Fig. 2.

An example of the dynamic programming recurrence. (a) The relationships between the dynamic programming matrix elements. (b) The suffixes, $c[i\text{;}\gamma ]$, of the coding sequences corresponding to the matrix elements shown in (a); e.g. $c[5\text{;}3]={\mathrm{y}}_1{\mathrm{k}}_2{\mathrm{h}}_1$. Capital letters (M, D, Y, K, H, and N) indicate amino acid codes, and their corresponding non-capital letters are codons: e.g. Y is tyrosine; ${\mathrm{y}}_1$ and ${\mathrm{y}}_2$ represent UAU and UAC, respectively. Each solid line indicates which matrix element (a set of score vectors) is used as the argument of each vmax operation. Dashed lines indicate the matrix elements that are skipped in the vmax operations due to the occurrence of a forbidden sequence motif C AAG CA (${\mathrm{y}}_2{\mathrm{k}}_2{\mathrm{h}}_1$ = UAC AAG CAU and ${\mathrm{y}}_2{\mathrm{k}}_2{\mathrm{h}}_2$ = UAC AAG CAC). This is an example of k = 3.