Table 2:
Assembly accuracy on D. melanogaster.
| wtdbg2 | Flye | Others | |||||||
|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||
| W | W-int | W-so | F | F-int | F-so | hifiasm | LJA | HiCanu | |
|
| |||||||||
| EA50max (×106) | 13.7 | 12.2 | 14.5 | 0.7† | 0.7 | 1.0 | 14.1 | 6.0 | 13.4 |
| EA75max (×106) | 4.0 | 3.8 | 9.9 | 0.3 | 0.3 | 0.5 | 4.8 | 1.7 | 13.0 |
| N. misassemblies | 1 | 1 | 4 | 3 | 3 | 5 | 1 | 1 | 0 |
| N. contigs | 405 | 369 | 363 | 856 | 951 | 576 | 1,871 | 925 | 1,110 |
| Genome fraction (%) | 90.1 | 89.2 | 89.1 | 94.9 | 94.9 | 94.7 | 95.7 | 95.5 | 95.3 |
| Duplication ratio | 1.02 | 1.01 | 1.55 | 1.68 | 1.68 | 1.98 | 1.99 | 1.74 | 1.81 |
All statistics are shown in homopolymer compressed space. “W-int” is wtdbg2 with the homopolymer modification but without simple omnitigs, and “F-int” is Flye with the post-construction steps disabled but without simple omnitigs.
†:
Note that this number is much worse than the NGA50 reported for D. melanogaster on https://github.com/fenderglass/Flye. This is likely due to us using the highly heterozygous cross of the A4 and ISO1 strains of D. melanogaster, while the variant assembled on the Flye website is plain ISO1.