Table 3.
Running time statistics. MS/MS samples were selected from Human, Shewanella, Dictyostelium and Yeast datasets. The results are shown in two modes: without clustering, and clustering using default values. The table holds the sizes of the original MS/MS datasets and the sequence databases used for identifications. For each experiment the table also holds the run-time (in cpu hours), and the relative speed-up factor achieved by clustering vs. the non-clustered search. The runtime was measured on a 3.0GHz desktop PC with 2GB of RAM.
| Dataset | Database size (amino acids) | Dataset size (# spectra) | Experiment type | Run Time (cpu hours)
|
Speed-up | ||
|---|---|---|---|---|---|---|---|
| clustering | search | total | |||||
| Human (1 run) | 26.7 M | 0.793 M | Non-clustered | - | 132.2 | 132.2 | |
| Clustered | 0.77 | 56.6 | 57.4 | 2.3 | |||
| Human (5 runs) | 26.7 M | 4.0 M | Non-clustered | - | 664.3 | 664.3 | |
| Clustered | 6.4 | 107.4 | 113.8 | 5.8 | |||
| Human (14 runs) | 26.7 M | 11.4 M | Non-clustered | - | 1812.7 | 1812.7 | |
| Clustered | 24.3 | 308.2 | 332.5 | 5.5 | |||
| Shewanella | 1.45 M | 14.5 M | Non-clustered | - | 286.3 | 286.3 | |
| Clustered | 26.8 | 28.5 | 55.3 | 5.2 | |||
| Dictyostelium | 7.36 M | 1.4 M | Non-clustered | - | 78.6 | 78.6 | |
| Clustered | 2.3 | 39.3 | 41.6 | 1.9 | |||
| Yeast | 4.9 M | 0.179 M | Non-Clustered | - | 7.5 | 7.5 | |
| Clustered | 0.1 | 3.5 | 3.6 | 2.1 | |||