Fig. 4

Detection Diversity Assessment. a Mapping rate of various small RNAs utilizing the 1000 ng input human brain data. Undetermined indicates that the read did not map to the annotations of the evaluated small RNA classes. Error bars show standard deviation. Significance is only shown for methods that had significantly different mapping rates compared to all other methods with the same direction of change. b Mapping rate of small RNAs for all starting input amounts for each method. The Y-axis shows the percentage of reads of each category and the X-axis shows each tested brain sample. Statistical tests for the differences in mapping rates are shown in Additional file 6: Table S4. c The bars show the number of unique miRNAs with greater than 10 normalized reads common to all triplicates for the 1000 ng data of the first batch. Points indicate the number of unique miRNAs for each triplicate and the standard deviation error bars shown are for these points. d Percentage of miRNAs that had quantifications above 10 in only 1 or 2 of the triplicates. e Bars show number of unique isomiRs with greater than 100 normalized reads common to all triplicates for the 1000 ng data of the first batch. Points indicate the number of unique isomiRs for each triplicate and the standard deviation error bars shown are for these points. f Percentage of isomiRs that had quantifications above 100 in only 1 or 2 of the triplicates. g Overlap of unique miRNAs with greater than 10 normalized reads in all triplicates for the 1000 ng data of the first batch. h Overlap of unique isomiRs with greater than 100 normalized reads in all triplicates for the 1000 ng data of the first batch. i Number of false isomiRs detected for each of the 962 synthetic sequences. j Number of normalized reads (expression) of the false isomiRs. k Percent variance of the number of isomiRs observed for each synthetic sequence explained by various sequence characteristics. The heatmap legend shows the percentage of variance from 0 to 9%