Figure 4.

Layout of predicted hairpins and observed cores in an RNU1 consensus sequence.
    Underlined
sequences correspond to predicted 2D hairpins, as declared by mfold and approximately as predicted and described by O'Gormann et al. [2]. Nucleotides highlighted in gray correspond to our declared "cores" of Illumina sequences (abundant shared subsequences in Illumina reads). In Sample 1, all cores but S had 45 to 154 instances, but S itself had 4736 instances. In the other two samples the counts were: 10-59 vs. 1317; 60-247 vs. 2454. In short, S was much more popular than any other sequenced cores apparently derived from RNU1.

Jeffries et al. BMC Bioinformatics 2010 11:602   doi:10.1186/1471-2105-11-602
Download authors' original image