Optimizing NGS Library Prep

A crucial element of any experiment using next-generation sequencing (NGS) is the NGS library, a collection of similar-sized DNA fragments with appended adapter sequences. Optimizing your NGS library preparation is a worthwhile investment of time and attention to detail. “It is difficult to narrow down which aspects of library preparation are more responsive to optimization, as all steps work together to produce a high-quality NGS library,” says Elizabeth Carpenter, senior support scientist at Roche Sequencing. “The process of optimizing the fragmentation parameters, adapter concentration, and even the number of PCR cycles all begins with the quality of the sample and the amount of sample used.” This article provides some expert advice for optimizing the preparation of NGS libraries.

Quality control and quantification for samples

Sample quality is indeed paramount when optimizing NGS library prep, as are other tools for quality control and quantification. Many vendors offer kits for cleaning and concentrating DNA samples prior to library preparation, and for working with extremely small quantities of input DNA. “Having a library prep kit that can withstand variable input amounts from small to large, as well as one that produces high-quality results with minimal PCR cycles, such as our Ultra II line of products, will help with challenging samples,” says Eileen Dimalanta, associate director of applications and product development, NGS at NEB.

Measuring starting DNA requires different tools. For verifying the quality and quantity of starting samples, Xiaojing Yang, group leader at Zymo Research, recommends the Thermo Scientific™ NanoDrop^® microvolume spectrophotometer, the Invitrogen Qubit Fluorometer, or Agilent’s automated TapeStation^® for electrophoretic sample analysis. Agilent’s Bioanalyzer can also be used.

Although using control DNA during NGS library prep might seem obvious, they are important for evaluating workflow quality when investigating test samples. “Scientists can monitor each run quality using E. coli Non-Methylated Genomic DNA as spiking-in control, and the Human Methylated & Non-Methylated DNA Set can help researchers validate their workflow from library preparation to sequencing,” says Yang. Using high-quality controls can help you spot weaknesses in your NGS workflow. “One of the best practices in quality control for NGS library prep is to process a high-quality sample alongside the test samples,” says Carpenter. “This will help to show how well the library preparation is proceeding under the most optimal conditions.”

Quality control and quantitation for libraries

Evaluating the quantity and quality of a newly created NGS library is also advisable. “qPCR is one of the most accurate quantitation methods to determine the optimal amount of library needed for sequencing, because it quantifies only those molecules that can be sequenced,” says Carpenter.

However, qPCR cannot measure fragment sizes. “We recommend that scientists use fragment analyzing equipment such as TapeStation to check fragment sizes and for primer dimers,” says Yang. Such capillary electrophoresis-based methods can measure both quantity and fragment size. “[Capillary electrophoresis-based methods] allow confirmation of narrow size distribution and expected insert size, and can alert the user to the presence of unwanted artifacts,” says Dimalanta. “[The resulting library traces] are also very helpful when troubleshooting, as they are quick to share with colleagues or technical support specialists.”

Grace DeSantis, senior director of product development at Illumina, recommends fluorescent or colorimetric quantification methods. “In some cases, quantification and normalization steps are reduced via use of bead-based normalization,” she says. “In addition, with our bead-linked transposome (BLT) products, quantification steps can be reduced or eliminated, due to the self-normalizing nature of this technology.”

Reducing bias

Ideally, library preparation proceeds without bias, accurately representing the genetic code despite challenging factors such as high GC content. “A good library-preparation kit should reduce bias as much as possible to capture the true molecular dynamics,” says Yang. “For example, both Zymo-Seq RRBS Library Kit™ and Zymo-Seq WGBS Library Kit™ optimize the library prep process to ensure unbiased methylation calling and reproducible CpG coverage, and Zymo-Seq RiboFree^® Total RNA Library Kit uses a patented technology to eliminate bias from the rRNA depletion process of NGS library prep.”

Another opportunity to reduce bias entails paying attention to a library’s fragment sizes. “It’s important to consider loading similar-sized libraries within one lane of a [sequencer’s] flow cell to reduce bias,” says DeSantis. “There is a tendency for smaller insert libraries to cluster more efficiently.” The sizes of library fragments can be “tuned” using bead- or gel-based size selection kits, but according to DeSantis, “these approaches result in library and diversity loss.” A better alternative, she says, is Illumina’s bead linked transposome technology, which reduces the number of workflow steps by fragmenting, tagging, and normalizing in one bead-based reaction, and concomitantly reduces diversity loss.

Adapter concentration and PCR cycles

PCR parameters such as adapter concentration and number of cycles are also opportunities for optimization. “The best way to optimize the adapter concentration is to determine the concentration most often used for a specific sample input, either from previous research or the recommended amount based on manufacturer's protocol, and then test a non-precious sample,” says Carpenter. Diluting the adapter concentration can also make library prep more efficient. “When lower input amounts are used, adapter dilution can be very helpful to prevent adapter dimer formation, and thus avoid the need for additional bead cleanups before sequencing,” adds Dimalanta.

Avoiding unnecessary PCR cycles during library prep is also beneficial. “For library amplification, it is ideal to minimize the number of PCR cycles to avoid both introduction of PCR bias and high levels of duplicates,” explains Dimalanta. “The number of cycles can vary depending on the original input amounts and the sample type, and should be high enough to provide sufficient library for a successful sequencing run, but low enough to avoid PCR artifacts and over-cycling.”

Throughout the NGS library-prep workflow, there are many opportunities for optimization, and hence multiple chances to increase your NGS success.