Table 3 Illustrative examples of possible changes made to convert unmodified motifs to specific modified counterparts, for downstream hypothesis testing. We use stacked letters like simple sequence logos. At these positions, N represents any base frequencies other than the base being modified. These make up the other positions in the motif’s PWM. \(\overset{\texttt {C}}{\texttt {N}} \rightarrow \texttt{m}\) indicates that a position containing cytosine is modified by setting all base frequencies other than m to 0 and setting the frequency of m to 1. Conversely, \(\overset{\texttt {C}}{\texttt {N}}\texttt{\raisebox{4pt}{N}} \rightarrow \ \overset{\texttt {h}}{\texttt {N}}\texttt{\raisebox{4pt}{G}}\) indicates that a position containing cytosine is modified by replacing the frequency apportioned to C with h, leaving the other base frequencies at that position unmodified. We portray the second base of each dinucleotide as having a frequency of 1. This second base, however, could also comprise different bases of various frequencies, including the base shown

From: Modeling methyl-sensitive transcription factor motifs with an expanded epigenetic alphabet