Table 3. Comparison of several methods using the largest subunit sequences of RNA polymerase.
| Method | CPU time (s) | Number of correctly aligned blocks |
|---|---|---|
| 76 sequences × 1182–2890 sites | ||
| CLW18d | 675.5 | 10 |
| CLW18q | 159.4 | 10 |
| NW-AP-2 | 54.95 | 8 |
| NW-NS-2 | 59.30 | 11 |
| FFT-NS-2 | 18.15 | 11 |
| FFT-NS-i | 173.1 | 11 |
| 24 sequences × 1206–2890 sites | ||
| T-COFFEE | 745.3 | 11 |
| CLW18d | 100.1 | 9 |
| CLW18q | 50.78 | 9 |
| NW-AP-2 | 20.79 | 10 |
| NW-NS-2 | 22.77 | 11 |
| FFT-NS-2 | 7.150 | 11 |
| FFT-NS-i | 46.00 | 11 |
The CPU time and the number of correctly aligned blocks by each method are shown for a large data set composed of 76 sequences and a small data set composed of 24 sequences. In the case of the large data set, the test of T-COFFEE was aborted due to a memory shortage. For all cases, default parameters were used, except for CLUSTALW, in which both cases default setting (CLW18d) and ‘quicktree’ option (CLW18q) were examined. All calculations were performed on a Linux operating system (Intel Xeon 1.7 GHz with 1 GB of memory). The gcc compiler (version 2.96) was used with the optimization option ‘-O3’.