P553: How much should I sequence? A validated mathematical model for determining sequencing depth targets in clinical germline NGS testing*

Clinical laboratories developing germline NGS tests repeatedly face the same question: how much sequencing depth is sufficient to ensure reliable variant calling across regions of interest (ROI) while meeting QC thresholds? In practice, decisions about per-sample depth and flow-cell pooling are often guided by tradition rather than quantitative analysis, leading to trial-and-error, wasted flow cells, and delayed launches. Here we present DeepSeq, a specification-driven mathematical framework that determines (i) the locus-level depth required for accurate heterozygote detection, (ii) the mean coverage needed to achieve defined coverage-assurance criteria across an ROI, and (iii) the optimal multiplexing level that satisfies these targets while limiting QC failures to a predefined rate.