The revised diploid genome sequence of an individual human: A workflow for near telomere-to-telomere assemblies using only Oxford Nanopore Technologies Simplex-reads.

hifiasm assembly from Oxford Nanopore Simplex reads only, without ultra-long reads or proximity ligation

The authors report HuRef2.0, a revision of the first diploid assembly of an individual genome (HuRef, 2007), built using only Oxford Nanopore Simplex reads and the hifiasm assembler, without ultra-long reads or proximity ligation. Three flowcells (~103× coverage) from fresh blood DNA produced 26 contigs with gapless assembly of 23 chromosomes; two gaps on chromosome Y were locally assembled to yield a telomere-to-telomere assembly with base accuracy above Q60, 2,393 phase blocks and an NG50 of 2.36 Mb. A single flowcell (~35× coverage) consistently produced an assembly more contiguous than GRCh38.p14. Quality was assessed by chromosome-level alignment to a reference, variant calling and annotation of difficult regions: 22q11, the extended MHC, rRNA gene clusters, the 4q35 subtelomeric region, chromosome Y and centromeres. At equivalent coverage, more accurate but shorter PacBio HiFi reads produced less contiguous assemblies than error-corrected ONT reads.

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

The demonstration concerns a single individual, but it shows that a near-complete assembly is achievable from standard blood DNA and a single technology, bringing de novo assembly closer to laboratory practice. The regions assessed, 22q11 and 4q35 foremost, are precisely those where short-read sequencing falls short in diagnostics; the abstract, however, measures no clinical benefit.

Analysis by Dr Thibaut Benquey