Next-generation sequencing: Sequencing depth vs. sequencing batching

In hematological oncology, where the detection of rare variants (such as measurable residual disease or emerging clonal populations) which often have low variant allele frequencies (VAF, the proportion of sequence reads that contain a specific variant relative to the total number of reads at that position) is often critical, the balance between depth and batch size is a key consideration.

Understanding sequencing depth

Sequencing depth is the number of times a nucleotide is read during a sequencing run and the mean target coverage depth (MTC) is the average number of reads for a specific targeted region; for example, an MTC of 1000x means that on average, that region in your sample has been sequenced 1000 times.

The limit of detection for NGS sequencing, and thus the capability to detect lower frequency variants, is directly related to the depth of sequencing performed. For example, if we consider a cancer sample where most of the cells are malignant (and thus carry the disease-associated variant at a high VAF), we can assume that a moderate sequencing depth would detect this variant. However, in cases such as measurable residual disease (MRD) testing, where variants may only be present in a very small fraction of cells (i.e. low VAF), a high sequencing depth is required to confidently detect these low frequency variants.

Several factors are relevant to depth requirements, including:

• Flow cell capacity: Different flow cells will offer different sizes of data output, which is related to how much depth you can achieve (i.e. a flow cell that can be used to output 40 GB of sequencing data will offer less depth than a flow cell that can output 120 GB of sequencing data for the same number of samples)
• Sample input: Degraded or low-input DNA may present with high levels of technical noise that may not be corrected for with higher sequencing
• Bioinformatics pipeline: Error-correction algorithms and variant calling thresholds can mitigate some of the issues inherent in lower-depth sequencing
• Cost constraints: Higher depth typically increases costs, so laboratories must balance the need for sensitivity with budgetary limitations

Sequencing batching explained

Sequencing batching is a strategy whereby multiple samples are pooled after preparation to be sequenced simultaneously. For example, multiple genetic samples may each be tagged with unique identifiers during the library preparation process. Samples can then be pooled together and sequenced on the same sequencing batch. The unique molecular identifiers are then used to differentiate samples after sequencing using a bioinformatic pipeline for analysis and differentiation of sample information.

This approach can potentially offer labs cost and time efficiencies by maximizing the use of a sequencing instrument’s capacity, reducing per-sample costs and user hands-on time as well as increasing sample throughput. But we must consider that a sequencer’s total capacity will be divided among those pooled samples. If too many samples are batched together, each sample may receive insufficient depth for reliable variant detection.

In hematological oncology, where the detection of rare variants can be clinically significant, balancing sequencing depth with batching strategies can be challenging. While the theory behind sequencing depth and batching is clear, practical implementation in a laboratory involves several additional considerations.

Library preparation and sample quality

• Quality control: High-quality input DNA and careful library preparation can ensure carry-through of unique molecules through to sequencing and reduce technical variability, which can otherwise necessitate even higher sequencing depths
• Unique Molecular Identifiers (UMIs): Incorporating UMIs can help distinguish true variants from PCR or sequencing artifacts, particularly in samples with low VAF increasing confidence for variant detection
• Pre-run validation: Pilot runs using known standards or reference samples can help validate the chosen batch size, depth and limit of detection, ensuring that the assay performs as expected in a clinical setting

Cost and turnaround time

• Budget constraints: High sequencing depth per sample increases costs. Laboratories should evaluate the level of detection required and associated potential clinical benefit to detect lower frequency variants against the expense of achieving this detection.
• Turnaround time: Larger batches may improve cost efficiency but can delay individual results if the laboratory waits for a full batch to accumulate before running a sequencing run.*
• Optimization strategies: Balancing cost with rapid turnaround times may require running smaller batches at higher depth for cases where rapid decision-making is critical, while larger batches may be acceptable for routine assays.*

(*however this can be dependent upon flow cell capacity)

Bioinformatics and data analysis

• Variant calling pipelines: The sensitivity of variant detection is not only determined by depth of sequencing, but also by the sophistication of bioinformatic pipelines. Error-correction methods, threshold settings, and statistical models all influence the final results.
• Quality metrics: Post-sequencing metrics such as uniformity of coverage, on-target percentage, and duplication rates must be monitored to ensure that the balance between depth and batching has not compromised assay quality.
• Continuous optimization: As new bioinformatic tools become available, laboratories should re-assess their pipelines to maintain or improve sensitivity without disproportionately increasing sequencing costs.

Through careful planning and ongoing optimization, laboratories can ensure that their NGS protocols are not only cost-effective but also sufficiently sensitive to detect subtle yet critical genomic changes. By understanding and applying these principles, you can design NGS workflows that are robust, sensitive, and tailored to the unique challenges of hematological oncology genomics.

View our range of SureSeq hematological malignancy NGS panels