Detecting DNA regulatory motifs by incorporating positional trends in information content

Table 2 Percentage of correctly identified sites for crp using different methods, varying number of starting points and different dataset lengths L

The rows display the results for each method. First row: MEME software with the options -nmotifs 1 -dna -mod oops -brief -noshorten -b 0 -adj none. Second and third rows: the variable change point algorithm with normal prior, λ = 0 and motif width and motif specification (uni- or bimodal) listed in Table 1. The number of starting points in the algorithm was set at 1 or 100. Starting points were selected according to the MEME procedure as described in the text. The entry labeled with an asterisk corresponds to a trial in which the correct motif was not found, but one or two sites were correctly predicted by chance, with a spurious motif.

ISSN: 1474-760X