Fig. 6
From: ReSeq simulates realistic Illumina high-throughput sequencing data
51-mer spectra of real and simulated data. The shape and position of peaks reflects the coverage distribution, while the exponential decrease at low frequencies is defined by systematic errors. The black lines show the minimum between the exponential decrease and signal peak for real data. The first value of R gives the sum of relative deviations for frequencies (x) up to the minimum. The second value of R gives the sum of absolute deviations for frequencies larger than the minimum. Both values are stated relative to ReSeq. When the signal peak of the simulation does not exist or starts before the minimum in real data the values lose their interpretability. This happens for ART (a, c), pIRS (c, d, l), NEAT (a, c-e, l), and BEAR (b, c, f–g, i, k–l)