5 Genomics and Genome Analysis
Learn how genomes are organized, how assemblies and annotations are produced and evaluated, and how comparative genomics uses sequence and genome structure to investigate biological relationships.
organization
A is an organism’s complete set of genetic material. Most organisms store this information in DNA, while some viruses use RNA. size and structure vary widely and do not directly indicate an organism’s biological complexity.
In bacteria and archaea, genomes are commonly organized as one or more circular or linear chromosomes. They may also include smaller DNA molecules called plasmids. Eukaryotic genomes are divided among linear nuclear chromosomes; many eukaryotes also have mitochondrial DNA, and plants and algae typically have chloroplast DNA. In eukaryotic chromosomes, DNA is packaged with proteins into chromatin.
Genomes contain more than protein-coding genes. They also include genes for functional RNAs, regulatory sequences, repetitive DNA, and other regions whose functions may be known, uncertain, or still under study. Eukaryotic genes often contain exons, which remain in mature RNA, interrupted by introns, which are generally removed during RNA processing. Regulatory elements can influence when, where, and how strongly genes are expressed.
From reads to assemblies
A sequencer produces many reads, each an observation of part of a . A computationally combines overlapping reads to reconstruct longer sequences. It is a data-based representation of the rather than the itself, and its accuracy depends on the available reads and methods.
Assembly sequences are described at different levels of continuity:
A read is an individual sequencing observation.
A is a continuous assembled sequence with no gaps.
A orders and orients contigs, potentially with estimated gaps between them.
A chromosome-level assembly assigns sequences to chromosomes, but may still contain unresolved gaps.
Short reads can provide high-accuracy sequence but may be difficult to place in repetitive regions. Long reads can span many repeats and help connect distant regions. Genetic maps or chromosome-conformation data may also help order and orient sequences into scaffolds or chromosomes. Assembly pipelines may identify and correct errors, remove contamination, or separate different chromosome copies, called haplotypes.
Evaluating and tracking assemblies
Assembly quality should be assessed using several kinds of evidence rather than a single number. is a continuity statistic: after sequences are sorted from longest to shortest, it is the length at which sequences of that length or longer account for half of the total assembly length. A higher can indicate greater continuity, but it does not by itself show that the sequence is correct or complete.
Other checks include total length, gap count, read coverage, expected gene recovery, and consistency with independent biological evidence. Together, these checks help evaluate different aspects of an assembly rather than relying on continuity alone.
Assemblies are versioned because later releases may correct errors, close gaps, or add sequence. When reporting an analysis, specify the organism and the exact assembly accession and version. Coordinates and annotations are tied to a particular assembly, so coordinates from different versions may not match directly.
identifies and describes features on an assembled sequence. It involves two complementary tasks:
Structural annotation locates features and defines their boundaries, including genes, transcripts, exons, coding sequences, and noncoding RNA genes.
Functional annotation assigns likely roles to features, such as a protein’s function, a gene’s pathway, or a sequence’s regulatory activity.
Annotation combines computational predictions with evidence. A predicted gene model can be compared with RNA-sequencing data, known transcripts, protein sequences, and conserved genes in related organisms. Different evidence can support different parts of a model: RNA evidence may support an expressed transcript, while protein similarity may help identify a coding region or infer a likely function.
Annotations are predictions and interpretations, not direct guarantees of biological function. A feature may be missing because the assembly is incomplete, or may be incorrectly split, merged, or assigned when evidence is limited. Automated annotation enables large-scale analysis, while manual review can improve selected models. Annotation sets may change as evidence or methods improve. When reproducibility matters, record the annotation provider and release alongside the assembly version.
Comparing genomes and genes
compares sequences and their features across organisms or across different individuals. The aim is to identify similarities and differences that can help explain evolution, gene function, structure, or biological traits.
Sequence alignment compares DNA or protein sequences and reveals matching and differing positions. A local alignment focuses on similar regions; a global alignment compares sequences across their full lengths. Insertions and deletions are represented as gaps.
means that two sequences share ancestry. are homologous genes that diverged through speciation, while are homologous genes that arose through gene duplication. Similarity alone does not establish whether a pair is orthologous or paralogous.
Regions that remain similar across species may be constrained by functional or structural requirements. Conservation is evidence to investigate, not proof of a particular function; rapidly evolving or species-specific functions may show little sequence conservation. —the conserved order and neighborhood of genes across genomes—can provide evidence of shared ancestry and help identify corresponding regions even when individual sequences have changed.
A cautious comparative workflow
Suppose a gene in Species A has a similar sequence to two genes in Species B. If Species B experienced a duplication, its two genes may be of each other, while both are related to the Species A gene. Comparing gene trees, neighboring genes, and broader alignments can help distinguish these relationships. Choosing the closest sequence match alone may be misleading.
A comparative analysis can proceed in this order:
Select the organisms and assemblies relevant to the question.
Check each ’s assembly quality, version, and annotation source.
Compare homologous sequences or use -wide alignments to locate corresponding regions.
Examine gene order, gene trees, and other evidence to assess orthology, duplication, and rearrangement.
Interpret conserved or divergent regions cautiously, considering assembly gaps, annotation differences, evolutionary distance, and the limits of the method.
Reliable conclusions depend on treating computational results as evidence-based interpretations. The exact assembly and annotation versions should be recorded because assemblies and annotations can change, and coordinates are tied to particular assembly versions.