METHODS AND TECHNOLOGIES

The reliable identification of single nucleotide variants (SNVs) and small insertions and deletions (InDels) with next-generation sequencing (NGS) technologies has been a cornerstone of routine diagnostics. However, conventional short-read sequencing methods face technical limitations when detecting larger genomic regions (>150bp) with gene dosage variations. To address this, molecular genetic diagnostics have traditionally relied on multiplex PCR combined with fragment length analysis (MLPA) for identifying copy number differences in specific target regions. Despite its high sensitivity, MLPA is constrained by its limited availability, with commercial kits covering only a small fraction of the genomic regions that can be analyzed via NGS.

An alternative solution lies in indirectly identifying copy number variations (CNVs) from existing NGS data. Various algorithms, both commercial and open-source, have been developed to detect CNVs and structural variants (SVs) using diverse methodologies (Pirooznia, Goes et al., 2015). The choice of detection method depends on the sequencing approach. Structural changes such as deletions, duplications, and complex rearrangements (e.g., translocations, inversions) create breakpoints in the DNA. To detect these structural changes, it is necessary to detect the breakpoints using whole genome sequencing (WGS) without PCR amplification. However, due to its extensive scope and high costs, WGS remains impractical for many molecular genetic diagnostic applications. Instead, targeted regions of interest (ROIs) are enriched and sequenced using panel sequencing. Read-depth algorithms are particularly effective for such applications. These algorithms normalize enrichment data across a patient cohort by determining the depth of coverage (number of sequenced DNA fragments) at each genomic position and comparing it to a reference dataset.

Leveraging this approach, a CNV analysis pipeline was developed in Martinsried, Germany, enabling the detection of gene dosage differences at the level of individual targets or exons (Becker, Ziegler et al., 2018). To enhance the pipeline’s specificity, it integrates three independent algorithms for CNV detection. This multi-algorithm approach ensures highly accurate identification of heterozygous, hemizygous, and homozygous CNVs within the analysis region (Ziegler, Becker et al., 2018). This methodology provides significant diagnostic advantages, particularly in scenarios where CNV analysis was previously unfeasible or restricted to a limited number of genes (Busse, Ibisler et al., 2018).