Genomics & Genome Structure

What is genomics?

The study of all an organism's genes, or its genome.

• Studies the structure, function, evolution, and mapping of genomes

• Broader than simply reading DNA sequence

Why is genomics a multidisciplinary field?

It integrates molecular biology, genetics, sequencing technology, statistics, computing, and data science.

• Requires both biological and computational expertise.

What is structural genomics?

Studies the physical nature of genomes, including the sequencing and mapping of genomes. It provides the structural and DNA-level variation data used in later analyses.

• Genome sequencing

• Genetic variants detection such as SNP, Indels, CNV

• QTL mapping

• GWAS

• Other applications of structure and quantitative genomics

What is functional genomics?

Studies the expression and function of the entire genome. It asks what genomic elements do rather than only where they are located.

• Transcriptome (RNA-sequencing)

• CHIP-Seq

• Genome Editing

• eQTL & Gene Enrichment and Network Analyses

• Epigenomics

• Meta-transcriptome

What is comparative genomics?

Compares genomes from different organisms. Similarities and differences across species can reveal conserved functions and evolutionary relationships or changes across genomes.

Comparative Genomics example: what gene illustrates an evolutionarily conserved gene required for normal muscle development?

Knowledge of gene structure/function in one species can be applicable or guide interpretation in another.

• Example: mutations in myostatin can alter muscle development across multiple species

What four factors have shaped the rise of genomics?

• Advances in sequencing technology and reducing in the cost of sequencing

• Computational and data-science innovation

  • Ex: availability of high-performance computing (HPC) facilities

• Advances in data analytics such as machine learning (ML)

• Shift from one-gene studies to genome-wide studies.

Why has reduced sequencing cost been important for genomics?

Lower sequencing cost makes it feasible to obtain genomic data from many individuals or entire populations. Large sample sizes improve discovery and prediction.

Why is high-performance computing important in genomics?

Provides the storage and processing power needed for large genomic datasets. Sequencing and variant analyses can involve billions of bases and many samples.

What is the major computational challenge in genomics identified in the lecture?

The major challenge is big-data computation and analysis. Generating sequence data is only useful if it can be processed and interpreted.

Name three cloud-computing providers listed as examples for genomics work.

Cloud platforms can supply scalable computing resources.

• Amazon Web Services

• Google Cloud

• DigitalOcean

• Microsoft Azure

What major sequencing platforms are named in the lecture, and how do their read types differ?

Read length and chemistry shape each platform's strengths.

• Sanger uses capillary electrophoresis

• Illumina uses short-read sequencing by synthesis

• Nanopore uses long reads

• PacBio HiFi uses highly accurate long reads

What general trend in sequencing output is illustrated by the Illumina platform slides?

• Sequencing platforms can generate extremely large amounts of nucleotide data per day

• This increasing output is a key source of genomics big-data challenges

What is a reference genome?

• A high-quality, preassembled, standardized genomic map used as a template.

• Sequencing reads and variants are commonly interpreted relative to this reference.

What is read depth, also called depth of coverage?

• The number of sequencing reads that cover a particular base or genomic position

• Deeper coverage generally increases confidence in a called base or variant.

What is breadth of coverage?

• The fraction of the reference genome that has been covered by reads

• A dataset can have deep coverage at some sites but poor coverage across the genome overall.

How do depth and breadth of coverage differ in the diagram?

• Depth is the vertical number of overlapping reads at a position

• Breadth is the horizontal extent of the reference genome covered by reads

  • The reference genome is a high-quality, pre-assembled, standardized map

• The two measures describe different aspects of sequencing completeness.

What is variant calling?

• The process of identifying differences between sequencing data and a reference genome

• it converts aligned reads into candidate genetic variants and genotypes.

What are the broad steps in a generalized variant-calling workflow?

1. Reads are aligned to a reference genome

2. Variants are called

3. The calls are filtered and annotated or interpreted

• Quality control is needed before a detected difference becomes a reliable variant.

What are the two broad classes of genetic variants in the lecture?

Short variants (i.e SNPs or INDELs) and structural variants (CNVs and STRs). The size and genomic effect of the variant determine its category.

Which variants are considered short genetic variants?

• SNPs or SNVs (single-nucleotide variant)

• INDELs (insertion/deletion)

• They involve a single nucleotide or a relatively small insertion or deletion.

Which variants are considered structural variants in the lecture?

• CNVs (copy-number variants)

• STRs (short tandem repeats)

• Microsatellites

• They involve larger-scale changes in genomic structure or repeat number.

What is a genotype?

The combination of alleles at one locus or multiple loci in an individual. Genotyping determines which alleles an animal carries.

• Each animal inherits two sets of alleles (one from dam, one from sire)

What is a SNP?

• A single-nucleotide polymorphism is a single-base substitution at a specific genomic position with a minor allele frequency greater than 1%.

• Using sequencing or genotyping assays it’s possible to determine which alleles (i.e. genotype) animals have for a particular SNP locus

What is the difference between homozygous and heterozygous at a SNP locus?

• Homozygous individual has two copies of the same allele, such as A/A or G/G,

• Heterozygous individual has two different alleles, such as A/G.

An animal has maternal allele A and paternal allele G at a SNP. What is its genotype and zygosity?

• Genotype is A/G

• It is heterozygous

• The two inherited alleles differ

What is a missense SNP?

Changes a codon so that one amino acid is substituted for another in a protein, which can alter protein function and phenotype.

• Example: DGAT1 in cattle which can produce alanine instead of lysine in position 232 of the protein sequence

• Individuals that express the lysine version produce more milk with higher fat content

What is an INDEL?

• A small insertion or deletion of DNA sequence, generally defined as less than 1,000 base pairs

• One genome copy may contain a short sequence that another lacks

• Can change peptide protein sequences and then gene function, especially when it shifts the codon reading frame

What is a frameshift mutation?

An insertion or deletion that changes the reading frame of downstream codons. It can dramatically alter the amino-acid sequence and often disrupt protein function.

What myostatin mutation causes double muscling in Belgian Blue and Piedmontese cattle?

• An 11-base-pair deletion in the third exon of the myostatin, or MSTN, gene causes loss of 102 AAs due to a frameshift

• The deletion eliminates functional myostatin, so the gene is incomplete autosomal dominance.

  • Means that the phenotype of heterozygotes is intermediate rather than identical to either homozygote

  • MSTN mutation's effect depends on how many mutant alleles are present.

What is a copy-number variant, or CNV?

A DNA segment for which different individuals have different copy numbers. The segment can range from about one kilobase to several megabases and may contain multiple genes.

How can a CNV genotype be homozygous or heterozygous?

An individual can carry two copies of the same copy-number allele or two different copy-number alleles. CNVs, like SNPs and INDELs, have genotypes.

What gene is associated with the CNV coat-color examples in pigs?

The KIT gene, which encodes the mast or stem-cell growth-factor receptor.

• Can produce a belted phenotype due to duplications of regulatory elements upstream and downstream of the KIT locus

What two mechanisms of CNV formation are named in the lecture?

• Nonallelic homologous recombination, or NAHR

• Fork stalling and template switching, or FoSTeS

  • Both can create copy-number changes during meiosis

What is a commercial SNP chip?

A genotyping array that measures many predefined SNP markers at once. it provides a lower-cost alternative to sequencing every base in every animal.

What is genotype imputation?

A statistical inference of unobserved genotypes using observed markers and a reference population. It can increase marker density without measuring every SNP directly.

What applications of structural or quantitative genomics are shown in the lecture?

• Genomic selection

• QTL mapping or GWAS

• Marker panels

• Biomarker development

• Heterosis

• Traceability and parentage verification

• Haplotypes and genetic recessives

• Genetic diversity, and understanding biology

What was the typical bovine-genomics approach in the 1990s vs 2001?

• 1990s emphasized DNA markers, linkage mapping, and marker-assisted selection. Studies often used one or a small number of markers.

• 2001: proposed genome-wide selection based on linkage disequilibrium. This shifted prediction from a few markers toward dense genome-wide marker information

What is linkage mapping?

Uses marker inheritance patterns to locate a specific gene trait (QTL) relative to one or a small number of markers. It relies on genetic linkage between the marker and trait locus.

How does GWAS differ from early linkage mapping?

GWAS tests associations between a trait and many genome-wide markers. Dense marker panels use linkage disequilibrium to improve resolution.

What is marker-assisted selection, or MAS?

MAS uses one or a small number of genetic markers linked to desirable trait loci to guide selection. It predates genome-wide genomic selection.

What has replaced many physical genetic maps and linkage maps in modern animal genomics?

Genome-sequence assemblies have replaced them in many species because assembled sequences provide a more complete genomic reference.

How has QTL mapping changed in modern genomics, and what has stayed the same?

• It’s shifted from few DNA-marker or interval mapping approaches to high-density panels and GWAS (genome-wide association)

  • But the underlying principle of associating genomic variation with traits remains the same

• Technology changed the resolution, not the core goal.

How does genomic prediction differ from traditional pedigree-based genetic evaluation?

• Genomic prediction estimates genetic merit using DNA markers alone or with other data, while traditional evaluation relies more heavily on pedigree and performance records

• Genome prediction is becoming the major genetic evaluation tool

• Dense markers capture realized genomic relationships.

Why do full-genome sequencing and molecular technologies help identify QTL?

• New molecular technologies include RNAseq and allele expression

• They can identify causative DNA polymorphisms for major genes and may also reveal genes with smaller effects

• Genome-wide data increase the chance of finding trait-associated variation.