Figure 1.

Workflow to identify B. rapa miRNA genes. Candidate miRNA genes were predicted based on a combination of small RNA deep sequencing (blue arrows) and similarity searches using known A. thaliana miRNA sequences (green arrows). Sequence reads generated from small RNA sequencing were aligned to the draft genome sequences of B. rapa followed by secondary structure modeling. The resulting reads were further analyzed for structural and compositional features, genomic position, and expression level validation. Similarity searches using known A. thaliana miRNAs followed by secondary structure modeling and structural and compositional feature analysis identified matches of conserved miRNA candidates in the B. rapa genome. Integration and clustering of both data sets resulted in the final miRNA candidate genes (brown arrows).