Next-Generation Sequencing (NGS)

Next-generation sequencing (NGS), also called massively parallel sequencing, enables simultaneous sequencing of millions of DNA fragments. NGS has transformed clinical genetics by making multi-gene panels, whole exome sequencing (WES), and whole genome sequencing (WGS) practical and cost-effective, dramatically increasing diagnostic yield for genetic disorders.

• Library preparation: Patient DNA is fragmented, adapters are ligated to the ends, and fragments are amplified. This prepares the DNA for the sequencing platform.
• Sequencing by synthesis (Illumina): The most widely used NGS platform. Fluorescently labeled nucleotides are incorporated one at a time into growing strands, with imaging after each cycle. Produces short reads (~150–300 bp).
• Coverage and depth: Coverage refers to the percentage of the target region that is sequenced. Depth (e.g., 30x, 100x) is the average number of times each base is read. Higher depth increases confidence in variant calling. Clinical germline testing typically requires 20–30x minimum; somatic testing requires 500–1000x or more.
• Types of clinical NGS:
  • Gene panels: Targeted sequencing of a curated set of genes relevant to a specific phenotype (e.g., cardiomyopathy panel, hereditary cancer panel). Highest depth, lowest cost, fewest incidental findings.
  • Whole exome sequencing (WES): Captures and sequences all ~20,000 protein-coding genes (~1–2% of the genome). Commonly used for undiagnosed rare diseases. Diagnostic yield ~25–40% in selected populations.
  • Whole genome sequencing (WGS): Sequences the entire genome including introns, regulatory regions, and intergenic sequences. Detects structural variants and non-coding variants that WES misses. Increasingly used as a first-line test.
• Germline vs somatic testing: Germline testing analyzes constitutional DNA (typically blood) for inherited variants. Somatic testing analyzes tumor DNA for acquired mutations to guide cancer treatment. Somatic testing requires higher depth due to tumor heterogeneity and admixture with normal tissue.
• Bioinformatics pipeline: Raw sequencing data undergoes (1) base calling, converting signals to nucleotide sequences; (2) alignment, mapping reads to a reference genome (e.g., GRCh37/hg19 or GRCh38/hg38); (3) variant calling, identifying positions that differ from the reference; (4) annotation, adding gene, protein, and database information; (5) filtering and interpretation, applying ACMG criteria and clinical correlation.
• Paired-end sequencing: Both ends of each fragment are sequenced, improving alignment accuracy and structural variant detection.

• Diagnostic odyssey: NGS (especially WES/WGS) is recommended for patients with undiagnosed suspected genetic conditions after targeted testing is negative
• Trio analysis: Sequencing the proband plus both parents enables identification of de novo variants and assists with phasing and interpretation
• Pharmacogenomics: Panel-based NGS for drug-metabolizing enzyme variants
• Hereditary cancer screening: Multi-gene panels for breast, ovarian, colon, and other hereditary cancer syndromes
• Prenatal/neonatal: Rapid WGS (rWGS) in critically ill neonates can provide diagnoses within 24–48 hours, changing management in ~30% of cases
• Reanalysis: Previously negative NGS data can be periodically reanalyzed as new gene-disease associations are discovered

• GC-rich and repetitive regions: NGS short reads have difficulty with highly repetitive sequences, GC-rich regions, pseudogenes, and segmental duplications
• Trinucleotide repeat expansions: Standard NGS cannot reliably size large repeat expansions (e.g., Huntington disease, fragile X). Long-read sequencing (PacBio, Oxford Nanopore) addresses this.
• Structural variants: Short-read NGS has limited sensitivity for balanced translocations, inversions, and some large insertions. WGS is better than WES for SV detection.
• Variant interpretation bottleneck: Generates thousands of variants per patient; filtering to the causative variant requires robust bioinformatics and clinical expertise
• VUS burden: The more genes sequenced, the more variants of uncertain significance identified
• Incidental/secondary findings: ACMG recommends reporting pathogenic/likely pathogenic variants in a defined set of actionable genes (ACMG SF list, currently v3.2), which may reveal unexpected conditions
• Coverage gaps: WES may miss variants in poorly captured exons; confirmatory Sanger sequencing may be needed
• Does not detect methylation abnormalities: Standard NGS does not assess epigenetic changes