Table 3.
Retrieval of candidate molecules from PubChem. The top part of the table corresponds to the CASMI setup except for replacing KEGG with PubChem. The middle part shows the results with 10 ppm mass window with PubChem. The bottom rows depict the best possible results that could be obtained if the optimal mass window width was known.
| Challenge | 1 | 2 | 3 | 4 | 5 | 6 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 |
| Mass window (ppm) | 300 | 200 | 500 | 300 | 300 | 300 | 500 | 500 | 500 | 500 | 600 | 500 | 300 | 500 |
| Candidates | 55,242 | 9,931 | 87,514 | 75,205 | 75,320 | 115,639 | 26,701 | - | - | 69,708 | 10,390 | 18,669 | 97,567 | 19,718 |
| Cat 2 rank | 70 | 2 | 355 | 12,699 | 4,050 | 6,881 | 11 | - | - | 13,796 | 341 | 13,689 | 71,410 | 8,350 |
| 10 ppm rank | 25 | - | 5 | 2,225 | 1,343 | 1,048 | 7 | - | - | 166 | 57 | 1,815 | 2,416 | 231 |
| Ideal ppm | 2.34 | 50 | 3.90 | 9.38 | 1.17 | 9.38 | 0.98 | - | - | 0.98 | 9.38 | 7.81 | 0.98 | 1.95 |
| Ideal rank | 23 | 2 | 4 | 2,225 | 1,337 | 1,048 | 7 | - | - | 15 | 57 | 1,780 | 775 | 198 |