The on-target rate can provide information about the specificity of the target enrichment. Usually, this rate is calculated as the ratio between the number of sequenced bases covering the target regions and the total number of mapped bases. Besides calculating the percent bases on-target, also called percent on-target bases, one can also calculate the percent reads on-target, which are also called percent on-target reads. This rate includes all sequenced reads that map to the target region, even if it is only a single base. The more bases or reads map to the target region, the higher the on-target rate (see figure 1). Low on-target rates are presumed to result from suboptimal probe design, low-quality substances, or problems during preparation steps, such as library preparation or hybrid capture.
Figure 1 | Explanation of the different on-target rates. In our example, 40 reads were sequenced. Of those 40 reads, 31 reads map (partially) to the target region (red). Not all bases of a read necessarily need to map to the target region to make the read an on-target read – already one base of a read mapping to the target region makes the read an on-target read. In the example, a read mapping rate of 31/40 ≈ 77.5% is achieved. The light red box highlights the bases that map to the target region, making them on-target bases.
During target enrichment, fragments are captured by probes that are complementary to the target region. The coverage at the targets’ boundaries drops compared to the middle of the targets. Enrichment probes are usually designed to cover the target (exon) boundaries to obtain sufficient coverage at the boundaries of the targets. This increases the coverage at the boundaries at the expense of a lower on-target rate since the captured fragments only partially overlap with the targets. Using the on-target rate as a quality parameter in targeted sequencing experiments can be deceptive. Coverage extending over the original target regions into the exon-flanking regions allows including these regions in the downstream analysis. Our long-standing experience in genetics and sequencing shows that relevant information can be found in the exonflanking regions. We do not necessarily want to optimize the on-target rate, since this would simultaneously decrease the exon-flanking regions that can be analyzed.
CeGaT’s on-target rate might be lower than the ones reported by competitors, but we provide additional details on the relevant regions flanking the targets.