Table 2:
Statistics of assembly results on real datasets
| Datasets | Toola | Assembly length (bp)b | #Ctgsc | N10 (kb)d | N50 (kb)d | N90 (kb)d | Genome fraction (%)e | #Misassembliesf | #SEsg | Real timeh | Peak memory (GB)i |
|---|---|---|---|---|---|---|---|---|---|---|---|
| C. elegans | xRead-nd | 106,383,147 | 50 | 6,183.59 | 3,561.75 | 970.09 | 97.989 | 186 | 18 | 360.4 (s) | 7.97 |
| xRead-wtpoa | 102,908,753 | 50 | 7,218.37 | 3,645.98 | 948.68 | 98.893 | 45 | 11 | 753.8 (s) | 18.93 | |
| xRead-pipe | 102,454,687 | 98 | 4,753.84 | 2,176.12 | 637.66 | 99.155 | 39 | 0 | — | — | |
| NextDenovo | 103,679,213 | 32 | 15,672.21 | 5,776.25 | 2136.32 | 99.547 | 76 | 2 | 7,979.5 (s) | 12.94 | |
| Wtdbg2 | 99,706,610 | 82 | 10,659.58 | 3,841.09 | 941.37 | 97.711 | 62 | 11 | 1,591.5 (s) | 20.47 | |
| Flye | 102,735,830 | 57 | 6,598.99 | 3,310.01 | 1,190.73 | 99.745 | 81 | 3 | 9,046.0 (s) | 45.57 | |
| Shasta | 99,144,317 | 66 | 5,428.85 | 2,831.68 | 950.70 | 97.369 | 38 | 1 | 611.5 (s) | 40.47 | |
| Hifiasm | — | — | — | — | — | — | — | — | — | — | |
| D. melanogaster | xRead-nd | 145,917,905 | 64 | 21,360.15 | 14,378.63 | 1,523.78 | 89.955 | 598 | 82 | 719.2 (s) | 12.56 |
| xRead-wtpoa | 146,420,770 | 64 | 21,611.07 | 14,367.83 | 1,508.29 | 90.973 | 393 | 77 | 1,210.0 (s) | 20.42 | |
| xRead-pipe | 142,950,657 | 151 | 15,209.22 | 10,453.23 | 532.13 | 90.979 | 333 | 1 | — | — | |
| NextDenovo | 136,332,796 | 31 | 27,936.35 | 22,718.49 | 2,307.26 | 92.572 | 205 | 13 | 7,444.5 (s) | 19.45 | |
| Wtdbg2 | 155,277,279 | 933 | 21,514.16 | 7,152.20 | 66.96 | 91.496 | 336 | 31 | 1,753.0 (s) | 27.92 | |
| Flye | 139,561,820 | 165 | 27,939.08 | 21,917.50 | 950.81 | 93.763 | 301 | 8 | 7,841.6 (s) | 50.84 | |
| Shasta | 133,568,381 | 152 | 27,932.29 | 21,763.69 | 944.65 | 91.181 | 242 | 6 | 457.5 (s) | 27.07 | |
| Hifiasm | — | — | — | — | — | — | — | — | — | — | |
|
H. sapiens
(ONT fast mode) |
xRead-nd | 2,858,898,337 | 971 | 27,779.62 | 6,055.25 | 1,386.12 | 90.401 | 3,018 | 358 | 20.89 (h) | 22.76 |
| xRead-wtpoa | 2,856,254,648 | 971 | 27,779.99 | 6,216.56 | 1,554.92 | 92.459 | 686 | 261 | 24.18 (h) | 37.05 | |
| xRead-pipe | 2,850,159,937 | 1410 | 21,912.39 | 4,486.01 | 1,155.18 | 92.405 | 372 | 29 | — | — | |
| NextDenovo | 2,782,862,247 | 638 | 57,559.56 | 25,059.84 | 3,780.30 | 91.614 | 329 | 108 | 71.89 (h) | 168.75 | |
| Wtdbg2 | 2,729,955,397 | 4257 | 31,781.95 | 11,412.92 | 1,354.65 | 88.623 | 522 | 400 | 20.12 (h) | 158.35 | |
| Flye | 2,847,205,020 | 3048 | 59,028.68 | 21,863.18 | 2,830.99 | 93.101 | 964 | 177 | 32.39 (h) | 217.85 | |
| Shasta | 2,768,613,021 | 5627 | 4,426.75 | 1,775.63 | 340.55 | 90.738 | 366 | 34 | 1.45 (h) | 293.24 | |
| Hifiasm | — | — | — | — | — | — | — | — | — | ||
| H. sapiens (PacBioHiFi) | xRead-nd | 2,972,271,147 | 2,478 | 31,619.87 | 12,035.12 | 662.49 | 95.209 | 3,269 | 253 | 7.16 (h) | 20.61 |
| xRead-wtpoa | 2,957,818,291 | 2,478 | 32,009.50 | 12,031.25 | 688.13 | 94.894 | 2,975 | 157 | 9.44 (h) | 22.88 | |
| xRead-pipe | 2,893,264,635 | 2,607 | 26,185.02 | 8,411.10 | 509.92 | 93.775 | 1,583 | 12 | — | — | |
| NextDenovo | 2,852,886,493 | 1,689 | 77,210.10 | 20,583.16 | 1,351.32 | 93.392 | 769 | 79 | 16.42 (h) | 86.67 | |
| Wtdbg2 | 2,769,287,788 | 2,295 | 44,134.56 | 14,168.42 | 2,005.26 | 90.804 | 619 | 135 | 10.32 (h) | 110.71 | |
| Flye | 2,918,011,841 | 3,121 | 59,296.15 | 24,686.38 | 2,047.19 | 94.656 | 2,678 | 107 | 25.88 (h) | 151.04 | |
| Shasta | 3,060,152,361 | 11,261 | 90,749.00 | 31,895.10 | 878.29 | 97.218 | 3,357 | 53 | 3.84 (h) | 547.01 | |
| Hifiasm | 3,089,368,991 | 664 | 139,105.11 | 87,025.29 | 9,357.14 | 98.353 | 3,693 | 854 | 6.07 (h) | 93.43 |
a
The assemblies were generated using 8 pipelines and benchmarked using QUAST and an in-house assessment script to assess the misassemblies.
b
The total number of bases in all contigs.
c
The total number of contigs.
d
N10/N50/N90: The length of the shortest contig at 10%/50%/90% of the assembly.
e
The percentage of aligned bases of the reference genome.
f
The number of all misassemblies, including both local sequence errors and structural errors.
g
The number of structural errors.
h
The overall real time of the tools’ cost on simulated datasets using 30 threads; the results marked by “s” and “h” indicate CPU hours and CPU seconds, respectively.
i
The peak memory of the tools (in GB) on simulated datasets.