Finding motifs — patterns of conserved residues — within nucleotide and protein sequences is a key part of understanding function and regulation within biological systems. This paper presents a review of current approaches to motif discovery, both evolutionary computation based and otherwise, and a speculative look at the advantages of the evolutionary computation approach and where it might lead us in the future. Particular attention is given to the problem of characterising regulatory DNA motifs and the value of expressive representations for providing accurate classification.
Michael A. Lones, Andrew M. Tyrrell