Additional file 8: of Finite-size effects in transcript sequencing count distribution: its power-law correction necessarily precedes downstream normalization and comparative analysis

Figure S4. Medians of Regressed slopes of first-fitted segment versus R2 fit for NUGC3 dilution and spike-in background datasets. Figure S4A and 4B show the median slope of the first-fitted segments versus the median R2 value of the spike-in background set and the NUGC3 dilution set respectively. For the necessary R2 computations, the reference replicate was taken as the replicate with the largest total reads within the data series. In both plots, the refined solution space of the optimum points-per-segment (PPS; as indicated besides the data points) is indicated by the error margins defined by the slope of the first highest-count segments from Table 1. Within this margin, the optimum PPS value is determined by the largest average R2 value where it is 20 for the spike-in background set and 45 for the NUGC3 dilution data sets. (PNG 316 kb