1: Sequencing Technologies & NGS Chemistries

Chapter 1: Sequencing Technologies & NGS Chemistries

High-Throughput Sequencing (HTS), often termed Next-Generation Sequencing (NGS), has transformed biological discovery from low-throughput Sanger capillary electrophoresis into massive parallel architectures yielding billions of base pairs in hours. Understanding the biochemical mechanisms underlying these platforms is essential for diagnosing error profiles, understanding insert size biases, and picking the right tool for genomic analysis.

1. Illumina Sequencing-by-Synthesis (SBS)

Illumina remains the dominant short-read sequencing platform worldwide. Its workflow comprises four sequential phases:

• Bridge Amplification / Exclusion Amplification (ExAmp): Genomic fragments flanked by adapters are hybridized to a patterned flow cell coated with P5 and P7 oligos. Clusters of identical clones are generated via isothermal bridge amplification or ExAmp chemistry to amplify fluorescence signals.
• Reversible Terminator Chemistry: All four dNTPs (A, C, G, T) are modified with a 3'-O-azidomethyl blocking group and a cleavable fluorescent fluorophore. In each sequencing cycle, a DNA polymerase incorporates exactly one complementary nucleotide.
• Fluorescence Imaging: The flow cell is scanned with lasers to excite the fluorophores. Two-channel systems (NextSeq, NovaSeq 6000) use red and green emissions (where yellow/both indicates T, green indicates G, red indicates C, and dark/none indicates A), while NovaSeq X utilizes ultra-rapid blue and green optics.
• Cleavage and Deprotection: Chemical reagents cleave the fluorophore and restore the 3'-OH group, preparing the nascent strand for the subsequent cycle.

2. Long-Read Sequencing: PacBio vs. Oxford Nanopore

While short-read Illumina sequencing (150 bp paired-end) excels at single nucleotide variant (SNV) detection, repetitive genomic regions, centromeres, and large structural variants require long-read technologies:

• Pacific Biosciences (PacBio) HiFi: Uses Zero-Mode Waveguides (ZMWs) and circular consensus sequencing (CCS). A double-stranded DNA template is ligated to hairpin adapters forming a dumbbell structure ("SMRTbell"). DNA polymerase continuously loops around the template multiple times (e.g., 10–15 passes), generating independent subreads whose consensus yields long reads (15–25 kb) with >99.9% base accuracy (Q30+).
• Oxford Nanopore Technologies (ONT): Translocates native single-stranded DNA or RNA molecules through protein nanopores embedded in an electrically resistant synthetic membrane. As nucleotides pass through the pore constriction, they perturb the ionic current. Deep learning neural networks (e.g. dorado, guppy) decode the raw current "squiggles" into base sequences. Reads can exceed 100 kb up to 4+ megabases ("ultra-long reads").

3. Comprehensive Platform Benchmark

# Platform Architecture Summary:
# -----------------------------------------------------------------------------------------
# Platform       Chemistry            Avg Read Length    Accuracy     Primary Strength
# -----------------------------------------------------------------------------------------
# NovaSeq X      2-Channel SBS        2 x 150 bp         Q30 (>99.9%) High throughput, lowest $/Gb
# PacBio Revio   SMRT CCS (HiFi)      15 - 20 kb         Q30 (>99.9%) Structural variants, haplotypes
# ONT PromethION Nanopore Current     10 - 100+ kb       Q20 - Q28    Native methylation, telomeres
# Ultima UG 100  Emulsion PCR / OLS   250 - 300 bp       Q30 (>99.9%) Ultra-cheap short reads ($1/Gb)
# -----------------------------------------------------------------------------------------