隨著下一代測(cè)序技術(shù)的飛速發(fā)展以及成本的迅速下降，在各種基于NGS的應(yīng)用中，對(duì)于高通量的需求也逐漸增加起來(lái)。作為整個(gè)NGS過(guò)程第一步，基因文庫(kù)的構(gòu)建與測(cè)序結(jié)果密切相關(guān)并且直接影響到后續(xù)的步驟。然而傳統(tǒng)的手工建庫(kù)操作是一項(xiàng)費(fèi)時(shí)、費(fèi)力工作，它不僅要求操作人員富有經(jīng)驗(yàn)，而且對(duì)液體操作的精度與準(zhǔn)度都有極高的要求。
作為業(yè)內(nèi)知名的液體操作專(zhuān)家，Eppendorf致力于為從事NGS相關(guān)工作的用戶(hù)提供更舒適的操作感受，并攜手試劑行業(yè)的領(lǐng)導(dǎo)者KAPA Biosystem公司，推出基于epMotion全自動(dòng)液體操作平臺(tái)的NGS建庫(kù)及定量解決方案，旨在為用戶(hù)提供一個(gè)更簡(jiǎn)單、可靠、可重復(fù)的NGS文庫(kù)制備體驗(yàn)。
以KAPA HyperPlus 系列試劑盒為例，采用epMotion 5075t平臺(tái)對(duì)4種不同起始濃度的E.coli基因組DNA進(jìn)行文庫(kù)構(gòu)建，每個(gè)濃度四組重復(fù)，整個(gè)過(guò)程（包括在位孵育）最快僅需2.5小時(shí)。

Automated KAPA HyperPlus DNA Library Preparation for Illumina^® Sequencing on the Eppendorf epMotion^® 5075t

Nathan Quon1, Cheng Liu2, Ph.D., Eleanor Cowley1, M.S., Rachel Kasinskas1, Ph.D. and Dan Stover1, M.S.
1Roche Sequencing & Life Science | Kapa Biosystems, Wilmington, USA; 2Eppendorf AG, Hamburg, Germany

Abstract
The rapid growth and declining costs of Next Generation Sequencing (NGS) have increased the demand for high throughput sequencing capacity. Automation of library preparation offers customers efcient and effective tools for meeting the increased throughput demands of the NGS world. Eppendorf has partnered with Kapa Biosystems, an industry leader in life science reagents, to develop an automated method for the KAPA HyperPlus Library Preparation Kit on the Eppendorf epMotion 5075t automated liquid handling system, as described in this application note.

Introduction
The KAPA HyperPlus Kit enables rapid construction of DNA libraries for Illumina sequencing. It is compatible with a wide range of sample types and input amounts (1 ng – 1 µg), making it one of the most versatile kits on the market. The novel one-tube chemistry streamlines the DNA fragmentation and library construction processes, yielding libraries of similar or better quality than those produced with KAPA HyperPrep Kits and Covaris®-sheared DNA. This highly optimized protocol, including engineered enzymes, optimally formulated buffers, and minimal cleanup steps, results in efcient conversion of input DNA to adapter-ligated library, enabling deep and uniform sequence coverage.
The method developed on the Eppendorf epMotion 5075t offers an automated solution for preparation of up to 48 sam ples, with the capability of scaling up to 96 samples per run. Modular programming gives the user flexibility to run a size selection step if desired, before or after PCR amplifcation. The system’s user-friendly interface also guides the user through the run setup, including placement of the labware and required reagent volumes. An optical sensor verifes that all labware is correctly placed before the run starts. Moreover, epMotion’s walk-away potential for generating PCR-free libraries is maximized by an on-deck ThermoMixer that provides homogeneous reaction mixtures, a thermal module that enables on-deck incubation steps, and a gripper that transports plates to various deck positions. Workflows with PCR require the use of an off-deck thermocycler.

Figure 1: Eppendorf epMotion 5075t automated liquid handling system.

Materials and Methods

Experimental Design

Sixteen libraries were prepared from varying input amounts of Escherichia coli (E.coli) genomic DNA to ensure the au tomated method performed within the KAPA HyperPlus Kit specifcations. All incubations were performed on-deck, with the exception of the library amplifcation. Molecular biol ogy grade mineral oil, an inert overlay, was used to prevent evaporation during the end repair and A-tailing incubation. Four replicates each of 1 ng, 10 ng, 50 ng and 200 ng inputs were fragmented at 37 °C for 35 minutes. End repair and A-tailing was performed at 65 °C for 30 minutes. The adapter concentrations were matched to the input amounts accord ing to the kit specifcations, as shown below in Table 1.
The ligation reaction was performed at 20 °C for 15 minutes, followed by a 0.8× post-ligation cleanup and 0.6× – 0.8× size selection. The number of library amplifcation cycles was adjusted for each input amount to achieve fnal library yields between 100 ng – 1 µg. The method was completed with a fnal 1× post-amplifcation cleanup.

Table 1: Experimental summary of epMotion 5075t validation run. Sixteen libraries were prepared: four replicates each from 1 ng, 10 ng, 50 ng, and 200 ng input amounts with appropriate adapter concentrations and PCR cycles to generate between 100 ng – 1 µg of final library.

Quality control (QC) samples at several stages in the workflow were collected during the validation of the automated method. QC samples were recovered after the post-ligation cleanup, size selection, and the post-amplifcation cleanup as shown in Figure 2. For each QC sample, 2 µL of sample was stored in 18 µL of 10 mM Tris-HCl (pH 8.0 – 8.5) to prevent degradation. Samples were quantifed using the KAPA Library Quantifcation Kit (LQK), which measures library fragments containing the sites necessary for flow cell hybridization and Illumina sequencing. This quantifcation is a costeffective alternative to sequencing that can indicate viability of the prepared libraries prior to downstream sequencing.
Post-ligation and post-size selection QC samples were diluted 1:10,000 and post-amplifcation QC samples diluted 1:100,000 to ensure that all data points fell within the standard curve of the KAPA LQK assay. All qPCR samples and standards were run in triplicate (outlier data points were excluded from the analyses) on the Eppendorf MasterCycler® thermal cycling device. Final library size distributions were determined by running 1:5 dilutions of the post-amplifcation libraries on the Agilent® 2100 BioAnalyzer® with the High Sensitivity DNA Assay.

Figure 2: KAPA HyperPlus workflow shown with optional QC checkpoints. The approximate total instrument time to run the 16 samples was 4 hours, including on-deck incubations. Size Selection, Library Amplifcation, and Post-Amplifcation Cleanup are optional steps that can be omitted, which would reduce the run time to ~2.5 hours.

Results and Discussion

The post-ligation qPCR results were used to calculate the percentage of starting material that was successfully adapter ligated, or the conversion rate as shown in Figure 3. Higher conversion rates are generally achieved with higher input amounts into library preparation, which results in libraries with greater complexity. With lower input amounts, a larger proportion of material may be lost to DNA adsorption to plas tic surfaces, which can contribute to decreased conversion rates. For inputs above 100 ng, conversion rates typically range between 50 – 100%. For inputs between 10 ng – 100 ng, conversion rates range between 10 – 50%. For inputs 1 – 10 ng, conversion rates range between 5 – 20%. Across all replicates, each input exceeded the expected conversion rates. The high conversion rates indicate that the Eppendorf LoBind® quality consumables (PCR plates, tubes) may improve sample recovery yields throughout the library construction process.

Figure 3: Post-Ligation Yields and Conversion Rates. All input amounts resulted in higher conversion rates than typically observed in manual experiments. 1 ng inputs had an average conversion rate of 28.86% compared to 5 – 20% typical conversion rates. 10 ng inputs and 50 ng inputs had average conversion rates of 96.81% and 170.91%, respectively, compared to 10 – 50% typical conversion rates. 200 ng inputs had an average conversion rate of 178.54% compared to 50 – 100% typical conversion rates. Conversion rates above 100% can be explained by highly ef cient adapter ligation with little to no sample loss during the post-ligation cleanup. The addition of the ~140 bp adapters to the DNA fragments increases the overall molecular weight of the samples. If most of the DNA fragments were adapter-ligated and carried through the post-ligation cleanup, then the net yield can be higher than the initial input concentration.

The post-size selection qPCR results were used to calculate the percentage of material retained from the post-ligation cleanup as shown in Figure 4. Typically, 5 – 20% of material is retained from size selection. Across all inputs, the size selection retention fell within the expected range.

Figure 4: Post-Size Selection Yields and Percentage Retained from Post-Ligation material. Following the 0.6× – 0.8× size selection, all input amounts resulted in 10 – 17% retention from the post-ligation cleanup, which falls within the expected range of 5 – 20%

The post-amplif cation qPCR results were used to calculate the amplif cation ef ciency as shown in Figure 5. Library amplif cation ef ciency is typically ≥ 80%, but can vary depending on the quality of sample and number of PCR cycles. With the chosen PCR cycling parameters, both the 1 ng and 10 ng input samples had an average amplif cation ef ciency greater than 80%. The 50 ng and 100 ng input samples had an average amplif cation ef ciency of approximately 74%, still being within the normal range. The number of PCR cycles should be optimized according to the requirements for downstream processes.

Figure 5: Post-Amplif cation Yields and Amplif cation Effi ciency. The number of PCR cycles varied across input amounts to achieve f nal library yields between

The Agilent 2100 BioAnalyzer was used to determine the fnal library size distributions shown in Figure 6. The 0.6× – 0.8× size selection was expected to retain fragments between 250 – 450 bp. Across all inputs, the average fnal library sizes were between 305 – 330 bp, with narrow and reproducible size distributions. The absence of adapter-dimer (~140 bp) and large fragments (> 450 bp) shows the size selection process to be highly effective.
 

Figure 6: Electrophoretic Size Analysis of Final Libraries across all inputs. Size analysis of all 16 replicates was performed. The average size of the 1 ng (top left) and 10 ng inputs (top right) were 305 bp +/- 8 bp and 317 bp +/- 1 bp, respectively. The average size of the 50 ng (bottom left) and 200 ng inputs (bottom right) were 330 bp +/- 5 bp and 330 bp +/- 7 bp, respectively. Within all replicates, the average size varied by less than 8%, which shows that size selection was highly robust.

Conclusion

NGS library construction is a critical step in sample prepa ration for sequencing on Illumina platforms. The increased demand for high-quality NGS libraries in high-throughput laboratories has necessitated the development of robust, automated methods for library preparation. Kapa Biosystems, an industry leader in NGS library preparation, has partnered with Eppendorf to automate the KAPA HyperPlus Kit on the Eppendorf epMotion 5075t. The experimental data shows that the automated method performs within the KAPA HyperPlus Kit specifcations at a wide range of input amounts from 1 ng – 200 ng. The highly flexible and modular automated solution for NGS library preparation will allow laboratories to easily scale up experi ments without sacrifcing quality in the process.

Figure 7: Eppendorf epMotion 5075t Deck Layout. This fgure shows the fully walk-away setup for fragmentation, end repair, A-tailing, adapter ligation and bead-based purifcation steps before the library amplifcation.