Fig. 1

Representation of a topic model adapted to the motif finding problem. Representation of a topic model adapted to the motif finding problem. This figure shows the basic structure of a topic model (in this case, a LDA). The terms specific for the case of the motif finding problem are stated in red under the original ones in blue, showing that the motif finding problem can be represented by a topic model by describing the DNA sequences as documents, the instances of each given motif as words in those documents, and the motifs as clusters of words or topics