Integrating massively parallel sequencing into diagnostic workflows and managing the annotation and clinical interpretation challenge.

Massively parallel sequencing has become a powerful tool for the clinical management of patients with applications in diagnosis, guidance of treatment, prediction of drug response, and carrier screening. A considerable challenge for the clinical implementation of these technologies is the management of the vast amount of sequence data generated, in particular the annotation and clinical interpretation of genomic variants. Here, we describe annotation steps that can be automated and common strategies employed for variant prioritization. The definition of best practice standards for variant annotation and prioritization is still ongoing; at present, there is limited consensus regarding an optimal clinical sequencing pipeline. We provide considerations to help define these. For the first time, clinical genetics and genomics is not limited by our ability to sequence, but our ability to clinically interpret and use genomic information in health management. We argue that the development of standardized variant annotation and interpretation approaches and software tools implementing these warrants further support. As we gain a better understanding of the significance of genomic variation through research, patients will be able to benefit from the full scope that these technologies offer.

A comparison of molecular and cytogenetic techniques for the diagnosis of pregnancy loss.

PURPOSE: To evaluate the sensitivity, specificity, advantages, and limitations of multiplex ligation-dependent probe amplification compared with conventional karyotype analysis in the investigation of contributing factors to recurrent pregnancy loss. METHOD: A cohort of 284 patients underwent side-by-side analysis of products of conception by both conventional karyotyping and multiplex ligation-dependent probe amplification with direct comparison of results. RESULTS: Multiplex ligation-dependent probe amplification was shown to enable a diagnosis for an additional 47 (16.5%) patients compared with conventional karyotype analysis. However, this advantage was offset by some disadvantages of the method, including a high false-positive rate (8/104; 7.7%), as demonstrated by single-arm probe abnormalities of uncertain clinical significance, as well as the inability to characterize structural rearrangements, such as Robertsonian translocations, which comprised 2.46% of samples (99% confidence interval = 0.09-4.83), and ploidy changes. The calculated performance characteristics of multiplex ligation-dependent probe amplification in this cohort yielded a sensitivity of 86.9% and specificity of 92.4%. CONCLUSIONS: The advantages of now widely accepted molecular methodologies, such as lower failure rates, faster turnaround times, and lower cost, must be complemented by adequate counseling, family follow-up, and specific diagnostic reporting practices. It is particularly important to specifically address the important limitations of the methodology, including the inability to characterize balanced structural rearrangements and ploidy changes, especially if multiplex ligation-dependent probe amplification is to be performed alone.