Table 3.
3DAC prediction using MRC docking.a
| Protein | MRC 3DAC separation | MRC 3DAC separation | Best Pockets (predictive %) | ||
|---|---|---|---|---|---|
| no cut-off | SRC variants | 2.8 units cut-off | SRC variants | ||
| CDK2 | 20/26 (77%; p=100) | ↗20/26 | 14/26 (54%; p=81) | ↗18/26 | 1y91 (77%) |
| CS: 0.68 | ↘12/26 | CS: 0.58 | ↘6/26 | 1ykr (77%) | |
|
| |||||
| THRB | 18/24 (75%; p=91) | ↗20/24 | 14/24(58%; p=81) | ↗16/24 | 3da9 (83%) |
| CS: 0.69 | ↘7/24 | CS: 0.61 | ↘5/24 | 2zi2 (79%) | |
|
| |||||
| HS90A | 6/17 (35%; p=31) | ↗16/17 | 4/17 (24%; p=25) | ↗13/17 | 2fwz (94%) |
| CS: 0.37 | ↘3/17 | CS: 0.31 | ↘0/17 | 2cct (76%) | |
|
| |||||
| FA10 | 8/11 (73%; p=89) | ↗10/11 | 7/11 (64%; p=89) | ↗9/11 | 3kqw (91%) |
| CS: 0.68 | ↘4/11 | CS: 0.64 | ↘4/11 | ||
|
| |||||
| LKHA4 | 9/23 (39%; p=59) | ↗17/23 | 8/23 (35%; p=66) | ↗15/23 | - |
| CS: 0.37 | ↘4/23 | CS: 0.35 | ↘2/23 | ||
|
| |||||
| BACE1 | 9/18 (50%; p=39) | ↗14/18 | 8/18 (44%; p=28) | ↗13/18 | 3l5d (78%) |
| CS: 0.60 | ↘7/18 | CS: 0.57 | ↘6/18 | 3lnk (78%) | |
|
| |||||
| SRC | 10/15 (67%; p=65) | ↗15/15 | 6/15 (40%; p=35) | ↗15/15 | 1o47 (100%) |
| CS: 0.58 | ↘1/15 | CS: 0.45 | ↘0/15 | 1o48 (80%) | |
|
| |||||
| MK14 | 4/7 (57%; p=75) | ↗7/7 | 3/7 (43%; p=65) | ↗7/7 | 3d83 (100%) |
| CS: 0.58 | ↘1/7 | CS: 0.51 | ↘0/7 | ||
|
| |||||
| CHK1 | 3/5 (60%; p=61) | ↗4/5 | 2/5 (40%; p=72) | ↗3/5 | 2c3k (80%) |
| CS: 0.66 | ↘1/5 | CS: 0.56 | ↘0/5 | 2hog (80%) | |
a
The results are presented as a ratio between successfully-predicted 3DACs and the total number of 3DAC-pairs within a protein family. The MRC performance as compared to all individual pocket variants is expressed in percentile (p). The overall CS is also reported. The performance of best-(↗) and worst-performing (↘) pocket variants in each family are indicated. Conformations displaying accuracy over 75% are explicitly listed