Fig. 2

The impact of WGA on metagenomic profiling with sub-nanogram DNA. a JSD between the expected and measured composition of sequins. The p value from the Wilcoxon test is shown. b Correlations between the input mass of sequins components and corresponding numbers of reads per million. Different levels of input amount are indicated by colors; R² values from linear regression models are shown, p values < 0.001. c Correlations between the abundance bias of sequins components and corresponding fragment lengths for libraries with 0.5 pg template. R² and p values from linear regression models are shown. d The length distribution of Nanopore sequencing reads aligned to A. baumannii and C. striatum in a sputum sample measured with WGA-based (right) and non-WGA-based (left) libraries. Dotted lines indicate the median lengths. e Relative abundances of dominant genera in the sputum sample