Skip to main content
. Author manuscript; available in PMC: 2021 Mar 28.
Published in final edited form as: Nat Biotechnol. 2020 Dec 7;39(3):309–312. doi: 10.1038/s41587-020-0711-0

Table 1.

Assembly statistics. HiFi read N50: 50% of HiFi reads are longer than this number. Contig NG50: minimum contig length needed to cover 50% of the known genome (GRCh38). Contig NGA50: similar to NG50, but based on contig alignment lengths to GRCh38 instead of contig sizes. Phasing switch error rate: percent adjacent SNP pairs are wrongly phased. Phasing hamming error rate: percent SNPs wrongly phased in comparison to true phases.

Sample HG002 (NA24385) NA12878 PGP1 HG00733
Long-read coverage 29.7 (HiFi) 30.1 (HiFi) 23.9 (HiFi) 33.4 (HiFi) 93.0 (CLR)
Long read N50 (bp) 13,480 10,004 12,974 11,769 33,090
Hi-C read coverage 38.5 44.8 261.7 35.5 67.1
Assembly algorithm Trio Canu Trio Peregrine DipAsm DipAsm DipAsm DipAsm Strand-seq Falcon-Phase
Scaffolding 3D-DNA HiRise HiRise 3D-DNA
Paternal / maternal contig size (Gbp) 2.96 / 3.04 2.81 / 2.88 2.98 / 2.97 2.97 / 2.97 2.98 / 2.98 2.93 / 2.93 2.90 / 2.90 2.89 / 2.89
Paternal / maternal contig NG50 (Mbp) 15.5 / 18.3 16.6 / 15.2 25.2 / 24.3 19.6 / 18.7 15.1 / 18.4 25.2 / 26.2 28.5 / 23.6 22.3 / 22.3
Paternal / maternal contig NGA50 (Mbp) 10.2 / 12.8 11.0 / 10.6 14.3 / 13.5 12.7 / 12.1 10.3 / 11.0 16.0 / 16.6 15.8 / 15.8 14.3 / 13.7
Phasing switch / hamming error rate (%) 0.38 / 0.23 0.38 / 0.31 0.50 / 0.49 0.15 / 2.13 0.21 / 1.63 0.16 / 0.60 0.30 / 0.70 0.43 / 35.8
SNP / INDEL false positive rate (×10⁻⁶) 1.9 / 31.6 2.6 / 32.0 2.4 / 27.7 2.0 / 4.2
SNP / INDEL false negative rate (%) 4.31 / 5.85 3.28 / 5.00 0.36 / 2.09 0.56 / 1.22 3.32 / - 4.00 / - 7.89 / -
SV sensitivity / precision (%) 90.7 / 92.8 90.6 / 92.6 93.4 / 92.6