Table 1.
The extent of overlap between each dataset in pairwise combination
| Dataset 1 | Dataset 2 | n 1 | n 2 | O | E | F | Z | P value |
|---|---|---|---|---|---|---|---|---|
| RESCUE |
PESE |
238 |
238 |
75 |
13.8 |
5.42 |
17.3 |
< 0.001 |
| RESCUE |
ESR |
238 |
285 |
55 |
16.6 |
3.32 |
10.1 |
< 0.001 |
| RESCUE |
Ke-ESE400 |
238 |
400 |
54 |
23.2 |
2.32 |
6.8 |
< 0.001 |
| PESE |
ESR |
238 |
285 |
48 |
16.6 |
2.90 |
8.9 |
< 0.001 |
| PESE |
Ke-ESE400 |
238 |
400 |
65 |
23.2 |
2.80 |
9.2 |
< 0.001 |
| ESR |
Ke-ESE400 |
285 |
400 |
33 |
27.8 |
1.19 |
1.0 |
0.12195 |
| RESCUE |
Ke-ESE |
238 |
1182 |
125 |
68.7 |
1.82 |
8.2 |
<0.001 |
| PESE |
Ke-ESE |
238 |
1182 |
137 |
68.7 |
1.99 |
9.9 |
<0.001 |
| ESR | Ke-ESE | 285 | 1182 | 98 | 82.2 | 1.19 | 2.1 | 0.015 |
n1 = number of motifs in dataset 1; n2 = number of motifs in dataset 2; O = number of motifs in common between dataset 1 and dataset 2; E = expected = (n1 * n2)/T; where T is the total number of possible hexamers, that is, 4,096; F = overlap factor = O/E; factor >1 indicates more overlap than expected of two independent groups. Z score is the difference between O and E normalised by the standard deviation (derived from simulation). P values in bold are those significant after Bonferonni correction assuming P <0.05/9.