Table 5.

Comparison of two-step procedure with other methodologies

Gene	Number of motifs	Two-step BS	BS	Two-step MS	MS	MAVID	TBA
cfos	2	0	0	0	0	0	0
hoxb2	17	8 (+5)	13	2	1	0	0
pax6	6	6	1*	0	0	6	6
scl	5	3 (+1)	1	0	0	0	0
Total	30	17 (+6)	15	2	1	6	6

Number of motifs: the number of motifs reported by Blanchette and Tompa [26] in cfos, Scemama et al. [46] in hoxb2, Kammandel et al. [47] in pax6 and Göttgens et al. [48] in scl. Two-step BS: the number of previously described motifs detected by the two-step procedure, combining data reduction and motif detection using BlockSampler. The numbers in parentheses are the number of motifs present in non-significant blocks. BS: the number of previously described motifs detected by BlockSampler in initial full-length datasets. Two-step MS: the number of previously described motifs detected by combining data reduction and motif detection using MotifSampler. MS: the number of previously described motifs detected by MotifSampler in initial full-length datasets. MAVID: the number of previously described motifs detected (correctly aligned) by MAVID. TBA: the number of previously described motifs detected by TBA. *Only part of a motif was detected.