The purpose of this paper was to compare the results obtained using different whole genome sequencing (WGS) kits for the detection of variants in cell-free DNA (cfDNA) of a patient with prostate cancer. For one kit, WGS metrics were also compared in specimens from healthy patients and those with breast, prostate, or colon cancer.
Comparison of commercially available whole-genome sequencing kits for variant detection in circulating cell-free DNA.
Mauger F, Horgues C, Pierre-Jean M, Oussada N, Mesrob L, Deleuze JF
Purpose
Conclusion
Despite dilution of the primers, the NEXTFLEX library preparation method still produced dimers (approximately 1% of clusters). The Accel PCR-Free method failed to produce a product and was not used further. The median coverage was highest when the library was prepared using the Accel-NGS method and this method produced the highest true positive rate (TPR) and positive predictive value (PPV) of single nucleotide variants (SNVs) and insertion/deletions (INDELS) for the reference DNA NA12878a. However, INDELS were not detected and not all of the SNVs were detected for the reference cfDNA HD780 and HD786, regardless of the methods/depths. Using specimens from patients with prostate cancer, the vast majority of copy number variations (CNVs) were detected, regardless of library preparation method, but the majority of SNVs and INDELs detected were found only with one library preparation method with the most unique variants were found using Accel-NGS. Importantly, despite these differences, the prostate specimen sequenced using the different workflows largely clustered together and separately from the other patient specimens.
The median coverage, the percentage of paired-end (PE) reads, and the insert size differed among the plasma specimens (different patients) when using the ThruPlex system.
Studies (1)
Purpose
The purpose of this study was to compare the results obtained using different WGS kits for variant detection in cfDNA of a patient with prostate cancer. For one kit, WGS metrics were also compared in specimens from healthy patients and those with breast, prostate, or colon cancer. K2EDTA plasma from two patients with breast cancer, one patient with prostate cancer, one patient with colon cancer, two healthy patients, and one pool of plasma from healthy males was obtained from a commercial source and stored at -80°C. Plasma was thawed and centrifuged at 16000 x g for 10 min to remove debris. DNA was extracted from 1 mL of plasma using the QIAamp Circulating Nucleic Acid Kit. Extracted cfDNA was quantified using the dsDNA HS Qubit Assay and fragment length was analyzed by bioanalyzer. Libraries from the prostate specimen were prepared using the TruPLEX Plasma-seq Kit, the QIAseq cfDNA All-in-One Kit (adapter diluted 70%), the NEXTFLEX Cell Free DNA-seq for Illumina Kit (adapters diluted 40%), the Accel-NGS 2 S PCR FREE DNA Library for Illumina Kit, and the Accel-NGS 2 S PLUS DNA Library for Illumina Kit and the optimal number of PCR cycles determined experimentally. Libraries for the other specimens were prepared only using the TruPLEX Plasma-seq Kit. Libraries were quantified using qPCR and sequenced using an Illumina HiSeq X Series sequencer. For further comparisons, libraries were constructed from sheared reference DNA NA12878 and reference cfDNA HD780 and HD786.
Summary of Findings
The average fragment size of extracted cfDNA was 167 bp ± 4 bp, but one breast cancer sample also had high molecular weight DNA of 10,000 bp and the pooled plasma from healthy donors had a peak at 8500 bp. Despite dilution of the primers, the NEXTFLEX library preparation method still produced dimers (approximately 1% of clusters). The Accel PCR-Free method failed to produce a product and was not used further. The median coverage was highest when the library was prepared using the Accel-NGS method but the percentage of paired-end reads was comparable among the library preparation methods. For the reference DNA NA12878a, the Accel-NGS method at 10X produced the highest TPR and PPV of SNVs (95.96% versus 81.44-88.37% and 99.42% versus 97.18-98.98%, respectively) and INDELs (87.47% versus 68.1-76.92%% and 96.06% versus 81.16-94.7%). The TPR was greater than 99% for SNVs and 93% for INDELS and the PPV greater than 99% for SNVs and 80% for INDELS for the NA12878a reference DNA at 30x, regardless of library preparation method. In contrast, for the reference cfDNA HD780 and HD786, INDELS were not detected at either depth regardless of library preparation method and not all of the SNVs present were detected using any of the methods/depths. Principle component analysis of the first two components (46% of the variation) clustered the specimens by source rather than by library preparation method.
Using specimens from patients with prostate cancer, the vast majority of CNVs were detected, regardless of library preparation method, but the majority of SNVs and INDELs detected were found only with one library preparation method with the most unique variants found using Accel-NGS (23,026 SNVs and 7410 INDELS). The largest overlap in detection was with the Accel-NGS and NEXTFLEX methods. The Accel-NGS Kit had a significantly smaller number of common Germline INDELS and somatic SNVs. Both Accel-NGS and the QIAseq method resulted in a significantly low proportion of common CNVs. Importantly, despite these differences, the prostate specimen sequenced using the different workflows largely clustered together and separately from the other patient specimens.
The median coverage (10.3X+/− 2.5X), the percentage of PE reads (90.4% +/−6.8%), and the insert size (163bp+/− 14.6bp) differed among the plasma specimens (different patients) when using the ThruPlex system.
Biospecimens
FluidPlasmaPreservation Types
FrozenDiagnoses
NeoplasticCarcinomaNormalPlatforms
DNAAutomated electrophoresis/BioanalyzerDNAFluorometryDNANext generation sequencingPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Next generation sequencing Specific Technology platform Accel-NGS 2S PCR FREE DNAAccel-NGS 2S PLUS DNANEXTFLEX Cell Free DNA-seqQIAseq cfDNA All-in-OneThruPLEX Plasma-seq