) in the analysed sample. This percentage is also referred to as the paternal contribution.
      Similarly, the B-allele for the category C2 SNPs is always of maternal origin and measuring BAF of C2 SNPs can potentially be used to determine the percentage of maternal copies of a chromosome or chromosome segment (%Mat) in the sample. This percentage is also referred to as the maternal contribution.
      For most types of aneuploidy, including trisomy, monosomy, uniparental disomy (UPD), and some forms of tetrasomy, the percentage of maternal alleles will differ from 50% (Figure 1). In the example (Figure 2), the measured BAF for the APCAD SNPs are shown for a male embryo with a mosaic trisomy 14, which was hardly detectable with Karyomapping. Even though there is considerable noise on the measured BAF values of individual SNPs, a shift in measured BAF can be observed for the population of C1 and C2 SNPs for chromosomes 14, X and Y.
      Figure 2
      Figure 2Results obtained with analysis of parental contribution for aneuploidy detection (APCAD) single nucleotide polymorphisms (SNP) from a male embryo with a mosaic trisomy 14. Chromosome 14 stands out when analysing APCAD SNPs (panels A to C). Panel A: the B-allele frequencies (BAF) for the APCAD SNPs is plotted against the genomic position in Mb. The BAF for these SNPs closely relates to the parental contribution. The paternal contribution is reflected by the BAF for the green SNPs (category C1) and the maternal contribution is reflected by the BAF for the orange SNPs (category C2). As log2R shows an increase in copy-number for chromosome 14 (not shown), the result is interpreted as a paternal (mosaic?) trisomy. The X is of maternal origin, whereas the Y is of paternal origin as can be expected for a male embryo; Panel B: boxplots for the data visualized in panel A. For each chromosome the boxplot for category C1 SNPs is shown left in green whereas the boxplot for the category C2 SNPs is shown right in orange. The parameter ‘delta_BAF’ for a chromosome equals the difference between the means of both categories represented by a white dot (meanBAF category C2 − meanBAF category C1). The delta_BAF is used to estimate the parental contribution and to estimate the degree of mosaicism (if applicable); Panel C: detail of panel A for chromosome 14. %Mat, the maternal contribution.

       Optimization of analysis of parental contribution for aneuploidy detection

      As the BAF of an individual SNP is too noisy when analysing DNA from a few cells (such as a trophectoderm biopsy), the SNPs located within a chromosome or chromosome segment were selected. Next, the mean BAF for SNP categories C1 and C2 were calculated separately (Figure 2). Frequently, however, it was observed that the measured BAF underestimated the true BAF in the sample. In multiple samples, the sum of the mean BAF of C1 SNPs (~%Pat/100) and mean BAF of C2 SNPs (~%Mat/100) was lower than 1 (0.95–1.00) for most if not all chromosomes (data not shown). As all DNA in the embryo is either maternal or paternal (%Mat + %Pat = 100%), this indicates that the measured BAF often underestimates the true BAF. This is possibly caused by differences in the detection of A and B fluorophores and the low amount of input material. This phenomenon is equally affecting the BAF of category C1 and C2 SNPs (data not shown). Hence, the difference between the mean BAF obtained for both categories (meanBAF category C2 − meanBAF category C1) was expected to be a more robust parameter compared with the mean BAF of category C1 (or C2) SNPs. This value, from here onwards referred to as the 'delta_BAF', was used for further analysis. It corresponds to the difference between the means (white diamond) shown in the middle of each box in panel B of Figure 2. The 'delta_BAF' is close to zero in euploid samples. A positive value is obtained if more maternal than paternal DNA is present, whereas a negative value indicates more paternal than maternal DNA. In the example shown in Figure 2, the delta_BAF is deviating with a value of –0.1882 for chromosome 14. This value, together with an observed increase in Log2R (data not shown) indicates a mosaic paternal trisomy for chromosome 14.
      The intention was to use the delta_BAF to measure the %Mat of a chromosome or chromosome segment. To this end, a cell mixture experiment was carried out with peripheral lymphocytes from two male siblings, of which one was affected with Down's syndrome (47,XY,+21) and the other one was euploid (46,XY). Four replicate series containing the nine possible contributions from each sibling were analysed (Table 1). After SNP array, the delta_BAF was calculated for all chromosomes. The results were plotted as the measured delta_BAF against the a-priori known percentage of maternal contribution (%Mat). In theory, for all autosomes except chromosome 21, the %Mat is 50% (no aneuploidy for these chromosomes, n = 756) whereas the %Mat for chromosome X is 100% (n = 36) in the male samples, and the %Mat for chromosome 21 is between 50 and 67% depending on the percentage of trisomy 21 cells that were tubed (n = 36).
      A clear relationship between the %Mat and the delta_BAF (data not shown) was observed and curve fitting was carried out. To assign a higher weight to a measurement when the underlying data were less noisy, the weight was calculated as the inverse of the SE for that delta_BAF value. The SE of the delta_BAF was calculated as the root of the sum of the squared SE of the mean of each category (=σC12/nC1+σC22/nC2). Curve fitting showed a better fit for a second order generalized linear model (Akaike information criterion [AIC] = –4046.8) over a first (AIC = –4016.4) order model (P < 0.001). A third order (AIC = –4044.9) model did not improve fit over the second order model (P = 0.76).
      The used whole genome amplification method, MDA (Qiagen Repli-G), generates large fragments of 10 to 100 kb. Therefore, the fit of the curve was examined to establish whether it improved after removing SNPs, which are clustered together and are undergoing the same amplification bias generated by the MDA. The fit improved after removing clustered SNPs with a minimal distance between consecutive SNPs between 10 and 35 kb based on the evaluation of R2 and between 10 and 20 kb based on the evaluation of AIC (Supplementary Figure 1). A minimum 20 kb distance was used in our subsequent analyses. The fitted curve was used for quantifying the %Mat in test samples from the observed delta_BAF (Figure 3).
      Figure 3
      Figure 3Relation between percentage maternal contribution (%Mat) and delta_BAF from the cell mixture experiment. Cells from a 47,XY,+21 cell line were tubed together with cells from a cell line from a sib with a 46,XY karyotype. The %Mat was calculated based on the input while the delta_BAF was calculated from the SNP array data after removal of closely clustered SNPs (<20kb). The central line corresponds to the best fit which was obtained using a second order generalized linear model. The 95% confidence interval for the model is also shown (outer full lines). The outer dashed lines indicate the prediction interval. Grey horizontal lines indicate cut-offs of delta_BAF at 0.0924 and 0.2341 corresponding with the estimated delta_BAF by the model for 25% and 75% of cells with trisomy respectively (%Mat equal to 55.6% and 63.6% respectively).
      Cut-offs were assigned at the boundaries delta_BAF = 0.0924 and delta_BAF = 0.2341, corresponding with the delta_BAF predicted by the model for 25% (or 55.6 %Mat) and 75% (or 63.6 %Mat) of cells with a maternal trisomy respectively (Figure 3). The 0.0924 cut-off is located above the 95% prediction interval for 50.0 %Mat (normal disomy) and the 0.2341 cut-off is located at the lower boundary of the prediction interval for 66.6 %Mat (complete trisomy). Below the threshold 0.0924, the value is considered normal (normal disomy). A value above 0.2341 indicates a full maternal trisomy. A value between both thresholds indicates a mosaic anomaly. With these cut-offs, all chromosomes of the training set with normal disomy (760/760) and complete trisomy (4/4) are correctly categorized. Also, chromosomes with trisomy in 50% of the cells (60 %Mat) are expected to be correctly categorized, as the prediction interval for 60 %Mat is located between both cut-off values. Chromosomes with low-level mosaic trisomy, however, are potentially categorized as disomy: three out of four samples with one trisomic cell (12.5% trisomy) and one out of four samples with two trisomic cells (25% trisomy) out of eight were categorized as normal disomy. Similarly, chromosomes with high-level mosaic trisomy can be labelled as a full trisomy with these cut-offs for the training set: one sample with 75% trisomic cells and three out of four samples with 87.5% trisomy would fall in the (full) trisomy category. All samples with 37.5, 50 and 62.5% trisomy were correctly categorized as mosaic trisomy. For detection of trisomies of paternal origin, the cut-offs used for maternal trisomies were mirrored (%Pat = 100 – %Mat) and cut-offs were assigned at the delta_BAF = –0.0924 and delta_BAF = –0.2341, corresponding to the delta_BAF predicted by the model for 25% (or 55.6 %Pat) and 75% (or 63.6 %Pat) of cells with a paternal trisomy, respectively.

       Retrospective analysis: validation of analysis of parental contribution for aneuploidy detection

      As a proof of principle, the APCAD method was applied to SNP array data from 349 embryos that had been generated with Karyomapping. The %Mat per chromosome was calculated from the delta_BAF based on the model described above for the 22 autosomes, X and Y, resulting in 8376 calculations in total. Between 116 and 1827 SNPs were available for calculating the maternal contribution of a chromosome, and each SNP category (C1 or C2) contained at least 47 SNPs. The relationship between the calculated %Mat per chromosome and the detected whole chromosome anomalies was investigated.
      An overview of the results for the autosomes is presented in Table 3. A %Mat higher than 63.6% in 56 out of 58 (96.6%) of the analysed maternal autosome BPH trisomies and %Mat lower than 36.4% for one paternal autosome BPH trisomy (31.0 %Mat) were detected. The median (66.1%) and mean (66.4%) maternal contribution for the maternal trisomies were close to the theoretical 66.6% (Figure 4). The %Mat values for chromosomes with a detected BPH trisomy were clearly different from the %Mat of chromosomes with presumed disomy (P = 2.2 × 10−16; Wilcoxon rank sum test). There were three outliers with 50.3, 58.5 and 77.7 %Mat (>1.5 interquartile range from first and third quartiles), indicating partial trisomy rescue in over 25% of cells (58.5 %Mat), complete trisomy rescue to disomy (50.3 %Mat) or trisomy rescue with loss of the paternal copy leading to more than 25% cells with UPD (77.7 %Mat).
      TABLE 3MATERNAL CONTRIBUTION (%MAT) OBSERVED FOR AUTOSOMES SPLIT BY THE TYPE OF CHROMOSOME ANOMALY DETECTED BY KARYOMAPPING
      Measured %MatChromosome anomaly
      Lower, %Upper, %TetrasomyBPH trisomy(mosaic) SPH trisomyMosaic monosomy(Mosaic) SegmentalMonosomyNo anomolies detected for this chromosomeTotal
      ≥0.0<27.30007154163
      ≥27.3<36.401*46201326
      ≥36.4<44.4005830117133
      ≥44.4≤55.6011312071367153
      >55.6≤63.60145230151184
      >63.6≤72.705593901894
      >72.7≤1001106143
      BPH trisomy or monosomy of paternal origin. BPH, both parental homologue; SPH, single parental homologue; %Mat, maternal contribution.
      025
      Total1592338645774367678
      a BPH trisomy or monosomy of paternal origin.BPH, both parental homologue; SPH, single parental homologue; %Mat, maternal contribution.
      Figure 4
      Figure 4The distribution of the maternal contribution (%Mat) observed for 58 autosomes with a maternal B-allele frequencies (BPH) trisomy (dark grey) and 7436 autosomes for which no clear chromosome anomaly is detected (no chromosomal anomaly was detected; light grey). Vertical black bars indicate cut-offs for normal disomy (between 44.4 and 55.6 %Mat) and maternal trisomy (>63.6 %Mat). Panel A: boxplots show that the %Mat of 58 autosomes with a maternal BPH trisomy is significantly different from the %Mat of chromosomes without detected anomalies (P = 2.2 × 10−16 with Wicoxon rank sum test); panel B: the maternal contribution for autosomes without detected anomalies approximates the shape of a normal distribution. Most (96%) of %Mat values is located between 44.4 and 55.6 %Mat (solid vertical lines); panel C: the same data topped at 20 counts. Maternal BPH trisomies show a maternal contribution of more than 63.6% in 56 out of 58 maternal BPH trisomies (96.6%). A %Mat between 63.6 and 72.7% (dashed vertical lines) is observed for 55 out of 58 maternal BPB trisomies (94.8%). NAD, no anomolies detected for.
      A skewed maternal contribution was also observed for all (57/57) autosome monosomies with a measured maternal contribution close to 0% for 54 maternal origin monosomies (0–3.3%, median 0.5%) and close to 100% for three paternal origin monosomies (values 97.9, 99.1 and 99.5 %Mat). For the single observed BPH tetrasomy, a %Mat value of 74.3% was observed, close to the theoretical 75% for a tetrasomy with three maternal copies (Table 3).
      As expected, chromosomes for which mosaic monosomy was observed showed a %Mat covering almost the complete spectrum with values between 7.8 and 86.8 %Mat (Table 3). With Karyomapping, a mosaic SPH trisomy cannot be distinguished from a SPH trisomy present in all cells of the sample. With cut-offs corresponding to over 75% aneuploid cells (63.6 to 72.7 %Mat or %Pat), 10 out of 23 SPH trisomies detected with Karyomapping were labelled as mosaic disomy/trisomy (Table 3).
      For 7436 autosomes, no chromosomal anomaly was detected with Karyomapping. The distribution of the %Mat of these chromosomes resembles a normal distribution at first sight (Figure 4) but outliers are overrepresented (4.6% of values >3 SD from mean). For 7136 out of 7436 (96.0%) chromosomes the %Mat was within the normal range, between 44.4% and 55.6% (Table 3). For 268 chromosomes, a parental (maternal or paternal) contribution between 55.6% and 63.6% was observed, a finding compatible with a mosaic, segmental chromosomal anomaly, or both, which was not detected based on the raw BAF and log2R profiles using Karyomapping. For 32 chromosomes from 13 samples a skewed parental contribution (>63.6%) was observed. These embryos are expected to be mosaic or aneuploid based on the %Mat.
      For 167 female samples, a %Mat between 44.4% and 55.6% was observed for 161 X chromosomes, whereas six out of 167 (3.7%) X chromosomes showed a %Mat or %Pat between 55.6 and 63.3% (data not shown). Detection of mosaicism for X or Y chromosomes in male samples using SNPs that are only located on X and Y, respectively, is not possible. Therefore, the %Mat was also analysed for the pseudoautosomal regions (PAR) even though these are relatively small regions (~3Mb) with only nine to 32 SNPs available for analysis. SNP categories (C1 and C2) contained each between three and 20 SNPs in all samples except two. In these two samples, no value for the parental contribution in the PAR was calculated. Given the small number of SNPs, the interquartile range for the measured %Mat for the PAR without detected aneuploidy for the sex chromosomes is higher compared with the autosomes (data not shown). A skewed parental contribution of more than 63.6%, however, was observed in all XXY trisomies (5/5), X chromosome monosomies (6/6), in a single instance of a terminal deletion on X (1/1) and in one out of three instances with a detected mosaic monosomy X (1/3) while no or limited skewing (%Mat between 36.4 and 63.6%) was observed for 328 out of 332 PAR regions for which no aneuploidy was detected with Karyomapping (data not shown).

      Discussion

       Optimization and validation of analysis of parental contribution for aneuploidy detection

      We have developed a new SNP-based approach for aneuploidy detection in embryos that involves the measurement of the %Mat and hence also the paternal contribution (100 – %Mat) for a chromosome in the sample. The relation between the delta_BAF (measured) and %Mat (known) was determined by mixing uncultured peripheral lymphocytes from a euploid (46,XY) and a Down's syndrome sibling (47,XY,+21) in different ratios, mimicking chromosomal mosaicism.
      Next, we retrospectively analysed an independent dataset of 349 embryonic trophectoderm samples. This analysis showed that the measured parental contribution for the different types of chromosomal anomalies was precise and accurate compared with the theoretically expected parental contribution. The median %Mat was 0.5% for maternal origin monosomies (theoretically 0%), 66.4% for maternal BPH trisomies (theoretically 66.7%) and 50.2% for chromosomes with normal disomy (theoretically 50%). It was 74.3% (theoretically 75%) for a single occurrence of a tetrasomy with three maternal copies and 31.0% for one occurring paternal BHP trisomy (theoretically 33.3%). We also showed that 96.0% of the autosomes with presumed disomy had a %Mat in the normal range between 44.4 and 55.6% whereas 55 out of 58 (94.8%) maternal BPH trisomies had a %Mat between 63.6 and 72.7%. For autosome monosomies the %Mat or %Pat was between 0 and 3.3% in all instances.
      A %Mat outside the normal range (>55 %Mat or %Pat) was observed for 4% (300/7436) of chromosomes for which no aneuploidy had been detected with Karyomapping. On the basis of the distribution of the %Mat values from the population of 7436 presumed normal chromosomes, these 300 values are outliers (>1.5 x interquartile range from first and third quartiles). Given the high prevalence of mosaicism in preimplantation embryos (
      • Vanneste E.
      • Voet T.
      • Caignec C.Le
      • Ampe M.
      • Konings P.
      • Melotte C.
      • Debrock S.
      • Amyere M.
      • Vikkula M.
      • Schuit F.
      • Fryns J.P.
      • Verbeke G.
      • D'Hooghe T.
      • Moreau Y.
      • Vermeesch J.R.
      Chromosome instability is common in human cleavage-stage embryos.
      ;
      • Mertzanidou A.
      • Wilton L.
      • Cheng J.
      • Spits C.
      • Vanneste E.
      • Moreau Y.
      • Vermeesch J.R.
      • Sermon K.
      Microarray analysis reveals abnormal chromosomal complements in over 70% of 14 normally developing human embryos.
      ,
      • Viotti M.
      Preimplantation Genetic Testing for Chromosomal Abnormalities: Aneuploidy, Mosaicism, and Structural Rearrangements.
      ) and the limited signal to noise ratio of the Log2R profile in Karyomapping, it can be expected that most of these outliers are caused by chromosomal mosaicism.

       Benefits and limitations of analysis of parental contribution for aneuploidy detection

      The APCAD method requires high-quality SNP genotypes from both partners together with raw SNP allele frequency values from embryos. DNA from family members (references) is not required. As these data are readily available from the Karyomapping procedure, or any other cPGT method, our analysis requires no additional wet lab work. For a typical PGT cycle with up to six embryos, APCAD results can be available within half an hour. The APCAD has the advantage that the maternal contribution can be measured without any assumptions. It is not influenced by the number of different maternal or paternal haplotypes present in the sample, e.g. SPH or BPH trisomy, the availability of reference samples or the location of recombination events.
      With APCAD, we were able to identify 32 chromosomes with a skewed maternal contribution (%Mat or %Pat >63.6%) for which no clear shift in Log2R had been observed by Karyomapping. These embryos are expected to be mosaic (>43% mosaic monosomy) or aneuploid (>75% SPH trisomy). Caution is recommended when considering these embryos for transfer.
      If a shift in copy number or Log2R is observed, the measured %Mat can be used to determine the (minimal) proportion of aneuploid cells. In this way mosaic trisomy (between 25 and 75% trisomic cells) and complete trisomy (>75% trisomic cells) can be distinguished. Also, the parental origin of this (mosaic) copy number gain or loss can be determined (Table 1). As expected, mosaicism was more frequently observed for SPH trisomies (10/23 [43.5%]), which are considered mainly of mitotic origin compared with BPH trisomies, which arise during meiosis (3/59; 5.1%). The additional chromosome copy was also more frequently of paternal origin in SPH trisomies (9/23 [39.1%]) compared with BPH trisomies (1/59 [1.7%]), further confirming that they are mainly of mitiotic origin. The detection of mosaic SPH trisomies is clinically relevant as mosaic embryos have been reported to have a lower implantation rate and higher miscarriage rate but can lead to the birth of an apparently healthy child. These embryos can be considered for transfer after counselling, depending on the chromosome involved (
      • Greco E.
      • Minasi M.G.
      • Fiorentino F.
      Healthy Babies after Intrauterine Transfer of Mosaic Aneuploid Blastocysts.
      ;
      • Fragouli E.
      • Alfarawati S.
      • Spath K.
      • Babariya D.
      • Tarozzi N.
      • Borini A.
      • Wells D.
      Analysis of implantation and ongoing pregnancy rates following the transfer of mosaic diploid-aneuploid blastocysts.
      ;
      • Lledó B.
      • Morales R.
      • Ortiz J.A.
      • Blanca H.
      • Ten J.
      • Llácer J.
      • Bernabeu R.
      Implantation potential of mosaic embryos.
      ;
      • Munne S.
      • Blazek J.
      • Large M.
      • Martinez-Ortiz P.A.
      • Nisson H.
      • Liu E.
      • Tarozzi N.
      • Borini A.
      • Becker A.
      • Zhang J.
      • Maxwell S.
      • Grifo J.
      • Babariya D.
      • Wells D.
      • Fragouli E.
      Detailed investigation into the cytogenetic constitution and pregnancy outcome of replacing mosaic blastocysts detected with the use of high-resolution next-generation sequencing.
      ;
      • Spinella F.
      • Fiorentino F.
      • Biricik A.
      • Bono S.
      • Ruberti A.
      • Cotroneo E.
      • Baldi M.
      • Cursio E.
      • Minasi M.G.
      • Greco E.
      Extent of chromosomal mosaicism influences the clinical outcome of in vitro fertilization treatments.
      ;
      • Victor A.R.
      • Tyndall J.C.
      • Brake A.J.
      • Lepkowsky L.T.
      • Murphy A.E.
      • Griffin D.K.
      • McCoy R.C.
      • Barnes F.L.
      • Zouves C.G.
      • Viotti M.
      One hundred mosaic embryos transferred prospectively in a single clinic: exploring when and why they result in healthy pregnancies.
      ;
      • Zore T.
      • Kroener L.L.
      • Wang C.
      • Liu L.
      • Buyalos R.
      • Hubert G.
      • Shamonki M.
      Transfer of embryos with segmental mosaicism is associated with a significant reduction in live-birth rate.
      ;
      • Munné S.
      • Spinella F.
      • Grifo J.
      • Zhang J.
      • Beltran M.P.
      • Fragouli E.
      • Fiorentino F.
      Clinical outcomes after the transfer of blastocysts characterized as mosaic by high resolution Next Generation Sequencing- further insights.
      ,
      • Zhang Y.X.
      • Chen J.J.
      • Nabu S.
      • Yeung Q.S.Y.
      • Li Y.
      • Tan J.H.
      • Suksalak W.
      • Chanchamroen S.
      • Quangkananurug W.
      • Wong P.S.
      • Chung J.P.W.
      • Choy K.W.
      The Pregnancy Outcome of Mosaic Embryo Transfer: A Prospective Multicenter Study and Meta-Analysis.
      ).
      The observation that three BPH trisomies were categorized as mosaic is particularly interesting. For two maternal BPH trisomies a %Mat lower than 63.6% was observed, compatible with the presence of more than 25% of cells with normal disomy. In one sample, the %Mat was markedly increased, compatible with the presence of more than 25% cells with UPD in the sample. Complete UPD was not observed. Trisomy rescue is a known mechanism causing UPD (
      • Eggermann T.
      • Soellner L.
      • Buiting K.
      • Kotzot D.
      Mosaicism and uniparental disomy in prenatal diagnosis.
      ), and UPD has been observed in blastocysts, albeit rarely (
      • Gueye N.A.
      • Devkota B.
      • Taylor D.
      • Pfundt R.
      • Scott R.T.
      • Treff N.R.
      Uniparental disomy in the human blastocyst is exceedingly rare.
      ). Our results, and those of others (
      • Handyside A.H.
      • McCollin A.
      • Summers M.C.
      • Ottolini C.S.
      Copy number analysis of meiotic and postzygotic mitotic aneuploidies in trophectoderm cells biopsied at the blastocyst stage and arrested embryos.
      ), show the first evidence for partial trisomy rescue in trophectoderm biopsies from blastocysts.
      The relationship between %Mat and the measured delta_BAF was established using samples with 50%Mat (disomy for chromosomes 1–20 and 22), between 50 and 66.7 %Mat (trisomy and mosaic trisomy 21) and with 100%Mat (X chromosome in male sample). No samples had a %Mat between 66.7 and 100%. Therefore, the measurement of %Mat in this range is potentially less accurate.
      A limitation of the current method, similar to haplotyping methods, is that aneuploidy may be more difficult to detect in case of consanguinity. If both partners share a haplotype that is identical by descent, the parents will never have SNPs with a different homozygous genotype (maternal AA and paternal BB or vice versa). Hence, the parental contribution cannot be measured for this region using the method described here.

       Future prospects

      In this study, the %Mat was calculated for complete chromosomes. We are also, however, developing a higher resolution approach to detect segmental anomalies. The observation that a %Mat or %Pat of more than 63.6% was observed for the 3Mb PAR regions in 16 out of 18 samples with detected sex chromosome anomalies while no or limited skewing (%Mat between 36.4% and 63.6%) was observed for the PAR region in 328 out of 332 samples without detected sex chromosome aneuploidy is promising in this sense.
      Ideally, APCAD will be combined with a method for accurate copy number detection, e.g. by massive parallel sequencing, for highly accurate aneuploidy detection. The integration of the proposed parameters %Mat and %Pat and copy number will allow the maternal copy number (CNmat) and paternal copy number (CNpat) to be determined for a chromosome or chromosome segment in the sample (CNmat =CN x %Mat/100; CNpat =CN x %Pat/100). These CNmat and CNpat have the potential to become an intuitive readout for aneuploidy detection in future applications.
      In conclusion, the calculation of the maternal contribution provides a useful universal tool for aneuploidy detection. The maternal contribution can be measured without any assumptions. It is not influenced by the number of different maternal or paternal haplotypes present in the sample, e.g. SPH or BPH trisomy, the availability of reference samples or the location of recombination events. It is complementary and independent of copy number (Figure 1). The method can be combined with Karyomapping (Vitrolife, Göteborg, Sweden) as demonstrated in this study, as well as with other currently applied cPGT methods, such as OnePGT (Agilent, Santa Clara, CA, USA), Haploseek or WGS for improved detection of (mosaic) aneuploidy and intuitive interpretation.

      Acknowledgements

      We would like to thank the patients who participated in this study. We are also grateful to all colleagues of the centres of medical genetics and reproductive medicine of the UZ Brussel for their direct or indirect contribution to this project and to the PGT programme in general. The APCAD method discussed in this study is subject of patent application WO2021180722.

      Appendix. Supplementary materials

      References

        • Backenroth D.
        • Zahdeh F.
        • Kling Y.
        • Peretz A.
        • Rosen T.
        • Kort D.
        • Zeligson S.
        • Dror T.
        • Kirshberg S.
        • Burak E.
        • Segel R.
        • Levy-Lahad E.
        • Zangen D.
        • Altarescu G.
        • Carmi S.
        • Zeevi D.A.
        Haploseek: a 24-hour all-in-one method for preimplantation genetic diagnosis (PGD) of monogenic disease and aneuploidy.
        Genet. Med. 2019; 21: 1390-1399
        • Ben-Nagi J.
        • Jones B.
        • Naja R.
        • Amer A.
        • Sunkara S.
        • SenGupta S.
        • Serhal P.
        Live birth rate is associated with oocyte yield and number of biopsied and suitable blastocysts to transfer in preimplantation genetic testing (PGT) cycles for monogenic disorders and chromosomal structural rearrangements.
        Eur. J. Obstet. Gynecol. Reprod. Biol. X. 2019; 4: 100055
        • Chen S.
        • Yin X.
        • Zhang S.
        • Xia J.
        • Liu P.
        • Xie P.
        • Yan H.
        • Liang X.
        • Zhang J.
        • Chen Y.
        • Fei H.
        • Zhang L.
        • Hu Y.
        • Jiang H.
        • Lin G.
        • Chen F.
        • Xu C.
        Comprehensive preimplantation genetic testing by massively parallel sequencing.
        Hum. Reprod. 2021; 36: 236-247
        • De Rycke M.
        • De Vos A.
        • Belva F.
        • Berckmoes V.
        • Bonduelle M.
        • Buysse A.
        • Keymolen K.
        • Liebaers I.
        • Nekkebroeck J.
        • Verdyck P.
        • Verpoest W.
        Preimplantation genetic testing with HLA matching: from counseling to birth and beyond.
        J. Hum. Genet. 2020;
        • Deleye L.
        • Gansemans Y.
        • De Coninck D.
        • Van Nieuwerburgh F.
        • Deforce D.
        Massively parallel sequencing of micro-manipulated cells targeting a comprehensive panel of disease-causing genes: A comparative evaluation of upstream whole-genome amplification methods.
        PLoS One. 2018; 13e0196334
        • Eggermann T.
        • Soellner L.
        • Buiting K.
        • Kotzot D.
        Mosaicism and uniparental disomy in prenatal diagnosis.
        Trends Mol. Med. 2015; 21: 77-87
        • Fragouli E.
        • Alfarawati S.
        • Spath K.
        • Babariya D.
        • Tarozzi N.
        • Borini A.
        • Wells D.
        Analysis of implantation and ongoing pregnancy rates following the transfer of mosaic diploid-aneuploid blastocysts.
        Hum. Genet. 2017; 136: 805-819
        • Gawad C.
        • Koh W.
        • Quake S.R.
        Single-cell genome sequencing: current state of the science.
        Nat. Rev. Genet. 2016; 17: 175-188
        • Greco E.
        • Minasi M.G.
        • Fiorentino F.
        Healthy Babies after Intrauterine Transfer of Mosaic Aneuploid Blastocysts.
        N. Engl. J. Med. 2015; 373: 2089-2090
        • Gueye N.A.
        • Devkota B.
        • Taylor D.
        • Pfundt R.
        • Scott R.T.
        • Treff N.R.
        Uniparental disomy in the human blastocyst is exceedingly rare.
        Fertil. Steril. 2014; 101: 232-236
        • Handyside A.H.
        • McCollin A.
        • Summers M.C.
        • Ottolini C.S.
        Copy number analysis of meiotic and postzygotic mitotic aneuploidies in trophectoderm cells biopsied at the blastocyst stage and arrested embryos.
        Prenat. Diagn. 2021; 41: 525-535
        • Kubicek D.
        • Hornak M.
        • Horak J.
        • Navratil R.
        • Tauwinklova G.
        • Rubes J.
        • Vesela K.
        Incidence and origin of meiotic whole and segmental chromosomal aneuploidies detected by karyomapping.
        Reprod. Biomed. Online. 2019; 38: 330-339
        • Lledó B.
        • Morales R.
        • Ortiz J.A.
        • Blanca H.
        • Ten J.
        • Llácer J.
        • Bernabeu R.
        Implantation potential of mosaic embryos.
        Syst. Biol. Reprod. Med. 2017; 63: 206-208
        • Masset H.
        • Zamani Esteki M.
        • Dimitriadou E.
        • Dreesen J.
        • Debrock S.
        • Derhaag J.
        • Derks K.
        • Destouni A.
        • Drüsedau M.
        • Meekels J.
        • Melotte C.
        • Peeraer K.
        • Tšuiko O.
        • van Uum C.
        • Allemeersch J.
        • Devogelaere B.
        • François K.O.
        • Happe S.
        • Lorson D.
        • Richards R.L.
        • Theuns J.
        • Brunner H.
        • de Die-Smulders C.
        • Voet T.
        • Paulussen A.
        • Coonen E.
        • Vermeesch J.R.
        Multi-centre evaluation of a comprehensive preimplantation genetic test through haplotyping-by-sequencing.
        Hum. Reprod. 2019; 34: 1608-1619
        • Mertzanidou A.
        • Wilton L.
        • Cheng J.
        • Spits C.
        • Vanneste E.
        • Moreau Y.
        • Vermeesch J.R.
        • Sermon K.
        Microarray analysis reveals abnormal chromosomal complements in over 70% of 14 normally developing human embryos.
        Hum. Reprod. 2013; 28: 256-264
        • Munne S.
        • Blazek J.
        • Large M.
        • Martinez-Ortiz P.A.
        • Nisson H.
        • Liu E.
        • Tarozzi N.
        • Borini A.
        • Becker A.
        • Zhang J.
        • Maxwell S.
        • Grifo J.
        • Babariya D.
        • Wells D.
        • Fragouli E.
        Detailed investigation into the cytogenetic constitution and pregnancy outcome of replacing mosaic blastocysts detected with the use of high-resolution next-generation sequencing.
        Fertil. Steril. 2017; 108 (e8): 62-71
        • Munné S.
        • Spinella F.
        • Grifo J.
        • Zhang J.
        • Beltran M.P.
        • Fragouli E.
        • Fiorentino F.
        Clinical outcomes after the transfer of blastocysts characterized as mosaic by high resolution Next Generation Sequencing- further insights.
        Eur. J. Med. Genet. 2020; 63103741
        • Natesan S.A.
        • Bladon A.J.
        • Coskun S.
        • Qubbaj W.
        • Prates R.
        • Munne S.
        • Coonen E.
        • Dreesen J.C.
        • Stevens S.J.
        • Paulussen A.D.
        • Stock-Myer S.E.
        • Wilton L.J.
        • Jaroudi S.
        • Wells D.
        • Brown A.P.
        • Handyside A.H.
        Genome-wide karyomapping accurately identifies the inheritance of single-gene defects in human preimplantation embryos in vitro.
        Genet. Med. 2014; 16: 838-845
        • Natesan S.A.
        • Handyside A.H.
        • Thornhill A.R.
        • Ottolini C.S.
        • Sage K.
        • Summers M.C.
        • Konstantinidis M.
        • Wells D.
        • Griffin D.K.
        Live birth after PGD with confirmation by a comprehensive approach (karyomapping) for simultaneous detection of monogenic and chromosomal disorders.
        Reprod. Biomed. Online. 2014; 29: 600-605
        • Rabinowitz M.
        • Ryan A.
        • Gemelos G.
        • Hill M.
        • Baner J.
        • Cinnioglu C.
        • Banjevic M.
        • Potter D.
        • Petrov D.A.
        • Demko Z.
        Origins and rates of aneuploidy in human blastomeres.
        Fertil. Steril. 2012; 97: 395-401
        • Spinella F.
        • Fiorentino F.
        • Biricik A.
        • Bono S.
        • Ruberti A.
        • Cotroneo E.
        • Baldi M.
        • Cursio E.
        • Minasi M.G.
        • Greco E.
        Extent of chromosomal mosaicism influences the clinical outcome of in vitro fertilization treatments.
        Fertil. Steril. 2018; 109: 77-83
        • Thornhill A.R.
        • Handyside A.H.
        • Ottolini C.
        • Natesan S.A.
        • Taylor J.
        • Sage K.
        • Harton G.
        • Cliffe K.
        • Affara N.
        • Konstantinidis M.
        • Wells D.
        • Griffin D.K.
        Karyomapping-a comprehensive means of simultaneous monogenic and cytogenetic PGD: comparison with standard approaches in real time for Marfan syndrome.
        J. Assist. Reprod. Genet. 2015; 32: 347-563
        • Treff N.R.
        • Zimmerman R.
        • Bechor E.
        • Hsu J.
        • Rana B.
        • Jensen J.
        • Li J.
        • Samoilenko A.
        • Mowrey W.
        • Van Alstine J.
        • Leondires M.
        • Miller K.
        • Paganetti E.
        • Lello L.
        • Avery S.
        • Hsu S.
        • Melchior Tellier L.C.A.
        Validation of concurrent preimplantation genetic testing for polygenic and monogenic disorders, structural rearrangements, and whole and segmental chromosome aneuploidy with a single universal platform.
        Eur. J. Med. Genet. 2019; 62103647
        • Vanneste E.
        • Voet T.
        • Caignec C.Le
        • Ampe M.
        • Konings P.
        • Melotte C.
        • Debrock S.
        • Amyere M.
        • Vikkula M.
        • Schuit F.
        • Fryns J.P.
        • Verbeke G.
        • D'Hooghe T.
        • Moreau Y.
        • Vermeesch J.R.
        Chromosome instability is common in human cleavage-stage embryos.
        Nat. Med. 2009; 15: 577-583
        • Victor A.R.
        • Tyndall J.C.
        • Brake A.J.
        • Lepkowsky L.T.
        • Murphy A.E.
        • Griffin D.K.
        • McCoy R.C.
        • Barnes F.L.
        • Zouves C.G.
        • Viotti M.
        One hundred mosaic embryos transferred prospectively in a single clinic: exploring when and why they result in healthy pregnancies.
        Fertil. Steril. 2019; 111: 280-293
        • Viotti M.
        Preimplantation Genetic Testing for Chromosomal Abnormalities: Aneuploidy, Mosaicism, and Structural Rearrangements.
        Genes (Basel). 2020; : 11
        • Zamani Esteki M.
        • Dimitriadou E.
        • Mateiu L.
        • Melotte C.
        • Van der Aa N.
        • Kumar P.
        • Das R.
        • Theunis K.
        • Cheng J.
        • Legius E.
        • Moreau Y.
        • Debrock S.
        • D'Hooghe T.
        • Verdyck P.
        • De Rycke M.
        • Sermon K.
        • Vermeesch J.R.
        • Voet T.
        Concurrent whole-genome haplotyping and copy-number profiling of single cells.
        Am. J. Hum. Genet. 2015; 96: 894-912
        • Zeevi D.A.
        • Backenroth D.
        • Hakam-Spector E.
        • Renbaum P.
        • Mann T.
        • Zahdeh F.
        • Segel R.
        • Zeligson S.
        • Eldar-Geva T.
        • Ben-Ami I.
        • Ben-Yehuda A.
        • Carmi S.
        • Altarescu G.
        Expanded clinical validation of Haploseek for comprehensive preimplantation genetic testing.
        Genet. Med. 2021; 23: 1334-1340
        • Zhang Y.X.
        • Chen J.J.
        • Nabu S.
        • Yeung Q.S.Y.
        • Li Y.
        • Tan J.H.
        • Suksalak W.
        • Chanchamroen S.
        • Quangkananurug W.
        • Wong P.S.
        • Chung J.P.W.
        • Choy K.W.
        The Pregnancy Outcome of Mosaic Embryo Transfer: A Prospective Multicenter Study and Meta-Analysis.
        Genes (Basel). 2020; : 11
        • Zore T.
        • Kroener L.L.
        • Wang C.
        • Liu L.
        • Buyalos R.
        • Hubert G.
        • Shamonki M.
        Transfer of embryos with segmental mosaicism is associated with a significant reduction in live-birth rate.
        Fertil. Steril. 2019; 111: 69-76

      Biography

      Pieter Verdyck obtained a PhD in human genetics in 2005 and has worked as a recognized genetic laboratory supervisor and molecular geneticist at the Centre for Medical Genetics of Universitair Ziekenhuis Brussel (UZ Brussel) since 2009. He is mainly active in PGT-M, PGT-SR and PGT-A.
      Key message
      A new method for reliable aneuploidy detection was developed independent of haplotype and copy number detection. In addition, it is useful to determine the parental origin and to estimate the proportion of aneuploid cells for mosaic copy number gains and losses.