Fig. 2

Standard NGS metrics for merged reads from SRSLY and NEBNext Ultra II libraries from healthy human cfDNA extracts H-69 and H-81. Unless otherwise stated, all libraries for each method were combined by cfDNA extract prior to analysis and filtered for PCR duplicates and a quality score equal to or greater than q20. (a) Insert distribution plots for cfDNA extracts H-69 and H-81, respectively. (b) Fold coverage by base percent across the human genome (hg19) for SRSLY and NEBNext by cfDNA extract. Combined libraries were subsampled to similar read depth prior to fold coverage calculations. Subsampled depth was set at 295 M reads, the limit of sequenced reads for SRSLY-H-81. (c) Preseq complexity estimate for SRSLY and NEBNext by cfDNA extract. Three libraries of equivalent sequencing depth per method were combined to estimate complexity, since more libraries were made via SRSLY than NEBNext. Files containing the PCR duplicate reads were used to facilitate complexity estimates (d) Normalized coverage as a function of GC content over 100 bp sliding scale across the human genome for SRSLY and NEBNext by cfDNA extract. Green histogram represents the human genome GC across the 100 bp sliding window. (e) Normalized, log-transformed base composition at each position of read termini starting 2 bp upstream and extending to 34 bp downstream of read start site for combined cfDNA extracts for SRSLY and NEBNext. All reads regardless of insert length considered