Skip to main content
Fig. 5 | BMC Genomics

Fig. 5

From: An overlooked phenomenon: complex interactions of potential error sources on the quality of bacterial de novo genome assemblies

Fig. 5

Visual representation of the pooled error rate effect sizes from the meta-analytic random-effects models in multiplicative GLMs on the total number of contigs (first row), size of largest contig (second row), NG50 (third row), and LG50 (fourth row). Error rate effect sizes were marginalized for different values of sequencing depth (25 × , 87.5 × , 150 × on the left, middle, and right columns, respectively), optical and PCR duplicate ratios (see on horizontal and vertical axes on the separate panels, respectively). Color intensity visualizes pooled error rate effect sizes, blue indicating negative, and red indicating positive effect. When present, the label (S) denotes suggestive (0.003 < P < 0.05), and NS denotes non-significant (P > 0.05) effects, whereas lack of labels indicate statistically significant (P < 0.003) effect of the pooled error rate estimate

Back to article page