Figure 7

Potential false uncapped 5′-ends caused by non-specific PCR amplification. Schemas depict models of uncapped transcripts (A) and capped transcripts (B) captured by the PARE protocol. The 5′ adaptor primer was perfectly annealed to cDNA corresponding to the 5′ RNA adaptor ligated to uncapped transcripts whereas it was partially annealed at its 3′-end to the internal region of capped cDNA. Distribution of normalized uncapped reads around the GTCCGAC sites in the CDS of Arabidopsis (C) and rice (D) genes with PARE libraries is visualized by MORPH. Reads surrounding motifs corresponding to the 3′ end of the 5′ adaptors used in degradome sequencing (E) and GMUCT (F) method are also visualized by MORPH. Motifs were boxed with the first nucleotide set as 1. Loci containing the motif of interest were identified from the CDS of all annotated genes and the number is shown in parentheses above the heat map. Only loci with a total read number greater than five in the 20-nt region are shown and the number of loci in each heat map is also indicated. Motifs were boxed with the first nucleotide set as 1. Loci were clustered based on the distribution of normalized read numbers across the 20-nt region by Ward’s method.