- Single-nucleotide variants (SNVs) are genetic variations in which a single nucleotide differs between DNA sequences. They occur when one nucleotide, such as adenine (A), cytosine (C), guanine (G), or thymine (T), is replaced by another at a specific genomic position. SNVs are among the most common forms of genetic variation in many organisms and are important in genetics, genomics, evolution, medicine, agriculture, and population biology. Their identification and interpretation are central activities in bioinformatics because even a single nucleotide change can influence gene function, gene regulation, protein structure, or an organism’s susceptibility to disease.
- An SNV is generally defined as a single-base difference between genomic sequences, without implying that the variant is common in a population. The term single-nucleotide polymorphism (SNP) is often used when a single-nucleotide variant is observed as a polymorphism within a population. In practice, the terms SNV and SNP are sometimes used interchangeably, although SNV is the broader term. A variant may be rare or common, inherited or newly arisen, and either biologically neutral or functionally important. The frequency of a variant is therefore an important consideration when describing its population distribution and potential significance.
- Single-nucleotide variants can be classified according to the nucleotide substitution involved. A transition occurs when a purine is replaced by another purine or a pyrimidine by another pyrimidine. The purines are adenine and guanine, while the pyrimidines are cytosine and thymine. Thus, A-to-G and C-to-T substitutions are examples of transitions. A transversion occurs when a purine is replaced by a pyrimidine or a pyrimidine by a purine. Examples include A-to-C, A-to-T, G-to-C, and G-to-T substitutions. Transitions are generally observed more frequently than transversions in many genomic datasets, although the relative rates vary among organisms, genomic regions, and biological processes.
- Single-nucleotide variants may occur in coding or noncoding regions of the genome. Coding-region variants are located within protein-coding sequences and may change the codon used during translation. A synonymous variant changes a codon without changing the encoded amino acid. Although traditionally considered less likely to affect protein function, synonymous variants can sometimes influence mRNA stability, splicing, translation efficiency, or gene expression. A missense variant changes a codon so that a different amino acid is incorporated into the protein. Its effect depends on the amino acid substitution, the position within the protein, the structural and functional importance of the region, and other biological factors.
- A nonsense variant introduces a premature stop codon into a protein-coding sequence. This can result in a shortened protein or trigger nonsense-mediated mRNA decay, a cellular mechanism that degrades certain transcripts containing premature termination codons. The biological consequences depend on the location of the variant, the transcript involved, and whether the resulting protein retains functional activity. Variants near the beginning of a coding sequence may have different consequences from those near the end.
- Single-nucleotide variants can also affect splice sites and regulatory regions. Splice-site variants may disrupt the recognition of exon–intron boundaries or alter the selection of splice sites, potentially producing abnormal transcripts. Variants in promoters, enhancers, silencers, untranslated regions, and other regulatory elements may influence transcription, RNA processing, translation, or mRNA stability. A variant outside a protein-coding region should not automatically be considered functionally unimportant because noncoding regions contain many regulatory elements and noncoding RNA genes.
- The effect of an SNV depends partly on its location within a gene and its relationship to the gene’s transcript structure. A single genomic position may overlap different transcripts, and the same variant can have different consequences for different transcript isoforms. For example, a nucleotide substitution may be intronic in one transcript but exonic in another. It may also be synonymous in one coding transcript and missense in a different transcript because of differences in reading frame or exon structure. Therefore, variant annotation should specify the transcript reference used and should consider biologically relevant transcript isoforms.
- SNVs may arise through several molecular mechanisms. DNA replication errors can introduce incorrect nucleotides when DNA polymerases copy genetic material. Although DNA proofreading and repair systems correct many such errors, some changes remain and become permanent mutations. Chemical modification of DNA bases, spontaneous deamination, oxidative damage, exposure to ultraviolet radiation, and certain chemical mutagens can also contribute to nucleotide substitutions. The distribution of SNVs across a genome reflects mutation processes, DNA repair, replication, selection, recombination, and population history.
- Some SNVs arise in germline cells and can be inherited by offspring, while others arise in somatic cells and remain limited to particular tissues or cell lineages. Germline variants contribute to inherited genetic diversity and may be transmitted across generations. Somatic SNVs are important in cancer research because they can accumulate during cell division and contribute to the development and progression of tumours. A variant detected in a tumour sample may be somatic, germline, or present in both, so comparison with an appropriate normal sample can be important when investigating cancer-associated variation.
- The frequency of an SNV in a population is commonly described using allele frequency. For a biallelic variant, the two alleles are often designated as the reference allele and the alternative allele. In a diploid organism, individuals may carry two reference alleles, one reference and one alternative allele, or two alternative alleles. These genotypes are commonly represented as 0/0, 0/1, and 1/1 in a typical diploid VCF record. The 0 represents the reference allele, while 1 represents the first alternative allele. This notation is a simplified description and must be interpreted in relation to ploidy, phasing, and the variant record.
- A single-nucleotide variant can be identified through several experimental approaches. Whole-genome sequencing examines DNA across most of the genome and can detect SNVs in coding and noncoding regions. Whole-exome sequencing focuses primarily on protein-coding regions and nearby sequences, reducing the amount of DNA that must be sequenced but excluding most intergenic and many regulatory regions. Targeted sequencing examines selected genes or genomic regions and is often used when particular variants or disease-associated genes are of interest. Genotyping arrays detect predefined variants at selected genomic positions and are widely used in population genetics, association studies, and genomic selection. However, arrays generally do not provide the same broad discovery capability as whole-genome sequencing.
- Short-read sequencing technologies, particularly Illumina sequencing, are widely used for SNV discovery because they can generate large quantities of accurate sequence data. Long-read technologies, including PacBio and Oxford Nanopore platforms, can also detect SNVs and may provide advantages in repetitive or structurally complex regions, especially when reads are sufficiently accurate or supported by consensus methods. The choice of technology depends on the research objective, genome complexity, required accuracy, sample type, sequencing depth, and available resources.
- The identification of SNVs from sequencing data usually begins with raw-read quality control. Sequencing reads are assessed for base quality, adapter contamination, sequence composition, and other technical problems. Tools such as FastQC and MultiQC are commonly used for quality assessment, while fastp and Cutadapt can remove adapters and low-quality sequence. Poor-quality reads and technical artifacts can reduce alignment accuracy and increase false-positive variant calls, making preprocessing an important step in reliable SNV analysis.
- After quality control, sequencing reads are aligned to a suitable reference genome. Short-read aligners such as BWA-MEM2 and Bowtie2 are commonly used for DNA sequencing data, while long-read aligners such as minimap2 are used for suitable long-read datasets. The resulting alignments are stored in SAM, BAM, or CRAM formats. Alignment files are generally sorted and indexed, and additional processing may include duplicate marking, read-group assignment, and quality assessment. Mapping quality is important because reads aligned to repetitive or highly similar genomic regions may not have a unique or reliable placement.
- Variant-calling software examines aligned reads and identifies genomic positions where the observed bases differ from the reference sequence. The caller evaluates evidence such as the number of reads supporting each allele, base quality, mapping quality, strand distribution, and local sequence context. Common SNV callers include GATK HaplotypeCaller, FreeBayes, bcftools, and DeepVariant. These tools use different algorithms and evidence models, so their results may not be identical. The appropriate caller depends on the organism, sequencing technology, ploidy, sample design, and analysis objective.
- GATK HaplotypeCaller uses local haplotype assembly to evaluate candidate variation in regions of the genome. FreeBayes uses a Bayesian framework to identify variants from aligned reads and can support different ploidy settings. The bcftools variant-calling tools provide a flexible approach based on read alignment data and can be used in both individual and population-level analyses. DeepVariant applies machine-learning methods to predict genetic variants from sequencing evidence. These tools can produce high-quality results, but their accuracy depends on input data quality, reference compatibility, parameter selection, and the biological context.
- Single-nucleotide variants are commonly stored in Variant Call Format (VCF) files. A VCF record includes the chromosome or sequence identifier, genomic position, variant identifier, reference allele, alternative allele, quality score, filtering status, and additional information. Genotype information can be included for individual samples. A simplified SNV record is shown below:
##fileformat=VCFv4.3
##reference=GRCh38
#CHROM POS ID REF ALT QUAL FILTER INFO
chr1 123456 . A G 60 PASS .
- This example describes a substitution from A to G at position 123456 on chromosome 1, relative to the GRCh38 reference assembly. The QUAL field represents the variant quality score, while FILTER indicates whether the record passed the specified filtering criteria. In a complete VCF file, metadata declarations should define the fields used in the header, and the file should identify the reference sequence and relevant sample information.
- Genotype information can be included in the VCF FORMAT and sample columns. For example:
##fileformat=VCFv4.3
##FORMAT=<ID=GT,Number=1,Type=String,Description=”Genotype”>
#CHROM POS ID REF ALT QUAL FILTER INFO FORMAT Sample1
chr1 123456 . A G 60 PASS . GT 0/1
- In this example, the sample has a heterozygous genotype, meaning that one chromosome copy carries the reference allele A and the other carries the alternative allele G, assuming a diploid organism and an unphased genotype call. A genotype of 0/0 would indicate two reference alleles, while 1/1 would indicate two copies of the alternative allele. A phased genotype may be represented using a vertical bar, such as 0|1, to indicate that the alleles have been assigned to particular haplotypes.
- Variant quality assessment is essential for distinguishing genuine SNVs from sequencing and alignment errors. Important metrics include read depth, genotype quality, variant quality, mapping quality, allele balance, strand bias, and base quality. Read depth describes the number of sequencing reads covering a genomic position. Genotype quality estimates confidence in the assigned genotype, while mapping quality reflects confidence in the placement of reads. Allele balance measures the proportion of reads supporting the alternative allele relative to the total reads supporting the reference and alternative alleles. For a typical heterozygous SNV in a diploid sample, the alternative allele fraction may be near 0.5, although biological and technical factors can produce deviations.
- SNV filtering is performed to remove low-confidence calls and reduce false-positive results. Filtering strategies may consider minimum read depth, genotype quality, mapping quality, allele balance, strand bias, and other caller-specific metrics. Some analyses use hard filters based on predefined thresholds, while others use statistical or machine-learning approaches such as variant quality score recalibration. Filtering criteria should be selected according to the sequencing platform, sample type, variant-calling method, and research question. Applying overly strict filters may remove genuine variants, while weak filtering can retain technical artifacts.
- Variant normalization is also important for reliable comparison and downstream interpretation. Although SNVs are usually straightforward to represent, multiallelic records and variants near insertions or deletions may require careful normalization. Reference and alternative alleles should be checked against the correct genome assembly, and chromosome naming conventions should be consistent. Tools such as bcftools norm can help normalize variant records and validate reference alleles. When combining datasets, differences in assembly versions, variant representation, and filtering procedures should be considered before comparing calls.
- After SNV discovery, annotation connects genomic positions with genes, transcripts, regulatory regions, and predicted biological effects. Tools such as Ensembl Variant Effect Predictor (VEP), SnpEff, and ANNOVAR can identify whether an SNV is synonymous, missense, nonsense, splice-related, intronic, intergenic, or located in a regulatory region. Annotation may also include population allele frequencies, known clinical classifications, conservation scores, and computational predictions of functional impact. These annotations provide evidence for interpretation but do not independently prove that a variant causes a disease or affects a particular trait.
- In medical genetics, SNVs are important because some are associated with inherited disorders, cancer susceptibility, drug response, and other clinical characteristics. A pathogenic SNV may disrupt protein function, alter splicing, affect gene expression, or change an important regulatory process. However, many SNVs are benign or have uncertain clinical significance. Clinical interpretation considers the variant’s frequency in relevant populations, molecular consequence, inheritance pattern, segregation with disease, functional evidence, phenotype compatibility, and available clinical literature. Resources such as ClinVar, ClinGen, gnomAD, and disease-specific databases can support interpretation, but their evidence and limitations must be assessed carefully.
- In cancer genomics, somatic SNVs are used to investigate tumour development, clonal evolution, treatment resistance, and potential therapeutic targets. Some mutations are considered driver mutations because they contribute to cancer development or progression, while others are passenger mutations that accumulate without providing a selective advantage to tumour cells. Distinguishing these categories requires statistical analysis, biological knowledge, functional evidence, and consideration of tumour purity and sequencing characteristics. Paired tumour-normal sequencing can help separate somatic variants from inherited germline variation.
- SNVs are also widely used in population genetics and evolutionary studies. Researchers examine allele frequencies and patterns of variation to investigate population structure, genetic diversity, demographic history, migration, admixture, natural selection, and evolutionary relationships. Genome-wide SNV datasets can be used for principal component analysis, ancestry estimation, genetic differentiation, and genome-wide association studies. However, population structure and relatedness can confound association analyses if they are not properly accounted for. Sampling design, marker selection, sequencing depth, and reference bias can also influence results.
- In plant and animal breeding, SNVs are important molecular markers for identifying genetic differences associated with economically important traits. They are used in genome-wide association studies, marker-assisted selection, genomic selection, parentage verification, breed characterization, and genetic diversity monitoring. In livestock, SNVs may be associated with milk production, growth, fertility, disease resistance, meat quality, and feed efficiency. In crop plants, they can help identify alleles associated with yield, flowering time, drought tolerance, disease resistance, and nutritional quality. A statistical association between an SNV and a trait does not necessarily mean that the variant directly causes the trait; it may be linked to another causal variant.
- SNV analysis has several limitations. Sequencing errors, low coverage, contamination, sample swaps, mapping ambiguity, and reference bias can affect variant detection. Rare variants may be difficult to distinguish from technical errors, especially in datasets with limited sequencing depth. Highly repetitive regions and paralogous genes can cause incorrect read placement, while structural variation may interfere with alignment and variant calling. Different variant callers may produce different results because they use different algorithms and assumptions. Independent validation may be appropriate for important findings, particularly in clinical or experimental applications.
- For reliable and reproducible SNV analysis, researchers should document the reference assembly, sequencing platform, read-processing steps, aligner, variant caller, software versions, parameters, and filtering criteria. Sample identifiers and metadata should be managed carefully, and quality-control results should be preserved. Variant records should be normalized and checked against the correct reference sequence before annotation or comparison. When reporting an SNV, the genomic coordinate, reference and alternative alleles, assembly version, and relevant transcript or protein description should be provided where appropriate.
- Single-nucleotide variants are fundamental units of genetic diversity and important markers for understanding genome function, inherited disease, evolution, and biological traits. Their analysis involves sequencing, alignment, variant calling, quality assessment, filtering, normalization, and functional annotation. Accurate interpretation requires attention to reference assemblies, population frequencies, transcript structure, technical limitations, and biological context. As sequencing technologies and computational methods continue to improve, SNV analysis will remain central to medical genomics, evolutionary research, precision medicine, agriculture, and modern bioinformatics.