Long-read sequencing and benchmarking of saliva against blood for genomic variant detection in autism

Genome-wide long-read sequencing of saliva-derived versus blood-derived DNA; higher nonhuman DNA content in saliva (10% versus 0.005% of reads)

This study is the first systematic evaluation of saliva-derived DNA for genome-wide long-read sequencing in autism: 105 new saliva samples were benchmarked against 68 new blood samples, supplemented by 189 previously published long-read blood samples and a matched blood-saliva quad. Saliva contained substantially more nonhuman DNA (mean 10% versus 0.005% of reads), reducing mean effective coverage from an estimated 28X to 24X, compared with 34X in blood. Numbers of validated (22,591 versus 23,064) and rare (209 versus 213) SVs per sample were comparable, as also observed in the quad at matched coverage. Saliva long-read sequencing identified four new pathogenic and likely pathogenic variants missed by short-read sequencing; the authors report variants affecting MECP2, DNMT3A, CHD8, LMBRD2, POGZ and PHF3, four additional candidates and a 9-18% increase in diagnostic yield.

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

Preprint, not peer reviewed. The practical message is strong: a saliva sample, easier to obtain than a blood draw in autistic children, does not degrade long-read SV detection, provided sequencing depth is increased to offset nonhuman DNA and lower assembly contiguity. The abstract details neither how the new diagnoses split between SVs and small variants nor the exact denominator of the 9-18% gain.

Analysis by Dr Thibaut Benquey