Alignment-free clustering of transcription factor binding motifs using a genetic-k-medoids approach

BMC Bioinformatics

Table 1 Retrieval accuracy

	GMACS	STAMP	MoSta	SMLR
bZIP (93)	0.92	0.94	0.90	0.92
C2H2 (74)	0.82	0.76	0.76	0.77
C4 (52)	0.98	0.98	0.98	0.91
Homeobox (50)	0.88	0.82	0.82	0.85
Forkhead (49)	0.92	0.90	0.92	0.83
bHLH (37)	0.89	0.81	0.92	0.88
Total (355)	0.90	0.87	0.88	0.86

The ability of the KFV metric to distinguish different motif structural classes in the TRANSFAC benchmark dataset is compared to three alternative approaches. For each class, the highest retrieval accuracy is shown in bold. GMACS scores the overall highest average accuracy and does particularly well on the complex C2H2 zinc-finger family which causes problems for some of the other approaches.

ISSN: 1471-2105