12. Analysis of the clinical utility of mate pair sequencing to further characterize congenital chromosome abnormalities

The field of clinical cytogenetic testing relies on chromosomal microarray (CMA) and karyotype to detect copy number imbalance and structural variation. Karyotype can both detect balanced and unbalanced rearrangements but cannot define breakpoint boundaries or gene content. CMA is less subjective and can more accurately define gene content but can only detect unbalanced rearrangements without structure or orientation information. Both methodologies can lead to variants of uncertain clinical significance for which no further clinical testing has historically been available. The Mayo Clinic Genomics Laboratory evaluated the ability of mate pair sequencing (MPseq) to refine breakpoints and more precisely describe complex rearrangements from cytogenetic studies. Here we provide a summary of MPseq's clinical utility in congenital cases with rearrangements of uncertain significance. We identified the rearrangement of interest in 28 samples analyzed. Of these 28 rearrangements, 46% (n=13) were translocations, 36% (n=10) were inversions, and 18% (n=5) were duplications. MPseq had high clinical utility, as sixteen cases (57%) were re-classified as pathogenic or likely pathogenic. Identifying affected genes at breakpoints was the most common form of result refinement. Interestingly, 21% (n=6) of samples had rearrangements that were more complex than originally expected based on cytogenetic studies. Finally, we identified five examples of possible positional effects, where the rearrangement is within close proximity to a known translocation or gene with clinical implications that overlap with the patient's phenotype. These results demonstrate that MPseq is an effective method to further characterize structural variants of uncertain clinical significance which have significant potential to represent pathogenic findings. As clinical genetic testing evolves in the direction of whole genome sequencing replacing most/all current genomic testing platforms, algorithms should be utilized to detect and characterize structural variants in order to increase diagnostic yield, and expression studies should be considered to further resolve cases in which position effects are possible.