NGS Library Size Selection: How Can It Boost Your NGS Results?

NGS library size selection is a critical step in next-generation sequencing (NGS) library preparation that directly impacts data quality, coverage uniformity, and sequencing efficiency. Library size selection isolates DNA fragments within a specific size range, removing both small fragments (including primer/adapter dimers) and larger fragments to boost sequencing efficiency, reduce costs, and improve data quality by ensuring only relevant, correctly sized DNA fragments are sequenced.

Library size selection prevents issues like wasted reads and challenges in alignment. It ensures the sequencer uses its capacity on meaningful data, leading to better assemblies and discoveries.

What is NGS Library Size?

Library size refers to the insert size + adapters (at both ends of the insert):

Insert size: DNA fragment of interest
Adapters: platform-specific sequences ligated to both ends of insert fragment

For example: a 300 bp insert with a 70 bp of adapter, the library size will be 300 + 70 + 70 = 440 bp.

Size Selection (Magnetic Beads)

Short-read Sequencing (Illumina, MGI)

Library Size Selection Kit (250-350 bp) – Ideal for illumina paired-end 100 (PE100) sequencing.
Library Size Selection Kit (300-450 bp) – Ideal for illumina paired-end 150 (PE150) sequencing .
Library Size Selection Kit (450-750 bp)–Ideal for illumina paired-end 300 (PE300) sequencing.

Long-read Sequencing (PacBio, Oxford Nanopore)

Library Size Selection Kit (>5 kb)– Ideal for long-read sequencing with DNA size >5 kb.
Library Size Selection Kit (>10 kb)– Ideal for long-read sequencing with DNA size >10 kb.

Other size selection

What is Library Size Distribution?

NGS Library Size Distribution refers to the range of DNA fragment lengths present in a NGS library after sample preparation. It is a critical quality control metric that directly impacts sequencing efficiency and data quality. An optimal size distribution can get better data by avoiding adapter dimers, increasing cluster efficiency, and maximizing usable reads.

Patterns of library size distribution: Ideal pattern is a tight peak around the target size, rather than a broad, messy spread.

Broad size distributions lead to:

- - Uneven amplification
  - Coverage bias
  - Dropouts in GC-rich or repetitive regions

Tighter size ranges give more consistent coverage, which is critical for:

- - Variant calling
  - Copy number analysis
  - RNA-seq quantification

Factors affecting size distribution:

Fragmentation method
- - Mechanical fragmentation: usually sonication, generates broader distribution
  - Enzymatic method: dependent on the type of the enzyme, some are more controlled
Sample type
- - FFPE sample: Often degraded, shorter DNA fragments
  - High-quality gDNA: Longer genomic DNA fragments
  - RNA-seq: Depends on RNA integrity

Measurement of library size distribution: Gel electrophoresis is used to visualize the size distribution

Instruments like Bioanalyzer and TapeStation
Traditional gel electrophoresis

Why Library Size Selection Can Boost Your NGS Results?

Sequencing platforms work best with fragments of certain sizes. Choosing the right fragment length helps:

Maximizes efficiency: Prevents sequencing capacity from being wasted.
- - Uniform fragment sizes: even cluster density
  - Reduced phasing/pre-phasing errors
  - Higher Q-scores and more stable runs

To maximize sequencing capacity, two approaches can be used:

1. 1. 1. Reduce adapter dimers and unwanted short fragments: Adapter dimers are typically ~120-140 bp:
      - Amplify extremely efficiently because of the short length
      - Can dominate a sequencing run if not removed
    2. Reduce larger fragments (for short-read sequencing). Large fragments cause low cluster density, and generate less reads and low-quality reads.

Improves data quality: A narrow, uniform fragment size leads to better cluster generation and more accurate read alignment,
- - Improves the reliability of genomic assemblies and variant detection (SNPs, indels).
  - Reduces Bioinformatics Burden: Cleaner, more uniform libraries mean less time spent filtering out junk reads and dealing with complex data during data analysis.
  - Increases reproducibility between samples: This is especially important in clinical, population, and differential expression studies.

When all libraries have similar size profiles:

- - - - Multiplexed samples more evenly
      - Reduced batch effects
      - More reliable comparisons across experiments

Optimizes for applications and sequencing platforms:

- - Applications: Different applications (whole genome, ChIP-Seq, RNA-seq, long-read) require specific fragment sizes; selection ensures the library matches the specific requirements.
  - Sequencing platforms: Different sequencers and sequencing kits have ideal fragment size ranges for efficient cluster generation and sequencing.

Size selection ensures the insert size fits the sequencing strategy:

- - - - Paired-end reads should overlap minimally or intentionally
        Avoids sequencing adapters instead of DNA inserts
        Optimizes fragment/read alignment

Examples of ideal library sizes:

- - - - Illumina PE100 sequencing: ~250–350 bp library size is ideal
        Illumina PE150 sequencing: ~300–450 bp library size is ideal
        Illumina PE300 sequencing: ~450–750 bp library size is ideal
        Long-read sequencing: PacBio, Oxford Nanopore etc., starts from several kb.

Note: Beads-based kits available for Illumina PE100, PE150, and PE300 with selection of perfect size ranges. For long-read sequencing, >5 kb and >10 kb beads-based size selection kits have been released, indicating beads-based size selection is possible for long DNA fragments.

Saves money: Reduces sequencing costs by focusing on the most valuable library material.

What Are the Common Issues without Proper Size Selection?

Too short
- - Sequencing adapter-dimers instead of actual library fragments, consume reads but produce no biological information.
  - Over-representation of short fragments, leading to inefficient use of sequencing reads
Too long: Poor cluster formation, low sequencing quality, produce few or low-quality reads
Broad distribution: Uneven coverage and data loss. Inaccurate assembly and difficult detection of structural variations.

How to Choose the Method for Library Size Selection?

1. Bead-based size selection: This is the most widely adopted method for high-throughput applications.

Technology: Magnetic beads/SPRI Beads
Selection type:
- - Left-side selection/clean-up: removes small fragments
  - Right-side selection/clean-up: removes large fragments
  - Double-sided selection/clean-up: selects a tight size range

Example of single-side size selection

Example of double-side size selection

Mechanism: It uses the principle of solid phase reversible immobilization (SPRI), where the ratio of PEG/salt/beads to DNA determines which fragment sizes bind to the beads. The beads method simplifies workflows by eliminating gels/spin columns and is adaptable for both single-sided and double-sided selections.
Pros: Highly scalable, reproducible, cost-effective, easily automated, and ideal for removing small fragments like adapter-dimers.
- - Automation & High-Throughput: Magnetic beads are ideal for automated liquid handlers, reducing manual pipetting and enabling processing of many samples efficiently.
  - Reproducibility: Automation leads to less variability and more consistent results across experiments.
  - Cost-Effectiveness: Reagent costs are often lower than other methods, and avoiding spin columns reduces overall costs.
  - Efficient Cleanup: Removes both small fragments (primer dimers, adapter dimers) and large fragments, enhances data quality, and maximizes usable reads.
  - Simplicity: Straightforward simple protocols.
Cons:
- - Less precise for very narrow size ranges compared to gel methods.
  - Large fragment selection (for PacBio, Oxford Nanopore) is a huge challenge for the beads-based methods. However recent progresses made it possible to get >5 kb and >10 kb.

2. Gel-based size selection: This approach physically separates fragments by electrophoresis.

Selection type
- - Agarose gel electrophoresis: Manual but highly customizable; allows precise excision of desired fragment sizes.
    - - Precise but labor-intensive and lower yield
      - Often used for special applications (e.g., mate-pair libraries)
  - Pippin Prep/BluePippin (Sage Science): Automated gel-based systems. Recent data shows these remain popular for challenging applications like ATAC-seq and mate-pair libraries.
Mechanism: DNA runs through an agarose gel, and the desired size range is physically extracted or electroeluted.
Pros:
- - Highly accurate for precise size ranges
  - reproducible
Cons:
- - Low throughput. Automated gel systems like Pippin Prep address some throughput issues but still have limitations compared to bead-based methods.
  - Labor-intensive and time-consuming.
  - Potential for contamination
  - DNA loss (typically 30-50%)

What is Coming Next?

The most recent developments highlight a strong emphasis on automation and integrated solutions. Long-read sequencing is also gaining more attention with the technology improving.

Automation is Key: The market for automated library preparation is expanding rapidly, driven by the need for consistency, reduced human error (e.g., pipetting errors), and increased throughput, especially for clinical diagnostics. Automated systems generally show higher consistency and library complexity than manual methods. The Beads-based size selection has the advantage over gel-based size selection.
Focusing on Long Reads: For long-read sequencing technologies (PacBio, Oxford Nanopore), the target fragment size is significantly larger (>10 kb) than the 200-500 bp for Illumina short-read platforms. The size selection for long reads focuses on preserving high molecular weight DNA and may use different methodologies that bypass traditional cleavage and amplification. Large fragment selection is still a hurdle for the beads-based methods, but recent technologies showed the successful selection of >5 kb and >10 kb.

Overall, magnetic bead-based automation is the dominant trend for short-read sequencing, offering a balance of efficiency, cost, and quality, while new chemistries and integrated systems continue to refine the process for various applications and sequencing platforms. For long-read sequencing, beads-based technology is also available now.

Conclusion

NGS library size selection remains a vital, albeit often overlooked, determinant of NGS success. While bead-based methods dominate due to simplicity and throughput, new technologies offer improved precision and recovery, particularly valuable for challenging samples. Recent data emphasizes that optimization should be application-specific, with increasing integration into automated workflows. As sequencing technologies evolve toward longer reads and lower inputs, size selection methods continue to adapt, balancing recovery, precision, and practicality.

Overall, library size selection is a cost-effective, high-impact step that transforms a potentially messy collection of DNA fragments into a high-quality, efficient sequencing library.

Ref

Boost Your NGS Results with Precise Library Size Selection