Table 2.

Probability table of transcription factor binding sites in training and control data sets

Sr. No	TFs	P(TFi) in training data set$\left({\underset{¯}{\mathit{X}}}_{\mathbf{159}\mathbf{\times }\mathbf{14}}\right)$	P(TFi) in control data set$\left({\underset{¯}{\mathit{Y}}}_{\mathbf{100}\mathbf{\times }\mathbf{14}}\right)$
1	Tal1b-E47S	54/159=0.34	6/100=0.06
2	NF-kappaβ	29/159=0.18	10/100=0.10
3	n-MYC	74/159=0.46	7/100=0.07
4	ARNT	61/159=0.38	17/100=0.17
5	USF	61/159=0.38	7/100=0.07
6	c-REL	58/159=0.36	20/100=0.20
7	MEF	55/159=0.35	10/100=0.10
8	FREAC	94/159=0.59*	14/100=0.14
9	AML-1	94/159=0.59*	25/100=0.25
10	c-FOS	119/159=0.74*	23/100=0.23
11	HNF-3beta	119/159=0.74*	31/100=0.31
12	SOX17	136/159=0.85*	44/100=0.44
13	HFH	150/159=0.94*	51/100=0.51*
14	SOX-5	151/159=0.95*	44/100=0.44

* TFs with P(TF_i) greater than 0.50 defined as group-2. The table present Sr. No 8 to 14 in Training data set as the only group of TFs with high probability of TFs so the most likely co-occurring TFs in CNS brain specific enhancers compared to TFs with P(TF_i) less than 0.5 in training and control data sets.