Table 2:
Gap-filling statistics for TGS-GapCloser and other gap-closing tools
| Input data | Unfilled gaps | Misassembly | Local misassembly | Scaffold NG50 (bp) | Scaffold NGA50 (bp) | Scaftig NG50 (bp) | Scaftig NGA50 (bp) | Runtime (min) | Peak memory (GB) |
|---|---|---|---|---|---|---|---|---|---|
| Draft assemblies | 2,600 | 176 | 126 | 1,561,142 | 196,307 | 9,687 | 9,464 | – | – |
| TGS-GapCloser | 288 | 187 | 324 | 1,426,438 | 383,995 | 194,512 | 149,166 | 12 | 16.37 |
| PBJelly | 1,730 | 664 | 741 | 1,240,439 | 83,803 | 29,715 | 19,247 | 3,137 | 9.93 |
| FGAP | 458 | 867 | 684 | 1,871,611 | 44,244 | 127,982 | 28,615 | 2,687 | 35.06 |
| GMcloser | 2,600 | 175 | 125 | 1,561,142 | 195,886 | 9,570 | 9,335 | 17,140 | 11.39 |
| Cobbler | 1,475 | 230 | 516 | 1,522,592 | 176,960 | 24,072 | 18,217 | 24 | 9.43 |
| LR_Gapcloser | 447 | 1,064 | 1,076 | 1,561,028 | 27,211 | 157,181 | 18,216 | 74 | 2.90 |
All datasets were run with 16 threads on the same computer. Note that QUAST accepts <10 continuous N's in the scaftig.