Next-Generation Sequencing (NGS) Explained for Beginners

Next-generation sequencing (NGS) reads millions of DNA fragments in parallel, so a lab can sequence a whole bacterial genome, a panel of cancer genes or a sample's entire transcriptome in one run. Sanger sequencing, by contrast, reads one fragment per reaction and remains the reference method for confirming a single short target.

NGS is now used in public health, food safety, hospital and research labs across India and the GCC. Even if you never load a sequencer, you may collect samples, interpret reports or audit the workflow.

Sanger sequencing vs NGS

Sanger sequencing uses chain-terminating dideoxynucleotides and capillary electrophoresis to read a single PCR product, typically several hundred to around a thousand bases. It is accurate, easy to interpret and cost-effective for a few targets, such as confirming a mutation or identifying an isolate from its 16S rRNA gene.

NGS sequences huge numbers of fragments at once and reads each base many times. That allows detection of mixed populations and low-frequency variants that Sanger cannot reliably see. The trade-off is more complex library preparation, much larger data files and a dependence on bioinformatics.

Short-read vs long-read sequencing

The most widely used short-read approach is sequencing by synthesis. Fragments are attached to a flow cell and amplified into clusters, then fluorescently labelled, reversibly terminated nucleotides are added one base per cycle and imaged. Reads are typically up to a few hundred bases, often from both ends of each fragment (paired-end). Ion semiconductor sequencing is another short-read method, detecting the hydrogen ions released as bases are added.

Nanopore sequencing passes a single DNA or RNA strand through a protein pore and reads changes in electrical current. Reads can reach tens of thousands of bases or more, data appear in real time and some devices are portable. Single-molecule real-time (SMRT) sequencing watches one polymerase incorporating labelled nucleotides; repeated passes around a circular template give highly accurate long reads, known as HiFi reads.

Long reads span repeats and structural variants and help close complete bacterial chromosomes and plasmids. Short reads remain the workhorse for high-accuracy, high-throughput work, and many labs combine both.

The NGS workflow step by step

1. Extraction: obtain enough intact DNA or RNA and check quantity and purity.
2. Library preparation: fragment the nucleic acid, add adapters and sample-specific indexes (barcodes), and amplify if needed. Targeted methods use PCR amplicons or hybridisation capture.
3. Pooling and sequencing: indexed libraries are pooled and loaded together.
4. Demultiplexing: reads are assigned to samples by index and written as FASTQ files.
5. Bioinformatics: trim adapters and low-quality bases, align reads to a reference (BAM files) and call variants (VCF files), or assemble genomes de novo.
6. Interpretation and reporting: compare with databases and previous results, then report in a form the user can act on.

Key QC metrics

• Q30: a Phred quality score of 30 means a 1 in 1,000 chance that a base call is wrong, or 99.9% accuracy. Runs are often judged by the percentage of bases at or above Q30.
• Depth of coverage: the average number of reads covering each position. The minimum depth depends on the application and should be fixed during validation.
• Breadth of coverage: the proportion of the reference or target covered at that minimum depth.
• Other checks: reads per sample, duplication rate, contamination screening and, for assemblies, contig number and expected genome size.

Applications of NGS

• Whole genome sequencing (WGS) for outbreaks: comparing bacterial genomes by single nucleotide polymorphisms (SNPs) or core genome MLST (cgMLST) shows whether cases are closely related. Many public health networks have replaced pulsed-field gel electrophoresis (PFGE) with WGS.
• Antimicrobial resistance: WGS detects resistance genes and mutations, supporting AMR surveillance. Phenotypic susceptibility testing is still needed, because a gene does not always predict resistance.
• Metagenomics: sequencing all DNA in a sample, or a marker gene such as 16S rRNA, to profile microbial communities in gut, water, soil or food environments.
• RNA-seq: measuring gene expression across the transcriptome.
• Oncology panels: targeted sequencing of tumour DNA to guide treatment selection.
• Non-invasive prenatal testing (NIPT): analysing cell-free DNA in maternal blood to screen for common chromosomal aneuploidies. High-risk results need confirmation by diagnostic testing, and in India prenatal testing is governed by the PC&PNDT Act, which prohibits sex determination.

Limitations to understand

• Equipment, reagents, computing and data storage are costly, and batching samples to fill a run can delay turnaround.
• Short reads struggle with repeats, structural variants and highly similar gene families.
• Pipelines need validation, version control and skilled staff, because small parameter changes can alter results.
• Clinical NGS produces variants of uncertain significance and incidental findings that need clear policies and genetic counselling.
• Contamination and index misassignment can create false low-level results, so negative controls matter.

Labs offering NGS under ISO 15189:2022 or ISO/IEC 17025:2017 must validate the whole process, wet lab and dry lab together. For bacterial typing in the food chain, ISO 23418:2022 sets general requirements for whole genome sequencing.

Worked example: linking Salmonella cases in a food safety authority lab

A food safety authority laboratory in the GCC is notified of Salmonella Enteritidis infections in several people over two weeks. Interviews point to a restaurant serving an egg-based sauce. Inspectors collect leftover sauce, raw eggs and environmental swabs, and hospital labs forward patient isolates.

The lab sequences all isolates on a short-read platform. Before interpretation, the analyst confirms each genome passes the SOP criteria for Q30, depth, genome size and contamination. Genomes are then compared by SNP analysis:

The genomic data support the epidemiological link, and the egg isolate points investigators upstream to the supplier. The lab does not treat a SNP cut-off as proof on its own: relatedness is reported alongside epidemiological and traceability evidence, and the case moves to traceability and recall actions.

Common mistakes

• Submitting degraded or low-yield DNA and expecting a good library.
• Judging a run by total output while ignoring Q30, depth and contamination checks.
• Treating a fixed SNP threshold as proof of a common source.
• Reporting resistance genes as if they were susceptibility results.
• Updating pipeline software or databases without revalidation.

Key takeaway: NGS reads millions of fragments in parallel, with short reads giving accuracy and throughput and long reads resolving complex regions. A sequencing result is only as good as its sample, QC metrics, validated pipeline and the context used to interpret it.

Frequently Asked Questions

What is the difference between NGS and Sanger sequencing?

Sanger sequencing reads one DNA fragment per reaction and is ideal for confirming a single target, such as a known mutation. NGS reads millions of fragments in parallel, so it can cover whole genomes, many genes or mixed samples at once and detect low-frequency variants. Labs often use Sanger to confirm selected NGS findings.

What does Q30 mean in sequencing?

Q30 is a Phred quality score indicating a 1 in 1,000 probability that a base call is incorrect, equivalent to 99.9% accuracy. Sequencing runs are often assessed by the percentage of bases at or above Q30. The acceptable percentage depends on the platform, application and the lab's validated acceptance criteria.

Is whole genome sequencing used in food safety?

Yes. Food safety and public health laboratories use WGS to type foodborne pathogens such as Salmonella, Listeria monocytogenes and E. coli, link patient cases to foods and production sites, and detect resistance and virulence genes. Results are interpreted together with epidemiological and traceability evidence rather than on genomic data alone.

Is NIPT a diagnostic test?

No. NIPT is a screening test that analyses cell-free DNA in the mother's blood to estimate the chance of certain chromosomal conditions. A high-risk result should be confirmed by a diagnostic test, such as amniocentesis or chorionic villus sampling, after counselling. Local laws on prenatal testing must also be followed.

Next steps

To turn FASTQ files into meaningful results, start with our Bioinformatics and Genomics Essentials course. If you are newer to molecular methods, PCR and Molecular Diagnostics Fundamentals covers the amplification and contamination control that NGS library work depends on.

Related reading: bioinformatics for biologists: where to start, foodborne pathogens: sources, symptoms and control and CRISPR gene editing explained. Need help choosing? Speak to a counsellor.