Comparison of library preparation methods for sequencing of circulating cell-free DNA
Cancers are disease of the genome, and identifying cancer mutations can guide patient care. Invasive tissue biopsies are traditionally used to sequence cancer genomes, but can be costly, painful, and may not represent the true heterogeneity of tumors. Circulating cell-free DNA (cfDNA) found in plasma may be useful as a non-invasive test for cancer diagnosis, prognosis, or even as a real-time treatment monitoring tool. High-throughput sequencing of cfDNA has shown potential for detecting actionable somatic mutations in many cancer types. There are several kits marketed for efficient and minimally biased library preparation but, there is very little data comparing their performance, especially at different cfDNA input levels. In this project, we compared three different library preparation kits (Rubicon ThruPLEX, Swift Biosciences, and KAPA Hyper Prep). Cell-free DNA was extracted from pooled control plasma using the QIAamp Circulating Nucleic Acids extraction kit and quantified using digital droplet PCR. Each of the three library preparation kits were tested in triplicate starting with 2, 5 and 10 ng of cfDNA input. The final concentrations of each individual library were measured using qPCR and normalized to account for the number of PCR cycles during amplification. Preliminary results suggest that the Swift kit outperforms the KAPA and Rubicon kits at 2ng input. In contrast, the KAPA Hyper kit shows an expected increase in library yield as a function of cfDNA input and generates the highest library yield at 10ng input. In future work, we plan to sequence these libraries to measure library diversity and evaluate additional protocol optimizations to improve library yield.