Figure 2
From: ConStruct: Improved construction of RNA consensus structures
Visualization of alignments by ConStruct. An alignment of SECIS elements created by CLUSTALW (A) and after manual optimization/correction using CONSTRUCT (B). In both cases predicted consensus structures and CONSTRUCT's GUI are shown. For an overview of colors used in CONSTRUCT see Table S4 in Additional file 1. Top left: Corresponding drawings of consensus structures (annotated with the consensus sequence) generated by CONSTRUCT; consensus base pairing probability is color-coded from white to red. Top right: Corresponding dotplots: the base pairing probability of individual sequences (dark blue for the selected sequence M_janaschii_sps and green for others) is shown top-right in CONSTRUCT's main window; yellow to red dots show the consensus pairing probability; white to light blue bars denote gaps. The lower-left triangle shows the MI normalized by pair entropy with a threshold of tCV = 30 % in rainbow-colors from yellow to red. The cursors in A and B (arrow in thermodynamics part and black square in MI part) point to a similar position. Bottom: Corresponding alignment windows. Nucleotides participating in a base pair to which the cursor points in the dotplot are automatically highlighted [colored by pairing probability from p = 0 (black) to p = 1 (red)]. The motif GAA (turquoise background), which is conserved in the internal loop, has been highlighted using the built-in regular expression search. Clicking with a mouse button to position 3 and to position 25 of the last sequence (M_jannaschii_fmfdh_B; see red cursors) in the alignment editor selects this subsequence; clicking once with left or right mouse button to the double-headed arrow moves the subsequence towards 5' or 3' end, respectively, by one position; in the top-right dotplot the corresponding base pairs are automatically positioned. Similarly, clicking to a 5' and a 3' nucleotide of two different sequences (for an example see blue cursors) selects all corresponding subsequences from the sequence range; if none of the subsequences ends in a gap and all are followed by a gap, the subsequence range is moved towards the gap by clicking to the double-headed arrow.