Figure 5
From: MARZ: an algorithm to combinatorially analyze gapped n-mer models of transcription factor binding
Performance of top five gapped n -mer matrices for HB across all threshold values. The x-axis corresponds to the threshold position used for each run of the MARZ algorithm. The y-axis corresponds to the RZ score obtained from each run. Of the 32 matrices, the five with the highest RZ score, along with the mononucleotide (m) and dinucleotide (m m) matrices for comparison, are shown. The ranking of each matrix is indicated in parentheses. All of these matrices consistently outperform the 0.5 RZ score for a matrix that fails to discriminate between true and false positive binding sites (random, dotted line), but do not perform evenly across the different threshold values.