Predicted secondary structure of the three <i>Kryptoperidinium foliaceum</i> endosymbiont mitochondrial introns modeled according to the conventions described in Burke et al.

<p><a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0043763#pone.0043763-Gordon2" target="_blank">[<b>47</b>] </a><b>and Michel et al. </b><a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0043763#pone.0043763-Rutherford1" target="_blank">[<b>48</b>]</a><b>.</b> (A) Group I introns. Left, the first cox1 intron; Right, the rnl intron. The K. foliaceum cox1 group I intron (left) had been previously mistakenly referred to as a group II intron <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0043763#pone.0043763-Imanian2" target="_blank">[27]</a>. (B) Group II intron. The second cox1 intron. Panels A and B: canonical Watson-Crick base pairings are denoted by dashes whereas guanine-uracyl pairings are marked by dots. Numbers inside variable loops indicate the sizes of these loops. Exon sequences are shown in lowercase letters. Panel A: splice sites between exon and intron residues are denoted by arrows; Panel B: the major structural domains are indicated by roman numerals and capital letters A to D, whereas tertiary interactions are represented by dashed lines, curved arrows, and/or Greek letters. Nucleotides potentially involved in the δ-δ′ interaction are boxed. Intron-binding and exon-binding sites are indicated by IBS and EBS, respectively. The putative site of lariat formation is denoted by an asterisk.