Systematic evaluation of long-read and short-read sequencing in neurological disorders diagnosis: a direct comparison study of 310 patients

Head-to-head comparison of long-read and short-read WGS in 310 families with undiagnosed neurological disorders (diagnostic yield, technical capabilities, cost)

This study directly compares long-read (lrGS) and short-read (srGS) WGS in 310 families with undiagnosed neurological disorders from the Hong Kong Genome Project. The overall diagnostic yield is 22.6% (70/310), with comparable detection rates: 92.2% for long-read and 96.1% for short-read. Long-read alone confirmed three repeat expansions through phasing and methylation, and improved the resolution of two complex structural variants; short-read alone detected six variants in homopolymeric regions. The unit cost per sample is comparable (USD 1,545 vs 1,580).

Synthesis written by Geno'X. For the full original abstract, please refer to the source publication.

A head-to-head comparison in 310 patients at equivalent cost is exactly the kind of evidence laboratories need to decide their WGS strategy. The message is nuanced and honest: long-read is not yet ready to replace short-read but excels at repeat expansions, methylation, and phasing — one more argument for positioning the genome (rather than panels) as a first-tier test in neurology.