Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: MARZ: an algorithm to combinatorially analyze gapped n-mer models of transcription factor binding

Figure 2

Comparison of a traditional frequency matrix and a gapped n -mer matrix for HB. (A) The traditional matrix treats each of the seven nucleotides (m) in the aligned HB binding site sequences independently and slides across the sequences with a window frame size equal to one nucleotide. The 4×7 matrix represents the frequency at which each of the four nucleotides is found at each of the seven positions in the binding site and can be used to generate a standard position weight matrix (PWM). (B) The matrix for the gapped n-mer mkkkkm considers each of the two outer nucleotides (m), but ignores the four inner nucleotides (k), with a sliding window frame size equal to six nucleotides. This generates a 16×2 matrix that represents the frequency at which the 16 possible nucleotide pairs are found at the two outer positions across the two possible frames in a seven nucleotide binding site sequence. It should be noted that the complex matrix constructed for each of the 32 different gapped n-mers is generated using the same principles as the mkkkkm example above, but is distinct based on the specific composition of the gapped n-mer. (C-D) Corresponding visualizations of each matrix at a HB binding site.

Back to article page