For an unsolved rare-disease case, when does reinterpretation add more value than new sequencing?

Reanalysis identified causative variants in 17 of 101 unsolved neuromuscular families. In a separate 20-family pediatric cohort, reported diagnostic variants were also recoverable by research short-read sequencing or reinterpretation. Two PHEX reports linked hidden variants to abnormal processing of RNA, the gene's working messages; a third resolved an inserted DNA segment with targeted assays after routine genome calls had not described it. Published June 30, 2026, the selected hereditary-cancer cohort investigated 134 high-risk participants using genome sequencing, RNA sequencing and structural analysis. It reported DNA-repair-gene variants classified as likely pathogenic or pathogenic in 18 patients, a FANCM deletion and six rare mobile-DNA insertions. These findings did not establish which candidates explained hereditary-cancer susceptibility. All three main datasets were available for 74 participants, and 125 of 134 were of Central-Northern European ancestry. This was not a comparison of validated diagnostic yield against reanalysis.

Detecting a variant and showing that it explains disease are different steps. Counting additional candidates as diagnoses would overstate the evidence for expanding testing beyond reanalysis.

Same-case comparisons against updated sequence reinterpretation; disease-specific validation of candidate findings; additional confirmed diagnoses attributable to each assay and variant class; complete assay coverage; standardized reanalysis intervals; broader ancestry representation; and cost and turnaround comparisons.