Verification of miRNA targets

We have used the GSTAr script from CleaveLand to predict potential miRNA targets. The degradome reads were aligned to all rice transcripts from MSU Rice Genome Annotation Project Version-7 [39] using Bowtie with 2 mismatches. We calculated the fragment abundance of degradome reads and category for each position in every gene. Subsequently, we merged the potential targeting information and degradome read alignments to obtain the list of valid targets for each miRNA and generated a list with target cut site, category, and abundance. Category 0 is when more than 1 read aligns at the cut site and is the only position where maximum reads have aligned. Category 1 is when more than one read aligns and is also where the maximum number of reads aligns and is among the other equal maximum peaks. Category 2 is when more than one read aligns at the cut site and the depth is above the average depth (average of abundances at all positions that have at least one read) and below the maximum abundance. Category 3 is when more than one read aligns at the cut site and the depth is below or equal to the average depth (average of abundances at all positions that have at least one read). Category 4 is when only one read aligns at the cut position. We combined the category 0 and 1 and named it as 1 since both the categories provided almost the same information.