Table 2.
Comparative time scales of brute force MD versus discontinuous approaches as reported in retrospective studies. Owing to the inherent difficulties in comparing timescales of several short trajectories, adaptive sampling is not considered in this table.
| Author (Year) | Complex | Multiple Ligands | No. of Runs | Aggregate Time | Productive Runs 1 | Time to Binding |
|---|---|---|---|---|---|---|
| Brute Force MD | ||||||
| Shan et al. (2011) | PP1/Src kinase | y | 7 | 115 µs | 3 | 15.1–1.9–0.6 µs |
| Dasatinib/Src kinase | y | 4 | 35 µs | 1 | 2.3 µs | |
| Buch et al. (2011) | Benzamidine/Trypsine | n | 495 | 49.5 µs | 187 | 15–90 ns |
| Dror et al. (2011) | Dihydroalprenolol/β2AR | y | 40 | 111.8 µs | 5 | NA |
| Alprenolol/β2AR | y | 10 | 14 µs | 1 | NA | |
| Propranolol/β2AR | y | 21 | 35.7 µs | 0 | - | |
| Isoprotenerol/β2AR | y | 1 | 15.0 µs | 0 | - | |
| Dihydroalprenolol/β1AR | y | 10 | 55.5 µs | 2 | NA | |
| Kruse et al. (2012) | ACh/M3 R | y | 1 | 25 µs | 1 | 9.5 µs |
| Tiotropium/M3 R | y | 3 | 18 µs | 0 | - | |
| Tiotropium/M2 R | y | 3 | 16.2 µs | 0 | - | |
| Decherchi et al. (2015) | DADMe-immucilin-H/PNP | y | 14 | 7 µs | 3 | 340 ns |
| Discontinuous Approaches | ||||||
| Sabbadin et al. (2014) | ZM241385/hA2A | n | 3 | - | 1 | 59 ns |
| T4G/hA2A | n | 3 | - | 1 | 62 ns | |
| T4E/hA2A | n | 3 | - | 1 | 105 ns | |
| Caffeine/hA2A | n | 3 | - | 1 | 15.2 ns | |
| Cuzzolin et al. (2016) | Ellagic Acid/CK2 | n | 3 | - | 0 | - |
| SAPS/GSTP1-1 | n | 3 | - | 2 | 27–19 ns | |
| Benzen-1,2-diol/PRDX5 | n | 3 | - | 3 | 17.4–31.2–18 ns | |
| (S)-naproxen/HSA | n | 3 | - | 0 | - | |
| (S)-fluoxetin/LeuT | n | 3 | - | 0 | - | |
| NECA/hA2A | n | 3 | - | 0 | - | |
| Zeller et al. (2017) | Oseltamivir/neuraminidase | n | 676 | 50.0 µs | ~20 | NA |
| Zanamivir/neuraminidase | n | 606 | 35.7 µs | ~20 | NA | |
1 “Productive” refers to simulations that reproduced the crystallographic pose within a given RMSD threshold.