Illumina’s NovaSeqX Plus system has arrived at Novogene and we are delighted to be offering pre-made library sequencing service for selected servcies on the new platform. There are a number of factors that need to be taken into consideration when constructing libraries, so if you are preparing to send us your premade libraries, this article will provide some tips and tricks for this preparation so that you get the best results from your sequencing. It will also discuss common issues that can arise during premade library sequencing and how to avoid or rectify them.

Library preparation consists of 4 steps:

1. Fragmentation: samples must be fragmented into uniform-sized pieces within a range suitable for the chosen sequencing platform.
2. End repair & A-tailing: an adenine base is added to either end of the newly generated fragments. This overhang facilitates the addition of adapters with a thymine overhang.
3. Addition of adapters: a ligase enzyme covalently binds adapters to the fragments, producing a full library compatible with binding to the flow cell of the sequencing instrument.
4. PCR amplification: if sample input concentration is low, PCR amplification will be required.

Fragmentation

There are two important factors to consider with the fragmentation process – the method used and size selection. Most sequencing instruments work best when supplied with fragments of a similar size. Three methods exist for fragmentation – physical, chemical, and enzymatic.

• Physical: considered a “clean” method as it doesn’t introduce any additional chemicals or enzymes to the system. This is the primary method used by service providers as it requires specialised equipment, so is not readily accessible to individuals preparing their own libraries. • Chemical: mostly used for preparing RNA libraries; however, it can introduce cations to the system which may interfere with the downstream sequencing process. • Enzymatic: most commonly used for in-house library preparation as it only requires standard lab equipment and is highly scalable. It requires a low input for PCR free library generation; however, it can cause fragmentation bias and the ratio of enzyme to substrate (nucleic acid) needs to be controlled for good results.

Size selection is important for eliminating sub-optimal fragments from the library before introducing them to the sequencing platform. It ensures the fragments are within the optimal range for the particular instrument in use. This step also minimises wastage of sequencing capacity. Fragments that are too short will result in the formation of primer dimers. If fragments are too long, the machine won’t be able to cover the length of them or produce any useful sequencing data. Two methods exist:

1. Gel-based: a reliable method but is low throughput and can eliminate only very small or very large fragments.
2. Magnetic bead-based: the most common method used in commercial kits. This method is cost-effective with higher throughput and higher size separation resolution.

Anatomy of a Library

Careful library preparation is crucial for successful sequencing and data analysis. The purpose of library preparation is to make the DNA fragments that are to be sequenced compatible with binding to the flow cell. After fragmentation, adaptors are added to either end of the generated fragments to allow them to bind to the flow cell and be sequenced.

The adaptors added to the fragments consist of three elements on either end – a P5 or P7 binding region, a read sequencing primer binding site, and an index sequence.

Figure 1. The composition of a library. Image courtesy of Illumina.

• The P5 and P7 binding regions are complementary to the flow cell oligos and allow for hybridisation and cluster generation. • The index regions are short sequences of 6-10 base pairs that allow for multiplexing i.e. for sequencing multiple samples in one run, and identifying them afterwards. This example has an index on either end and is therefore a “dual index” library. • The read 1 primer binding site is where the forward read primer for sequencing binds and the read 2 primer binding site is where the reverse read primer binds. A “single-end” library will contain only the read 1 primer binding site, whereas a “paired-end” library will contain both read 1 and read 2 primer binding sites. This allows for forward and reverse sequencing for higher coverage.

Library Quantification

It is crucial to measure library concentration so that dilutions and calculations can be made accurately. Two main approaches for this are utilised – qPCR and fluorometric quantification methods such as Qubit.

qPCR measures the library concentration using primers that anneal to the P5 and P7 regions of the adapters. This means it can also indicate if the structure of the library is correct, as no amplification during the process could suggest that these regions don’t exist within your library. Use triplicates for each sample and at least two separate dilutions. A positive control, such as a previously sequenced library, is highly recommended. Novogene’s minimum requirement for library concentration is 2nM.

Fluorometric quantification methods use fluorescence-based dyes that selectively bind to either double-stranded DNA, single-stranded DNA or RNA. They are a more accurate method for the quantification of libraries with a broad range of fragment sizes, however they can overestimate the concentration as it measures all dsDNA within the pool – if any dimers or other unwanted dsDNA is within the system it will be picked up and quantified. It is also recommended to use a positive control with these methods.

Library Quality Control

Bioanalyzer instruments produce a trace that indicates the fragment size distribution of the library. They are best used for quality control of libraries with a narrow fragment size distribution as their accuracy decreases as size range increases. This method can show if the library fragments are too small or too large, if beads have carried over from the size selection step, or if PCR artefacts are present in the mixture. It also shows if the sample contains fragments at all, as samples can sometimes gets lost throughout the process, for example, during bead clean-up.

The trace will also show characteristic peaks for adapter and primer dimers if they are present. Adapter dimers contain the full adapter sequence that will be able to bind to the flow cell and generate clusters and sequencing data. They are caused by insufficient starting material at the beginning of library preparation, poor quality starting material or inefficient bead clean-up. Primer dimers don’t contain the adapter sequence so can’t bind to the flow cell and won’t be sequenced. Therefore, adapter dimers cause more issues. They use up sequencing capacity resulting in lower coverage of actual fragments and can negatively impact the sequencing data, sometimes causing a run to stop prematurely. They must be removed with additional clean-up steps with beads or gel purification.

Library normalisation and pooling

Once the libraries have been quantified and quality controlled, the concentration needs to be normalised before they can be pooled. The standard C1V1 = C2V2 calculation can be used to work out the dilution factor that will result in uniform concentration across the libraries that are to be pooled. Then add equal volumes of the libraries you wish to pool to a microcentrifuge tube and pipette 10 times to mix thoroughly.

Sequencing complications

Index hopping occurs when a high volume of samples are multiplexed together, causing the incorrect assignment of a library to a different index, aligning the fragment with the incorrect sequencing template.

How to avoid: • Remove free adapters during prep. • Store libraries at -20°C • Only pool libraries right before sequencing • Use unique dual indexing strategy – allows for better identification.

Index dropout is a rare phenomenon which appears as an unusually small number of reads assigned to a sample. It is not always clear what causes it but some possibilities are:

• Contamination • Insufficient starting DNA concentration • Interference between adapter sequences and index sequences or read sequences. • Insufficient nucleotide diversity

An additional sequencing run is usually required in this case, once the library has been cleaned up of concentrated.

Find out more about our pre-made library sequencing services here and contact us here about running your pre-made libraries on the new NovaSeqX Plus system.