Background: Interstitial deletions of 3q29 have been recently described as a microdeletion syndrome mediated by nonallelic homologous recombination between low-copy repeats resulting in an ~1.6 Mb common-sized deletion. Given the molecular mechanism causing the deletion, the reciprocal duplication is anticipated to occur with equal frequency, although only one family with this duplication has been reported.

Chromosome analysis is an important component to the diagnosis of congenital anomalies, developmental delay, and mental retardation. Routine chromosome analysis identifies aneuploidy and structural rearrangements greater than 5 Mb but cannot identify abnormalities of the telomeric regions or microdeletions reliably. Molecular cytogenetic techniques were developed to overcome these limitations. High-resolution comparative genomic hybridization (CGH)-based microarrays (array CGH) were developed to increase the resolution of chromosomal studies and to provide a comprehensive assay by using large-insert clones as the target for analysis. We constructed a microarray for the clinical diagnosis of medically significant and relatively common chromosomal alterations. Nine hundred six bacterial artificial chromosome (BAC) clones were chosen, the chromosomal locations of which were confirmed by fluorescence in situ hybridization (FISH). FISH-testing showed that 7% of the clones were mismapped based on map locations obtained from two publicly available databases (58 mapped to the wrong chromosome and three mapped to a different locus on the same chromosome), 16% cross-hybridized to other chromosomes, and 12% did not hybridize or showed poor hybridization signals under uniform FISH conditions. Thus, from a total of 906 BAC clones that were evaluated, only 589 (65%) were deemed adequate for arraying on this clinical device. The performance of this array was tested in a set of blinded experiments on a cohort of phenotypically normal individuals and on individuals with known chromosome abnormalities. The array identified deletion/duplication polymorphisms not seen by FISH in the phenotypically normal individuals and detected single copy dosage differences in all of the cases with known chromosomal abnormalities. All abnormalities detected by the array were confirmed by FISH with BACs from the appropriate loci. Our data demonstrate that the rigorous assessment of BACs and their use in array CGH is especially important when the microarray is used for clinical diagnosis. In addition, this study illustrates that when constructed carefully with proper attention to the quality of the BACs that are arrayed, array CGH is an effective and efficient tool for delineating chromosomal aberrations and an important adjunct to FISH and conventional cytogenetics.

Microdeletions of 1q43q44 result in a recognizable clinical disorder characterized by moderate to severe intellectual disability (ID) with limited or no expressive speech, characteristic facial features, hand and foot anomalies, microcephaly (MIC), abnormalities (agenesis/hypogenesis) of the corpus callosum (ACC), and seizures (SZR). Critical regions have been proposed for some of the more prominent features of this disorder such as MIC and ACC, yet conflicting data have prevented precise determination of the causative genes. In this study, the largest of pure interstitial and terminal deletions of 1q43q44 to date, we characterized 22 individuals by high-resolution oligonucleotide microarray-based comparative genomic hybridization. We propose critical regions and candidate genes for the MIC, ACC, and SZR phenotypes associated with this microdeletion syndrome. Three cases with MIC had small overlapping or intragenic deletions of AKT3, an isoform of the protein kinase B family. The deletion of only AKT3 in two cases implicates haploinsufficiency of this gene in the MIC phenotype. Likewise, based on the smallest region of overlap among the affected individuals, we suggest a critical region for ACC that contains ZNF238, a transcriptional and chromatin regulator highly expressed in the developing and adult brain. Finally, we describe a critical region for the SZR phenotype which contains three genes (FAM36A, C1ORF199, and HNRNPU). Although ~90% of cases in this study and in the literature fit these proposed models, the existence of phenotypic variability suggests other mechanisms such as variable expressivity, incomplete penetrance, position effects, or multigenic factors could account for additional complexity in some cases.

Genetic testing has the potential to guide the prevention and treatment of disease in a variety of settings, and recent technical advances have greatly increased our ability to acquire large amounts of genetic data. The interpretation of this data remains challenging, as the clinical significance of genetic variation detected in the laboratory is not always clear. Although regulatory agencies and professional societies provide some guidance regarding the classification, reporting, and long-term follow-up of variants, few protocols for the implementation of these guidelines have been described. Because the primary aim of clinical testing is to provide results to inform medical management, a variant classification program that offers timely, accurate, confident and cost-effective interpretation of variants should be an integral component of the laboratory process. Here we describe the components of our laboratory's current variant classification program (VCP), based on 20 years of experience and over one million samples tested, using the BRCA1/2 genes as a model. Our VCP has lowered the percentage of tests in which one or more BRCA1/2 variants of uncertain significance (VUSs) are detected to 2.1% in the absence of a pathogenic mutation, demonstrating how the coordinated application of resources toward classification and reclassification significantly impacts the clinical utility of testing.

Purpose: The advent of molecular cytogenetic technologies has altered the means by which new microdeletion syndromes are identified. Whereas the cytogenetic basis of microdeletion syndromes has traditionally depended on the serendipitous ascertainment of a patient with established clinical features and a chromosomal rearrangement visible by G-banding, comparative genomic hybridization using microarrays has enabled the identification of novel, recurrent imbalances in patients with mental retardation and apparently nonspecific features. Compared with the "phenotype-first" approach of traditional cytogenetics, array-based comparative genomic hybridization has enabled the detection of novel genomic disorders using a "genotype-first" approach. We report as an illustrative example the characterization of a novel microdeletion syndrome of 1q41q42. Methods: We tested more than 10,000 patients with developmental disabilities by array-based comparative genomic hybridization using our targeted microarray. High-resolution microarray analysis was performed using oligonucleotide microarrays for patients in whom deletions of 1q41q42 were identified. Fluorescence in situ hybridization was performed to confirm all 1q deletions in the patients and to exclude deletions or other chromosomal rearrangements in the parents. Results: Seven cases were found with de novo deletions of 1q41q42. The smallest region of overlap is 1.17 Mb and encompasses five genes, including DISP1, a gene involved in the sonic hedgehog signaling pathway, the deletion of which has been implicated in holoprosencephaly in mice. Although none of these patients showed frank holoprosencephaly, many had other midline defects (cleft palate, diaphragmatic hernia), seizures, and mental In 1963, Black and Carter 1 posited that the "elfin"-like facies characteristic of two clinical entities, supravalvular aortic stenosis and infantile hypercalcemia, suggested that the two disorders may be related. Nine years later, Beuren 2 demonstrated that supravalvular aortic stenosis and idiopathic infantile hypercalcemia, were, in fact, components of the same disorder,

Segmental duplications, which comprise approximately 5%-10% of the human genome, are known to mediate medically relevant deletions, duplications, and inversions through nonallelic homologous recombination (NAHR) and have been suggested to be hot spots in chromosome evolution and human genomic instability. We report seven individuals with microdeletions at 17q23.1q23.2, identified by microarray-based comparative genomic hybridization (aCGH). Six of the seven deletions are approximately 2.2 Mb in size and flanked by large segmental duplications of >98% sequence identity and in the same orientation. One of the deletions is approximately 2.8 Mb in size and is flanked on the distal side by a segmental duplication, whereas the proximal breakpoint falls between segmental duplications. These characteristics suggest that NAHR mediated six out of seven of these rearrangements. These individuals have common features, including mild to moderate developmental delay (particularly speech delay), microcephaly, postnatal growth retardation, heart defects, and hand, foot, and limb abnormalities. Although all individuals had at least mild dysmorphic facial features, there was no characteristic constellation of features that would elicit clinical suspicion of a specific disorder. The identification of common clinical features suggests that microdeletions at 17q23.1q23.2 constitute a novel syndrome. Furthermore, the inclusion in the minimal deletion region of TBX2 and TBX4, transcription factors belonging to a family of genes implicated in a variety of developmental pathways including those of heart and limb, suggests that these genes may play an important role in the phenotype of this emerging syndrome.

The chance of detecting a chromosome abnormality in a prenatal population that has already been screened by routine cytogenetics is approximately 1.3%. However, given that many of the abnormal array results in the neonatal population were among those with dysmorphic features as the primary indication for testing, which are not easily identifiable by ultrasound, offering prenatal testing by aCGH to a wider population would likely result in a higher detection rate.

Telomeric chromosome abnormalities are a substantial cause of mental retardation and birth defects. Although subtelomeric fluorescence in situ hybridization (FISH) probes have been widely used to identify submicroscopic telomeric rearrangements, array-based comparative genomic hybridization (array CGH) has emerged as a more efficient and comprehensive detection method. Due to the clinical relevance of telomeric abnormalities, it has been proposed that array CGH using panels of BAC clones that map to regularly spaced intervals along the length of each telomere could be used to characterize subtelomeric aberrations more precisely in a single experiment. We have added 1,120 FISH-mapped BAC clones to our microarray to enhance the coverage of the 41 unique human subtelomeric regions. Contigs of clones were selected in increments of approximately 0.5 Mb beginning with the most distal unique sequence for each subtelomere and extending on average approximately 5.7 Mb toward the centromere. We have used this microarray to characterize 169 clinically significant subtelomeric abnormalities identified out of nearly 7,000 consecutive clinical cases analyzed by array CGH in our diagnostic laboratory. The expanded telomere coverage was sufficient to define the breakpoints of over half (56%) of the chromosome abnormalities. However, 44% of the subtelomeric aberrations extended beyond the size of this expanded coverage suggesting that many subtelomeric abnormalities are >5 Mb in size and that greater representation may be of even more value. In addition to identifying 6 cases of complex rearrangements, we have identified 42 cases of interstitial deletions that would have been missed by subtelomere FISH panels that use a single clone to the most distal unique sequence for each region. Microarrays designed to investigate regions known to be involved in chromosome abnormalities will enhance the detection of cytogenetic abnormalities at unprecedented resolution and frequency.