Table 4.
The hypergeometric test for the partition effect of the deacetylation sites for the co-regulated (ChIP-chip experiments) gene pairs. For every correlation threshold, the number of pair-wise correlations for the genes with the same regulator and being in the same partition group that are above the threshold was counted. The number of pair-wise correlations for the genes with the same regulator and being in the neighboring partition groups that are above the threshold was also counted. The probability of randomly drawing such numbers of gene pairs above the threshold from all of the gene pairs with the same regulator is determined by a hypergeometric distribution.
Total gene pairs sharing one regulator | Gene pairs sharing one regulator and both are in the same deacetylation partition group | Threshold for correlations | Number of gene pairs sharing one regulator above the threshold | Number of gene pairs sharing one regulator and both are in the same deacetylation partition group above the threshold | P-values based on the hypergeometric tests |
476,649 | 1,811 | 0.6 | 14,738 | 134 | <4.75e-011 |
0.7 | 8,419 | 68 | 5.71e-009 | ||
0.8 | 4,556 | 29 | 0.00329 | ||
0.9 | 621 | 7 | 0.00293 | ||
Total gene pairs sharing one regulator | Gene pairs sharing one regulator and they belong to neighboring deacetylation partition groups | Threshold for correlations | Number of gene pairs sharing one regulator above the threshold | Number of gene pairs sharing one regulator and they belong to neighboring deacetylation partition groups above the threshold | P-values based on the hypergeometric tests |
476,649 | 1,019 | 0.6 | 14,738 | 34 | 0.287 |
0.7 | 8,419 | 18 | 0.438 | ||
0.8 | 4,556 | 8 | 0.638 | ||
0.9 | 621 | 1 | 0.383 |