De novo deletions and duplications detected by array CGH: a study of parental origin in relation to mechanisms of formation and size of imbalance

Advanced maternal age: copy number variations and pregnancy outcomes

Objective: Adverse pregnancy outcomes are closely related to advanced maternal age (AMA; age at pregnancy ≥35 years). Little research has been reported on aneuploid abnormalities and pathogenic copy number variations (CNVs) affecting pregnancy outcomes in women with AMA. The purpose of this study was to assess CNVs associated with AMA in prenatal diagnosis to determine the characteristics of pathogenic CNVs and assist with genetic counseling of women with AMA. Methods: Among 277 fetuses of women with AMA, 218 (78.7%) were isolated AMA fetuses and 59 (21.3%) were non-isolated AMA fetuses and showed ultrasound anomalies from January 2021 to October 2022. Isolated AMA was defined as AMA cases without sonographic abnormalities. Non-isolated AMA was defined as AMA cases with sonographic abnormalities such as sonographic soft markers, widening of the lateral ventricles, or extracardiac structural anomalies. The amniotic fluid cells underwent routine karyotyping followed by single nucleotide polymorphism array (SNP-array) analysis. Results: Of the 277 AMA cases, karyotype analysis identified 20 chromosomal abnormalities. As well as 12 cases of chromosomal abnormalities corresponded to routine karyotyping, the SNP array identified an additional 14 cases of CNVs with normal karyotyping results. There were five pathogenetic CNVs, seven variations of uncertain clinical significance (VOUS), and two benign CNVs. The detection rate of abnormal CNVs in non-isolated AMA cases was increasing (13/59; 22%) than in isolated AMA cases (13/218; 5.96%) (p < 0.001). We also determined that pathogenic CNVs affected the rate of pregnancy termination in women with AMA. Conclusion: Aneuploid abnormalities and pathogenic CNVs affect pregnancy outcomes in women with AMA. SNP array had a higher detection rate of genetic variation than did karyotyping and is an important supplement to karyotype analysis, which enables better informed clinical consultation and clinical decision-making.

Have maternal or paternal ages any impact on the prenatal incidence of genomic copy number variants associated with fetal structural anomalies?

The objective of this study was to determine whether maternal or paternal ages have any impact on the prenatal incidence of genomic copy number variants (CNV) in fetuses with structural anomalies. We conducted a non-paired case-control study (1:2 ratio) among pregnancies undergoing chromosomal microarray analysis (CMA) because of fetal ultrasound anomalies, from December 2012 to May 2020. Pregnancies with any pathogenic copy number variant (CNV), either microdeletion or microduplication, were defined as cases. Controls were selected as the next two pregnancies with the same indication for CMA but with a normal result. Logistic regression was used, adjusting by use of assisted reproductive technology (ART) and parental smoking. Stratified analysis was performed according to CNV type (de novo/inherited and recurrent/non-recurrent). The study included 189 pregnancies: 63 cases and 126 controls. Mean maternal age in cases was 33.1 (SD 4.6) years and 33.9 (SD 6.0) years in controls. Mean paternal mean age was 34.5 (SD 4.8) years in cases and 35.8 (SD 5.8) years in controls. No significant differences in maternal or paternal age were observed, neither in stratified analysis according to the CNV type. Moreover, the proportion of cases were not significantly different between non-advanced and advanced ages, either considering paternal or maternal ages. The presence of pathogenic CNV at CMA in fetuses with structural anomalies was not found to be associated with advanced paternal or maternal age.

De novo copy number variants and parental age: Is there an association?

PURPOSE

To investigate whether increased parental age is associated with an increased risk for de novo copy number variant (CNV) formation in offspring.

METHODS

CNV calls from 2323 individuals referred to Signature Genomic Laboratories for clinical microarray-based comparative genomic hybridization were investigated; 17% of the samples were prenatal and 83% were postnatal. The de novo CNV data were further split into de novo CNVs bound by low copy repeats (LCRs) and those not bound by LCRs.

RESULTS

No association was found between CNV occurrence and paternal age in both the prenatal (p = 0.6795) and postnatal (p = 0.1741) cohorts. Maternal age was significantly higher with de novo CNV occurrence in our postnatal cohort (p = 0.0126), an effect which may be driven by formation of de novo CNVs that are bound by LCRs (p = 0.0026). Furthermore, a significant positive correlation was observed between maternal age and de novo CNVs (Point-Biserial R² = 0.0503, p = 0.0152).

CONCLUSIONS

This large-scale study did not find any evidence for the influence of increased paternal age on de novo CNV formation, while increased maternal age appeared to increase risk for de novo, non-complex CNV occurrence in offspring with intellectual disability/developmental delay. Further studies and continued technological advances will help yield more information on the risk factors for de novo CNVs.

A retrospective analysis of the prevalence of imprinting disorders in Estonia from 1998 to 2016

Imprinting disorders (ImpDis) represent a small group of rare congenital diseases primarily affecting growth, development, and the hormonal and metabolic systems. The aim of present study was to identify the prevalence of the ImpDis in Estonia, to describe trends in the live birth prevalence of these disorders between 1998 and 2016, and to compare the results with previously published data. We retrospectively reviewed the records of all Estonian patients since 1998 with both molecularly and clinically diagnosed ImpDis. A prospective study was also conducted, in which all patients with clinical suspicion for an ImpDis were molecularly analyzed. Eighty-seven individuals with ImpDis were identified. Twenty-seven (31%) of them had Prader-Willi syndrome (PWS), 15 (17%) had Angelman syndrome (AS), 15 (17%) had Silver-Russell syndrome (SRS), 12 (14%) had Beckwith-Wiedemann syndrome (BWS), 10 (11%) had pseudo- or pseudopseudohypoparathyroidism, four had central precocious puberty, two had Temple syndrome, one had transient neonatal diabetes mellitus, and one had myoclonus-dystonia syndrome. One third of SRS and BWS cases fulfilled the diagnostic criteria for these disorders, but tested negative for genetic abnormalities. Seventy-six individuals were alive as of January 1, 2018, indicating the total prevalence of ImpDis in Estonia is 5.8/100,000 (95% CI 4.6/100,000-7.2/100,000). The minimum live birth prevalence of all ImpDis in Estonia in 2004-2016 was 1/3,462, PWS 1/13,599, AS 1/27,198, BWS 1/21,154, SRS 1/15,866, and PHP/PPHP 1/27,198. Our results are only partially consistent with previously published data. The worldwide prevalence of SRS and GNAS-gene-related ImpDis is likely underestimated and may be at least three times higher than expected.

Prenatal diagnosis of a rare de novo 1q22-q25.1 chromosomal deletion syndrome using oligo array CGH

We present prenatal diagnosis of a case with a rare de novo interstitial deletion of 1q21-q25.1 by oligo array CGH and provide detailed information on unbalanced gene content and the breakpoints. The affected fetus was delivered at 37 weeks' gestation with a unique clinical phenotype.

Pronounced maternal parent-of-origin bias for type-1 NF1 microdeletions

Neurofibromatosis type 1 (NF1) is caused, in 4.7-11% of cases, by large deletions encompassing the NF1 gene and its flanking regions within 17q11.2. Different types of large NF1 deletion occur which are distinguishable by their breakpoint location and underlying mutational mechanism. Most common are the type-1 NF1 deletions of 1.4 Mb which exhibit recurrent breakpoints caused by nonallelic homologous recombination (NAHR), also termed unequal crossover. Here, we analyzed 37 unrelated families of patients with de novo type-1 NF1 deletions by means of short tandem repeat (STR) profiling to determine the parental origin of the deletions. We observed that 33 of the 37 type-1 deletions were of maternal origin (89.2% of cases; p < 0.0001). Analysis of the patients' siblings indicated that, in 14 informative cases, ten (71.4%) deletions resulted from interchromosomal unequal crossover during meiosis I. Our findings indicate a strong maternal parent-of-origin bias for type-1 NF1 deletions. A similarly pronounced maternal transmission bias has been reported for recurrent copy number variants (CNVs) within 16p11.2 associated with autism, but not so far for any other NAHR-mediated pathogenic CNVs. Region-specific genomic features are likely to be responsible for the maternal bias in the origin of both the 16p11.2 CNVs and type-1 NF1 deletions.

Mosaicism for structural non-centromeric autosomal rearrangements in disease-defined carriers: sex differences in the rearrangements profile and maternal age distributions

Background

Mosaicism for an autosomal structural rearrangement (Rea) associated with clinical manifestation of chromosomal imbalance is rare. Consequently, there is a lack of basic epidemiological characterization of this kind of mosaicism, such as population rate, cytogenetic profile of Reas involved, maternal age distribution, and sex (male to female) ratio among Rea carriers. The objectives of the present study were: (i) determination of the Rea profile in clinically affected individuals, (ii) comparative analysis of the cytogenetic profile and involvement of single chromosomes to rearrangements in affected and previously reported asymptomatic carriers, (iii) analysis of the male/female ratio in carriers of various types of Rea, and, (iv) examination of parental ages distributions according to carriers’ sex.

Results

Two hundred and forty six disease-defined cases of mosaicism for autosomal non-centromeric Rea with a normal cell line of known sex were identified from the literature. There was a significant difference in single chromosome involvements compared to structural rearrangements between affected and asymptomatic carriers of unbalanced Rea, p =0.0030. In affected carriers, chromosome 18 was most frequently involved in structural rearrangements (12.6% of 246 instances). The least frequently rearranged were chromosomes 16 and 21 (0.8% and 1.2%, respectively). In asymptomatic carriers, the most frequently rearranged were chromosomes 5 and 21 (13% of 51 instances each). Among carriers of “loss” or “gain/loss” of genomic material, a female predominance was observed (50 M/89 F, different from population ratio of 1.06 at p = 0.0002). Carriers of either “gain” or balanced Rea demonstrated typical male predominance (41 M/30 F and 18 M/16 F), not different from 1.06. Maternal and paternal ages were reported in 129 and in 109 cases, respectively. There was a significant difference in maternal age distribution between male and female carriers, with mean maternal age of 25.2 years vs 28.3 years (p = 0.032). However, there was no difference in paternal age, with mean paternal age of 29.4 in both groups.

Conclusion

The data suggested that structural rearrangements of certain chromosomes involved in mosaicism may not be tolerated by the embryo, while others have higher survival prospects. Maternal age appears to be a risk factor for somatic mosaicism of structural Rea in female offspring or might cause an adverse effect on male embryo viability.

Electronic supplementary material

The online version of this article (doi:10.1186/s13039-017-0321-9) contains supplementary material, which is available to authorized users.

CNV analysis in Chinese children of mental retardation highlights a sex differentiation in parental contribution to de novo and inherited mutational burdens

Rare copy number variations (CNVs) are a known genetic etiology in neurodevelopmental disorders (NDD). Comprehensive CNV analysis was performed in 287 Chinese children with mental retardation and/or development delay (MR/DD) and their unaffected parents. When compared with 5,866 ancestry-matched controls, 11~12% more MR/DD children carried rare and large CNVs. The increased CNV burden in MR/DD was predominantly due to de novo CNVs, the majority of which (62%) arose in the paternal germline. We observed a 2~3 fold increase of large CNV burden in the mothers of affected children. By implementing an evidence-based review approach, pathogenic structural variants were identified in 14.3% patients and 2.4% parents, respectively. Pathogenic CNVs in parents were all carried by mothers. The maternal transmission bias of deleterious CNVs was further replicated in a published dataset. Our study confirms the pathogenic role of rare CNVs in MR/DD, and provides additional evidence to evaluate the dosage sensitivity of some candidate genes. It also supports a population model of MR/DD that spontaneous mutations in males' germline are major contributor to the de novo mutational burden in offspring, with higher penetrance in male than female; unaffected carriers of causative mutations, mostly females, then contribute to the inherited mutational burden.

Contrasting mechanisms of de novo copy number mutagenesis suggest the existence of different classes of environmental copy number mutagens

While gene copy number variations (CNVs) are abundant in the human genome, and often are associated with disease consequences, the mutagenic pathways and environmental exposures that cause these large structural mutations are understudied relative to conventional nucleotide substitutions in DNA. The members of the environmental mutagenesis community are currently seeking to remedy this deficiency, and there is a renewed interest in the development of mutagenicity assays to identify and characterize compounds that may induce de novo CNVs in humans. To achieve this goal, it is critically important to acknowledge that CNVs exist in two very distinct classes: nonrecurrent and recurrent CNVs. The goal of this commentary is to emphasize the deep contrasts that exist between the proposed pathways that lead to these two mutation classes. Nonrecurrent de novo CNVs originate primarily in mitotic cells through replication-dependent DNA repair pathways that involve microhomologies (<10 bp), and are detected at higher frequency in children of older fathers. In contrast, recurrent de novo CNVs are most often formed in meiotic cells through homologous recombination between nonallelic large low-copy repeats (>10,000 bp), without an associated paternal age effect. Given the biological differences between the two CNV classes, it is our belief that nonrecurrent and recurrent CN mutagens will probably differ substantially in their modes of action. Therefore, each CNV class may require their own uniquely designed assays, so that we as a field may succeed in uncovering the broadest possible spectrum of environmental CN mutagens.

Incomplete penetrance and phenotypic variability of 6q16 deletions including SIM1

6q16 deletions have been described in patients with a Prader-Willi-like (PWS-like) phenotype. Recent studies have shown that certain rare single-minded 1 (SIM1) loss-of-function variants were associated with a high intra-familial risk for obesity with or without features of PWS-like syndrome. Although SIM1 seems to have a key role in the phenotype of patients carrying 6q16 deletions, some data support a contribution of other genes, such as GRIK2, to explain associated behavioural problems. We describe 15 new patients in whom de novo 6q16 deletions were characterised by comparative genomic hybridisation or single-nucleotide polymorphism (SNP) array analysis, including the first patient with fetopathological data. This fetus showed dysmorphic facial features, cerebellar and cerebral migration defects with neuronal heterotopias, and fusion of brain nuclei. The size of the deletion in the 14 living patients ranged from 1.73 to 7.84 Mb, and the fetus had the largest deletion (14 Mb). Genotype-phenotype correlations confirmed the major role for SIM1 haploinsufficiency in obesity and the PWS-like phenotype. Nevertheless, only 8 of 13 patients with SIM1 deletion exhibited obesity, in agreement with incomplete penetrance of SIM1 haploinsufficiency. This study in the largest series reported to date confirms that the PWS-like phenotype is strongly linked to 6q16.2q16.3 deletions and varies considerably in its clinical expression. The possible involvement of other genes in the 6q16.2q16.3-deletion phenotype is discussed.

Deletions in the 3' part of the NFIX gene including a recurrent Alu-mediated deletion of exon 6 and 7 account for previously unexplained cases of Marshall-Smith syndrome

Marshall-Smith syndrome (MSS) is a very rare malformation syndrome characterized by typical craniofacial anomalies, abnormal osseous maturation, developmental delay, failure to thrive, and respiratory difficulties. Mutations in the nuclear factor 1/X gene (NFIX) were recently identified as the cause of MSS. In our study cohort of 17 patients with a clinical diagnosis of MSS, conventional sequencing of NFIX revealed frameshift and splice-site mutations in 10 individuals. Using multiplex ligation-dependent probe amplification analysis, we identified a recurrent deletion of NFIX exon 6 and 7 in five individuals. We demonstrate this recurrent deletion is the product of a recombination between AluY elements located in intron 5 and 7. Two other patients had smaller deletions affecting exon 6. These findings show that MSS is a genetically homogeneous Mendelian disorder. RT-PCR experiments with newly identified NFIX mutations including the recurrent exon 6 and 7 deletion confirmed previous findings indicating that MSS-associated mutant mRNAs are not cleared by nonsense-mediated mRNA decay. Predicted MSS-associated mutant NFIX proteins consistently have a preserved DNA binding and dimerization domain, whereas they grossly vary in their C-terminal portion. This is in line with the hypothesis that MSS-associated mutations encode dysfunctional proteins that act in a dominant negative manner.

The rate of nonallelic homologous recombination in males is highly variable, correlated between monozygotic twins and independent of age

Nonallelic homologous recombination (NAHR) between highly similar duplicated sequences generates chromosomal deletions, duplications and inversions, which can cause diverse genetic disorders. Little is known about interindividual variation in NAHR rates and the factors that influence this. We estimated the rate of deletion at the CMT1A-REP NAHR hotspot in sperm DNA from 34 male donors, including 16 monozygotic (MZ) co-twins (8 twin pairs) aged 24 to 67 years old. The average NAHR rate was 3.5×10⁻⁵ with a seven-fold variation across individuals. Despite good statistical power to detect even a subtle correlation, we observed no relationship between age of unrelated individuals and the rate of NAHR in their sperm, likely reflecting the meiotic-specific origin of these events. We then estimated the heritability of deletion rate by calculating the intraclass correlation (ICC) within MZ co-twins, revealing a significant correlation between MZ co-twins (ICC = 0.784, p = 0.0039), with MZ co-twins being significantly more correlated than unrelated pairs. We showed that this heritability cannot be explained by variation in PRDM9, a known regulator of NAHR, or variation within the NAHR hotspot itself. We also did not detect any correlation between Body Mass Index (BMI), smoking status or alcohol intake and rate of NAHR. Our results suggest that other, as yet unidentified, genetic or environmental factors play a significant role in the regulation of NAHR and are responsible for the extensive variation in the population for the probability of fathering a child with a genomic disorder resulting from a pathogenic deletion.

Many genetic disorders are caused by deletions of specific regions of DNA in sperm or egg cells that go on to produce a child. This can occur through ectopic homologous recombination between highly similar segments of DNA at different positions within the genome. Little is known about the differences in rates of deletion between individuals or the factors that influence this. We analysed the rate of deletion at one such section of DNA in sperm DNA from 34 male donors, including 16 monozygotic co-twins. We observed a seven-fold variation in deletion rate across individuals. Deletion rate is significantly correlated between monozygote co-twins, indicating that deletion rate is heritable. This heritability cannot be explained by age, any known genetic regulator of deletion rate, Body Mass Index, smoking status or alcohol intake. Our results suggest that other, as yet unidentified, genetic or environmental factors play a significant role in the regulation of deletion. These factors are responsible for the extensive variation in the population for the probability of fathering a child with a genomic disorder resulting from a pathogenic deletion.

De novo dup p/del q or dup q/del p rearranged chromosomes: review of 104 cases of a distinct chromosomal mutation

We compiled 104 constitutional de novo or sporadic rearranged chromosomes mimicking recombinants from a parental pericentric inversion in order to comment on their occurrence and parental derivation, meiotic or postzygotic origin, mean parental ages, and underlying pathways. Chromosomes involved were 1-9, 13-18, 20-22, and X (64 autosomes and 40 X chromosomes). In the whole series, mean paternal and maternal ages in cases of paternal (proved or possible; n=29) or maternal (proved or possible; n=36) descent were 31.14 and 28.31 years, respectively. Rearranged X chromosomes appeared to be of paternal descent and to arise through intrachromosomal non-allelic homologous recombination (NAHR), whereas rec-like autosomes were of either maternal or paternal origin and resulted from mechanisms proper of non-recurrent rearrangements. Except for some mosaic cases, most rearranged chromosomes apparently had a meiotic origin. Except for 8 rearranged X chromosomes transmitted maternally, all other cases compiled here were sporadic. Hence, the recurrence risk for sibs of propositi born to euploid parents is virtually zero, regardless of the imbalance's size. In brief, recombinant-like or rea chromosomes are not related to advanced parental age, may (chromosome X) or may not (autosomes) have a parent-of-origin bias, arise in meiosis or postzygotically, and appear to be mediated by NAHR, nonhomologous end joining, and telomere transposition. Because rearranged chromosomes 10, 11, and Y are also on record, albeit just in abstracts or listed in large series, we remark that all chromosomes can undergo this distinct rearrangement, even if it is still to be described for pairs 12 and 19.

Enhanced maternal origin of the 22q11.2 deletion in velocardiofacial and DiGeorge syndromes

Velocardiofacial and DiGeorge syndromes, also known as 22q11.2 deletion syndrome (22q11DS), are congenital-anomaly disorders caused by a de novo hemizygous 22q11.2 deletion mediated by meiotic nonallelic homologous recombination events between low-copy repeats, also known as segmental duplications. Although previous studies exist, each was of small size, and it remains to be determined whether there are parent-of-origin biases for the de novo 22q11.2 deletion. To address this question, we genotyped a total of 389 DNA samples from 22q11DS-affected families. A total of 219 (56%) individuals with 22q11DS had maternal origin and 170 (44%) had paternal origin of the de novo deletion, which represents a statistically significant bias for maternal origin (p = 0.0151). Combined with many smaller, previous studies, 465 (57%) individuals had maternal origin and 345 (43%) had paternal origin, amounting to a ratio of 1.35 or a 35% increase in maternal compared to paternal origin (p = 0.000028). Among 1,892 probands with the de novo 22q11.2 deletion, the average maternal age at time of conception was 29.5, and this is similar to data for the general population in individual countries. Of interest, the female recombination rate in the 22q11.2 region was about 1.6-1.7 times greater than that for males, suggesting that for this region in the genome, enhanced meiotic recombination rates, as well as other as-of-yet undefined 22q11.2-specific features, could be responsible for the observed excess in maternal origin.

8p23.1 duplication syndrome; common, confirmed, and novel features in six further patients

The 8p23.1 duplication syndrome is a relatively rare genomic condition that has been confirmed with molecular cytogenetic methods in only 11 probands and five family members. Here, we describe another prenatal and five postnatal patients with de novo 8p23.1 duplications analyzed with oligonucleotide array comparative genomic hybridization (oaCGH). Of the common features, mild or moderate developmental delays and/or learning difficulties have been found in 11/12 postnatal probands, a variable degree of mild dysmorphism in 8/12 and congenital heart disease (CHD) in 4/5 prenatal and 3/12 postnatal probands. Behavioral problems, cleft lip and/or palate, macrocephaly, and seizures were confirmed as additional features among the new patients, and novel features included neonatal respiratory distress, attention deficit hyperactivity disorder (ADHD), ocular anomalies, balance problems, hypotonia, and hydrocele. The core duplication of 3.68 Mb contains 31 genes and microRNAs of which only GATA4, TNKS, SOX7, and XKR6 are likely to be dosage sensitive genes and MIR124-1 and MIR598 have been implicated in neurocognitive phenotypes. A combination of the duplication of GATA4, SOX7, and related genes may account for the variable penetrance of CHD. Two of the duplications were maternal and intrachromosomal in origin with maternal heterozygosity for the common inversion between the repeats in 8p23.1. These additional patients and the absence of the 8p23.1 duplications in published controls, indicate that the 8p23.1 duplication syndrome may now be considered a pathogenic copy number variation (pCNV) with an estimated population prevalence of 1 in 58,000.

Replicative mechanisms of CNV formation preferentially occur as intrachromosomal events: evidence from Potocki-Lupski duplication syndrome

Copy number variations (CNVs) in the human genome contribute significantly to disease. De novo CNV mutations arise via genomic rearrangements, which can occur in 'trans', i.e. via interchromosomal events, or in 'cis', i.e. via intrachromosomal events. However, what molecular mechanisms occur between chromosomes versus between or within chromatids has not been systematically investigated. We hypothesized that distinct CNV mutational mechanisms, based on their intrinsic properties, may occur in a biased intrachromosomal versus interchromosomal manner. Here, we studied 62 genomic duplications observed in association with sporadic Potocki-Lupski syndrome (PTLS), in which multiple mutational mechanisms appear to be operative. Intriguingly, more interchromosomal than intrachromosomal events were identified in recurrent PTLS duplications mediated by non-allelic homologous recombination, whereas the reciprocal distribution was found for replicative mechanisms and non-homologous end-joining, likely reflecting the differences in spacial proximity of homologous chromosomes during different mutational processes.

Contribution of global rare copy-number variants to the risk of sporadic congenital heart disease

Previous studies have shown that copy-number variants (CNVs) contribute to the risk of complex developmental phenotypes. However, the contribution of global CNV burden to the risk of sporadic congenital heart disease (CHD) remains incompletely defined. We generated genome-wide CNV data by using Illumina 660W-Quad SNP arrays in 2,256 individuals with CHD, 283 trio CHD-affected families, and 1,538 controls. We found association of rare genic deletions with CHD risk (odds ratio [OR] = 1.8, p = 0.0008). Rare deletions in study participants with CHD had higher gene content (p = 0.001) with higher haploinsufficiency scores (p = 0.03) than they did in controls, and they were enriched with Wnt-signaling genes (p = 1 × 10(-5)). Recurrent 15q11.2 deletions were associated with CHD risk (OR = 8.2, p = 0.02). Rare de novo CNVs were observed in ~5% of CHD trios; 10 out of 11 occurred on the paternally transmitted chromosome (p = 0.01). Some of the rare de novo CNVs spanned genes known to be involved in heart development (e.g., HAND2 and GJA5). Rare genic deletions contribute ~4% of the population-attributable risk of sporadic CHD. Second to previously described CNVs at 1q21.1, deletions at 15q11.2 and those implicating Wnt signaling are the most significant contributors to the risk of sporadic CHD. Rare de novo CNVs identified in CHD trios exhibit paternal origin bias.

Replication stress and mechanisms of CNV formation

Copy number variants (CNVs) are widely distributed throughout the human genome, where they contribute to genetic variation and phenotypic diversity. De novo CNVs are also a major cause of numerous genetic and developmental disorders. However, unlike many other types of mutations, little is known about the genetic and environmental risk factors for new and deleterious CNVs. DNA replication errors have been implicated in the generation of a major class of CNVs, the nonrecurrent CNVs. We have found that agents that perturb normal replication and create conditions of replication stress, including hydroxyurea and aphidicolin, are potent inducers of nonrecurrent CNVs in cultured human cells. These findings have broad implications for identifying CNV risk factors and for hydroxyurea-related therapies in humans.