Poster Note As presented at AGBT 215, Marco Island, FL KAPA HYPER PLUS: A single-tube NGS library prep workflow integrating enzymatic fragmentation results in high yields and low sequencing bias. Authors Bronwen Miller 1, Maryke Appel 2, Victoria Van Kets 1, Beverly van Rooyen 1, Heather Whitehorn 1, Martin Ranik 1, Piet Jones 1, Adriana Geldart 2, Rachel Kasinskas 2 and Eric van der Walt 1 1 Kapa Biosystems, 271 Victoria Road, 2nd Floor, Salt River, Cape Town, South Africa 2 Kapa Biosystems, 2 Ballardvale St, Suite 25, Wilmington, MA
INTRODUCTION Continuous improvements to library preparation for next-generation sequencing (NGS) are necessary to achieve the highest data quality. One of the crucial steps within library preparation is the initial DNA fragmentation, which can be accomplished through either mechanical or enzymatic processes. Mechanical methods for DNA fragmentation are difficult to scale or automate, and require large investments in expensive instrumentation. Current enzymatic solutions for DNA fragmentation typically exhibit sequence bias, provide poor control over fragment length distribution, and are highly sensitive to input amount. To address these challenges, we have developed the KAPA Library Preparation Kit by integrating an enzymatic DNA fragmentation technology with fast and efficient library construction to provide a streamlined, easy-to-automate, single-tube solution for preparing NGS libraries from 1 ng 1 µg of dsdna.
Kapa Biosystems has developed a kit that integrates an enzymatic DNA fragmentation technology with fast and efficient library construction to provide a streamlined, easy-to-automate, single-tube solution. FIGURE 1 KAPA Library Preparation Kit Workflow KAPA Kit Total time: ~2.5 hours Nextera Total time: ~2.5 hours Total time: ~3.5 hours Single Tube Fragmentation End Repair and A-tailing Adapter Ligation Tagmentation Bead Cleanup Library Amplification (required) Fragmentation End Repair and A-tailing Adapter Ligation Single Tube The KAPA Kit includes low-bias enzymatic fragmentation, eliminating the need for mechanical DNA shearing methods which are difficult to automate and require expensive instrumentation. The singletube workflow enables DNA fragmentation and library construction in 2.5 hours. Bead Cleanup Bead Cleanup Bead Cleanup Library Amplification (optional) Library Amplification (optional) Bead Cleanup Bead Cleanup
FRAGMENTATION IS TUNABLE AND REPRODUCIBLE Existing commercial enzymatic fragmentation solutions for NGS library construction are extremely sensitive to sample type and input amount, but the KAPA Kit provides reproducible results across a range of inputs. Library insert sizes can be adjusted by varying the fragmentation time. 2 FIGURE 2 Reproducible library fragment size distributions are obtained from a variety of DNA sample inputs E.coli gdna Human gdna Amplicon Fluorescence 2 15 1 5 1 ng 15 min 1 ng 15 min 1 ng 3 min 1 ng 3 min Fluorescence 4 1 1 ng 15 ng 15 min min 35 1 1 ng 15 ng 15 min min 3 1 1 ng 3 ng 3 min min 25 1 1 ng 3 ng 3 min min 2 15 1 5 Fluorescence 5 4 3 2 1 1 ng 15 min 1 ng 15 min 1 ng 3 min 1 ng 3 min 35 1 15 2 3 4 5 6 1 2 138 bp -5 35 1 15 2 3 4 5 6 1 2 138 bp 35 1 15 2 3 4 5 6 1 2 138 bp Various input amounts of E.coli gdna, human gdna, or a 1.8 kb amplicon were processed using the Kit with fragmentation times of 15 or 3 minutes at 37ºC. After library amplification and a single 1X SPRI bead cleanup, samples were analysed using an Agilent High Sensitivity DNA Assay. FIGURE 3 Defined fragmentation parameters yield consistent library insert sizes using samples from multiple species across a wide range of GC content Reproducible Fragment Sizes 8 B. pertussis 1 ng Average Size (bp) 7 6 5 4 3 % GC B. pertussis 5 ng E. coli 1 ng E. coli 5 ng Human 1 ng Human 5 ng C. difficile 1 ng C. difficile 5 ng P. falciparum 5 ng 2 1 5 min 15 min 45 min Fragmentation Time 1 ng or 5 ng of Bordetella pertussis (68% GC), Clostridium difficile (29% GC), Escherichia coli (51% GC), Plasmodium falciparum (2% GC) or human gdna were fragmented for 5, 15 or 45 minutes at 37ºC, yielding average insert fragment sizes of ~7, ~35, and~2 bp respectively.
HIGH LIBRARY YIELDS LEAD TO IMPROVED SEQUENCING RESULTS The workflow results in greater conversion of input DNA to adapter-ligated library. This higher library construction efficiency generates more complex libraries and improved sequencing metrics, including lower duplicate rates. FIGURE 4 Conversion rates vary widely for commercial library construction products Conversion 1 9 8 7 6 5 4 3 2 1 Conversion Rate Ranges Covaris + Illumina Nano Prep 1 µg 1 ng 1 ng 1 ng Input Amount The integrated workflow results in greater conversion of input DNA to adapter-ligated library compared to Covaris-sheared DNA processed using the Illumina TruSeq Nano Prep Kit or the KAPA Hyper Prep Kit. Conversion rates are highest for for both high-and low-input applications. FIGURE 5 exome capture libraries generate fewer duplicate reads than libraries produced using Covaris 5A Duplication Rates (Double-capture Protocol) 5B Duplication Rates 9 9 Duplicates 8 7 6 5 4 3 24S (Degraded) 241S (Degraded) 43S (HMW) 414S (HMW) Duplicates 8 7 6 5 4 3 2 2 1 1 gdna FFPE A Libraries were prepared from 2 ng of high molecular weight (HMW) FFPE DNA (43s, 414s) or degraded FFPE DNA (24s, 214s) and captured with the Roche Nimblegen double-capture protocol (Data courtesy of Dr. Brian Walker at the Institute of Cancer Research) B Libraries were prepared from 5 ng hgdna or 5 ng FFPE DNA and captured with the Nimblegen SeqCap EZ HGSC VCRome panel.
Roche Nimblegen capture libraries prepared with or Covaris and Hyper Prep were compared to libraries prepared using Nextera Rapid Capture. Sequencing metrics show a decrease in off-target reads and an improvement in sensitivity of SNP detection. 3 FIGURE 6 Key sequencing quality metrics demonstrate that performs well for exome capture 6A Percent Off-Target 6B Sensitivity of SNP Detection 25 2 Nextera 1 95 15 9 1 5 85 gdna FFPE 8 Covaris + Hyper Prep Nextera A Off-target reads were calculated using Picard CalculateHSMetrics. B Single nucleotide polymorphisms (SNPs) were called using LoFreq with default settings and a minimum per-base depth of coverage cut-off of 1, and compared to the NA12878 human gdna high confidence variant set. Sensitivity of SNP detection (true positive rate) was calculated as true positives divided by the sum of true positives and false negatives. The total number of SNPs analyzed was greater than 25,.
MINIMAL SEQUENCE COVERAGE BIAS Bias and coverage uniformity were investigated for libraries prepared for exome sequencing from human DNA. FIGURE 7 Read start-site bias for human exome libraries Covaris and Hyper Prep Nextera 5 5 5 A A A 4 C G 4 C G 4 C G T T T 3 3 3 2 2 2 1 1 1-1 1 2 Position 3-1 1 2 Position 3-1 1 2 Position 3 A T G C Nucleotide content over a 3 bp window (-1 to +2 relative to read alignment start) illustrates that start-site complexity of is similar to that of Covaris, and superior to that of Nextera. For human exome libraries, the workflow shows improved coverage uniformity in comparison to current enzymatic-fragmentation methods and leads to significantly lower AT-dropout, especially for FFPE samples. GC-dropout is similar for all three workflows with high-quality gdna samples, whereas slightly outperformed Nextera with the FFPE sample. FIGURE 8 AT- and GC- dropout for gdna and FFPE samples AT-dropout GC-dropout 12 12 1 1 Nextera Nextera 8 8 6 6 4 4 2 2 gdna FFPE gdna FFPE AT- and GC-dropout were calculated using Picard GCBiasSummaryMetrics.
FIGURE 9 exome sequencing libraries show excellent coverage-depth distribution.4 Fraction of Target.35.3.25.2.15.1.5 Nextera Data for all libraries were down-sampled to equal number of reads (27 million per technical replicate). Roche Nimblegen capture libraries prepared from hgdna with or Covaris and Hyper Prep were compared to libraries prepared with Nextera Rapid Capture. Coverage distribution was similar for Covaris and libraries, while Nextera libraries displayed much broader coverage distribution. A sharper peak and smaller tails indicate more uniform coverage.. x 2x 4x 6x 8x 1x Depth of Coverage GC bias was also investigated in libraries prepared for whole-genome sequencing from microbial DNA. Libraries were prepared from 1 ng of E. coli, B. pertussis or C. difficile gdna. DNA was either sheared via Covaris to 6 bp and converted to library with the KAPA Hyper Prep kit, or prepared with the KAPA Kit with a 5 min fragmentation time. Additional comparisons were performed with the Nextera kit and the NEBNext Ultra Kit with Fragmentase. shows limited GC bias compared to Covaris, and much less GC bias than Nextera and NEBNext Ultra. FIGURE 1 GC bias comparison for microbial samples C. difficile (29% GC) E. coli (51% GC) B. pertussis (68% GC) Normalized Coverage 3. 2.5 2. 1.5 1..5. 1 15 2 25 3 35 GC% of 1 Base Windows 3 25 2 15 1 5 4 45 5 Number of Windows (x 1 3 ) Normalized Coverage 1.2 1.1 1..9.8.7.6 3 25 2 15 1 5 25 3 35 4 45 5 55 6 65 7 GC% of 1 Base Windows Number of Windows (x 1 3 ) Normalized Coverage 1.8 3 1.6 1.4 1.2 1..8.6 25 2 15 1.4 5.2. 45 5 55 6 65 7 75 8 85 GC% of 1 Base Windows Number of Windows (x 1 3 ) Nextera Fragmentase + NEBNext Ultra GC bias for C. difficile (left), E. coli (middle), and B. pertussis (right) was assessed by calculating the GC content of the reference in 1 bp bins and plotting normalized coverage across these bins for Covaris and Hyper Prep,, NEBNext Ultra and Nextera workflows using Picard CollectGCBiasMetrics. In the absence of sequencing bias, all bins would be equally represented, indicated by a horizontal distribution centered on a normalized coverage of 1. Distribution of GC content in the genome is indicated by the grey histograms.
CONCLUSIONS The KAPA Kit provides a fast and efficient, single-tube enzymatic fragmentation and library preparation solution. This kit combines all the advantages of high-quality, unbiased DNA fragmentation, the speed and scalability of tagmentation, and Kapa s high library yields and low amplification bias in one integrated workflow. This workflow enables the following benefits: Automation-friendly DNA fragmentation and library prep in as little as 2.5 hours Tunable and reproducible fragmentation that is flexible for DNA sample inputs from 1 ng 1 µg Industry-leading library construction efficiency resulting in reduced bias and maximum sequence coverage High-quality results from challenging sample types and applications such as FFPE DNA
Notes
Contact Us at: Headquarters, United States Wilmington, Massachusetts Tel: 781.497.2933 Fax: 781.497.2934 sales@kapabiosystems.com International Office Cape Town, South Africa Tel: +27.21.448.82 Fax: +27.21.448.653 sales@kapabiosystems.com Kapa Technical Support kapabiosystems.com/support 216 Kapa Biosystems. All trademarks are the property of their respective owners. PN191 A179 4/16