Background PCR-free whole-genome sequencing (WGS) remains the benchmark approach for minimizing bias and providing comprehensive genome characterizations. Achieving balanced genome coverage without amplification can be difficult when DNA quantity or quality is limited. To define the best parameters for sensitive, low-bias PCR-free sequencing in clinically relevant contexts, we evaluated multiple enzymatic library preparation chemistries and workflows using reference DNA.

Methods Using NA12878 gDNA, PCR-free libraries were prepared using one sonication and seven enzymatic fragmentation workflows. Fragmentation parameters and post-ligation SPRI ratios were tuned to generate comparable ~450 bp insert sizes. Libraries were sequenced on an Illumina NovaSeq 6000, and data quality was assessed across key metrics including GC bias, promoter coverage, insert size distributions, library yield, read alignment rates, sequence artifacts (hairpins and chimeras), and variant-calling performance. Input mass requirements were also evaluated down to 75 ng to model low-input scenarios.

Results: Watchmaker produced the highest post-size selection yield, nearly doubling the yield observed with other enzymatic fragmentation workflows. A 75 ng input processed with the Watchmaker workflow yielded libraries comparable to those produced from 300 ng inputs using other enzymatic products. Sequencing data from Watchmaker libraries showed minimal adapter contamination, indels, softclips, chimeras, improper pairs, and hairpin artifacts. Indicators of variant-calling accuracy were consistent with a sonication control. Coverage analysis demonstrated that Watchmaker libraries achieved the highest coverage in GC-rich promoter regions, consistent with prior observations.