Near-perfect genome sequencing in medical genetics

Medical genetics currently operates through a fragmented diagnostic cascade built around short-read sequencing technologies that carry well-documented blind spots, including regions of high sequence homology, tandem repeats and segmental duplications, as well as large or complex structural variants, invisible base modifications and a lack of variant phasing. We propose that long-read genome sequencing should be considered as one pillar of a broader technological convergence encompassing diploid genome assembly, pangenome references and artificial intelligence-driven variant interpretation, termed near-perfect genome sequencing (NPGS). We further propose a Bayesian framework in which genomic completeness itself constitutes interpretive evidence for variant classification. This principle has direct implications for the interpretation of variants of uncertain significance in clinical practice. We highlight the potential of NPGS across postnatal, prenatal and oncological settings and outline a staged implementation roadmap toward the one-test paradigm. We also address real-world implementation challenges, including cost, computational demand, equity and ethical considerations.