2  Fundamentals of Viromics

Viromics uses sequencing and bioinformatics to study the virome — the viral community associated with a sample or environment. If a microbial sample is a city, then:

• virology studies one resident in detail (how a single virus infects and replicates);
• metagenomics surveys every resident’s genetic material at once; and
• viromics focuses on the viral population — who the viruses are, what they can do, and which hosts they interact with.

That shift in focus changes almost everything downstream, from how you prepare a library to how you decide a contig is really viral.

Learning objectives — by the end of this chapter you will be able to:

• distinguish viromics from classical virology and general metagenomics;
• compare the major viral genome types and Baltimore groups;
• explain why viral genomes require different bioinformatics strategies from bacterial genomes;
• read the sequence signals that separate lytic infection from temperate lifestyles; and
• use Linux commands to filter and summarize a tab-separated dataset.

ImportantA virome is an operational measurement

No sequencing protocol captures every virus equally well. The observed virome depends on sampling, particle enrichment, nucleic-acid extraction, DNA or RNA selection, library preparation, sequencing depth, and the analysis workflow. A reported virome is therefore the viral community detected under a particular protocol, not a complete inventory of every virus present.

2.1 Viromics, virology, and metagenomics

Field	Main question	Typical data	City analogy
Virology	How does a virus infect, replicate, and affect its host?	isolates, cultures, PCR assays, microscopy	one resident, studied closely
Metagenomics	What genetic material is present in a community?	total-community DNA or RNA	every resident at once
Viromics	Which viruses and viral functions are represented in a community?	virus-enriched data or viral sequences from metagenomes	the viral population

Unlike bacteria and archaea, viruses do not share a universal marker gene comparable to 16S rRNA. Viral sequence identification therefore combines several imperfect signals: viral hallmark genes, similarity to reference sequences, gene density and orientation, sequence composition, genome architecture, read coverage, and host-association evidence (Minot et al. 2011). No single signal is sufficient in every case, especially for short contigs and highly novel viruses.

2.2 Viral genome types

Viral genomes may consist of DNA or RNA and may be single- or double-stranded. They can also be linear or circular, and segmented or non-segmented. These properties affect library preparation, assembly, annotation, and the databases used for classification.

Genome type	Meaning	Example
dsDNA	double-stranded DNA	many bacteriophages
ssDNA	single-stranded DNA	microviruses
dsRNA	double-stranded RNA	reoviruses
+ssRNA	positive-sense RNA	coronaviruses
-ssRNA	negative-sense RNA	influenza viruses
+ssRNA-RT	positive-sense RNA with reverse transcription	retroviruses
dsDNA-RT	double-stranded DNA with reverse transcription	hepadnaviruses

2.3 Virus genomes versus bacterial genomes

Many viral genomes are compact and encode functions that help the virus enter a host, replicate, assemble particles, evade host defenses, and spread. Bacterial genomes are generally larger and encode the systems needed to maintain a cellular organism, including translation, energy generation, membrane transport, and stress responses. These are useful trends rather than universal rules: giant viruses can have large genomes, and some viruses carry auxiliary metabolic genes that redirect host metabolism during infection.

Feature	Viral genome	Bacterial genome	Example
Typical genome size	a few kb to hundreds of kb, with exceptions above 1 Mb	hundreds of kb to several Mb	SARS-CoV-2 is about 30 kb; E. coli K-12 is about 4.6 Mb
Nucleic acid type	DNA or RNA	DNA	influenza has RNA, T4 phage has dsDNA
Strandedness	single or double stranded	typically double stranded	ssDNA phages exist, bacteria are generally dsDNA
Shape	linear, circular, segmented, non-segmented	usually circular chromosome, sometimes multiple replicons	hepatitis B partially dsDNA, Vibrio has two chromosomes
Gene density	very high	moderate	viral genomes often have overlapping genes
Ribosomes	absent	present in the cell	viruses use host ribosomes
Metabolic capacity	none independent; some encode metabolic genes	extensive cellular metabolism	viruses redirect host resources
Reproduction	depends on a host cell	growth followed by cell division	phages replicate only in suitable hosts

2.3.1 Genome structure examples

• T4 bacteriophage: a tailed bacteriophage with a linear dsDNA genome.
• PhiX174: a bacteriophage with a small, circular ssDNA genome; also widely used as an Illumina sequencing control.
• SARS-CoV-2: positive-sense ssRNA, about 30 kb, one of the largest RNA virus genomes.
• Influenza A virus: negative-sense ssRNA with a segmented genome.
• Escherichia coli: a bacterium with a circular chromosome and thousands of genes that support cellular life.

Figure 2.1: Simplified virus structure showing the viral genome, capsid, and optional envelope.

Figure 2.2: A typical prokaryotic cell. Comparing an independent bacterial cell with a host-dependent virus highlights why viruses lack ribosomes and independent metabolism.

These differences explain why viral genome interpretation often emphasizes hallmark genes, genome architecture, terminal repeats, coverage patterns, and host association. Bacterial genome interpretation can additionally draw on broadly conserved genes, metabolic pathways, and taxonomic markers. The Baltimore classification complements sequence-based taxonomy by organizing viruses according to how they produce mRNA (Baltimore 1971).

2.4 Baltimore classification

The Baltimore system asks a practical question: how does the virus produce positive-sense mRNA that host ribosomes can translate? The answer highlights which polymerases or reverse-transcription steps are required during infection.

TipA cooking analogy for Baltimore groups

Think of mRNA as a recipe the host ribosome can cook from directly:

• +ssRNA (IV) arrives as a ready-to-use recipe — translate immediately.
• -ssRNA (V) arrives written backwards; a viral polymerase must first rewrite it forwards.
• dsRNA (III) comes as a sealed double recipe; a viral polymerase transcribes a usable copy.
• dsDNA (I) brings a whole cookbook that host machinery transcribes.
• ssDNA (II) is half a cookbook that is first completed into dsDNA.
• Retroviruses (VI) carry an RNA recipe they reverse-transcribe into a DNA cookbook first.
• dsDNA-RT (VII) keep a DNA cookbook but copy it through an RNA intermediate.

Group	Genome	Route to mRNA
I	dsDNA	dsDNA is transcribed into mRNA
II	ssDNA	ssDNA is converted to dsDNA, then transcribed
III	dsRNA	viral RNA-dependent RNA polymerase produces mRNA
IV	+ssRNA	the genome can function directly as mRNA
V	-ssRNA	viral RNA-dependent RNA polymerase produces +mRNA
VI	+ssRNA-RT	RNA is reverse-transcribed into DNA, then transcribed
VII	dsDNA-RT	repaired dsDNA is transcribed; an RNA intermediate is reverse-transcribed during replication

NoteClassification systems answer different questions

Baltimore groups describe the route to mRNA; they are not taxonomic ranks. Formal virus taxonomy instead organizes viruses into realms, kingdoms, phyla, classes, orders, families, genera, and species. Two viruses can share a Baltimore group without being close evolutionary relatives.

2.5 Lytic and lysogenic lifestyles

During a lytic infection, a phage replicates, assembles new particles, and releases progeny — often by lysing the host cell. A temperate phage can instead enter lysogeny, in which its genome persists as a prophage integrated into the host chromosome or maintained as an extrachromosomal element. Stress or other signals may induce the prophage to resume productive infection (Roux et al. 2015).

Roux, Simon, Francois Enault, Bonnie L. Hurwitz, and Matthew B. Sullivan. 2015. “VirSorter: Mining Viral Signal from Microbial Genomic Data.” PeerJ 3: e985. https://doi.org/10.7717/peerj.985.

flowchart TD
  A[Phage contacts host cell] --> B{Lifestyle}
  B --> C[Lytic cycle]
  C --> D[Genome replication]
  D --> E[Capsid assembly]
  E --> F[Cell lysis and release]
  B --> G[Lysogenic state]
  G --> H[Prophage persists in or alongside host genome]
  H --> I[Host replication]
  I --> J[Induction can return to lytic cycle]

flowchart TD
  A[Phage contacts host cell] --> B{Lifestyle}
  B --> C[Lytic cycle]
  C --> D[Genome replication]
  D --> E[Capsid assembly]
  E --> F[Cell lysis and release]
  B --> G[Lysogenic state]
  G --> H[Prophage persists in or alongside host genome]
  H --> I[Host replication]
  I --> J[Induction can return to lytic cycle]

Figure 2.3: Lytic and lysogenic bacteriophage cycles.

Why does this matter for bioinformatics? Because each lifestyle leaves different fingerprints in your data:

Lifestyle	Expected sequence signals
Lytic / free particles	free viral genomes assembling as separate contigs, circular or complete genomes, high read coverage, hallmark genes such as capsid, terminase, portal, and polymerase
Lysogenic / prophage	viral regions flanked by host DNA, integrase or recombinase genes, attachment (att) sites, mixed host–virus signals on one contig

Recognizing these signals explains why prophages can appear inside bacterial contigs and why tools such as CheckV trim host flanks from proviruses.

2.6 Viromes across environments

Environment	Typical viral signals	Bioinformatics challenge
Human gut	bacteriophages, eukaryotic viruses	high inter-individual variation
Soil	phages, plant viruses, fungal viruses	inhibitors and very high diversity
Marine	marine phages	huge diversity and novelty
Wastewater	mixed human, animal, bacterial viruses	mixed sources and contamination
Plant	plant RNA viruses, phages, mycoviruses	host RNA or DNA background
Animal	wildlife and livestock viruses	zoonotic surveillance context

Phages are not passive passengers. They drive microbial mortality, nutrient cycling, and horizontal gene transfer, and in the human body they form distinct sub-viromes of the gut, oral cavity, skin, and respiratory tract.

2.7 Why viral dark matter matters

Many viral sequences have no close match in current reference databases — a knowledge gap often called viral dark matter. Reference-only searches therefore miss or weakly classify many viral contigs. Modern workflows combine complementary approaches: tools such as VirSorter2 (Guo et al. 2021) and geNomad (Camargo et al. 2024) identify candidate viral sequences, while CheckV estimates completeness and contamination of assembled viral genomes (Nayfach et al. 2021). Predictions remain hypotheses whose confidence depends on contig length, database coverage, and agreement among multiple signals.

Camargo, Antonio Pedro, Simon Roux, Frederik Schulz, Michal Babinski, Yan Xu, Bin Hu, Patrick S. G. Chain, Stephen Nayfach, and Nikos C. Kyrpides. 2024. “Identification of Mobile Genetic Elements with geNomad.” Nature Biotechnology 42: 1303–12. https://doi.org/10.1038/s41587-023-01953-y.

When a reference match is missing, these clues still help build confidence that a contig is viral:

• viral hallmark genes (capsid, large terminase subunit, portal, and polymerase);
• high gene density with few cellular housekeeping genes;
• circularity or direct terminal repeats;
• CRISPR spacer matches to a known host;
• consistent, even read coverage; and
• absence of ribosomal RNA and other cellular markers.

2.7.1 Why viromics is harder than bacterial metagenomics

Challenge	Why it complicates analysis
No universal marker gene	there is no viral 16S; identification needs multiple signals
Enormous diversity	viral sequence space is vast and largely uncharted
Small genomes	short contigs carry less evidence per sequence
Reference bias	databases are incomplete, so novel viruses are missed
Contamination	host and bacterial DNA can dominate a library
Host DNA background	plant, animal, or human reads can swamp viral reads
RNA vs DNA workflows	RNA viromes need reverse transcription and are less stable
Prophages	integrated viruses blur the host–virus boundary

2.8 Linux activity: explore genome types

This activity creates a small tab-separated values (TSV) file, selects RNA genome types, and summarizes the genome categories. It requires only a POSIX-like shell and standard command-line tools. First, set up the standard project folders you will reuse throughout the book.

⬇ setup_project.sh

# One-time: create the standard project structure used across the book
bash setup_project.sh          # or: bash ~/Downloads/setup_project.sh

2.8.1 1. Create the dataset

mkdir -p ~/viromics_course/phase1_fundamentals
cd ~/viromics_course/phase1_fundamentals

cat > viral_genome_types.tsv << 'EOF'
Virus   Group   Genome  Example
T4_phage    I   dsDNA   Bacteriophage
Microvirus  II  ssDNA   Bacteriophage
Rotavirus   III dsRNA   Animal_virus
Coronavirus IV  +ssRNA  Animal_virus
Influenza   V   -ssRNA  Animal_virus
HIV VI  +ssRNA-RT   Retrovirus
Hepatitis_B VII dsDNA-RT    Hepadnavirus
EOF

column -t -s $'\t' viral_genome_types.tsv   # optional pretty printing

2.8.2 2. Select RNA genome types

awk -F'\t' 'NR==1 || $3 ~ /RNA/' viral_genome_types.tsv

-F'\t' tells awk that fields are separated by tabs. NR==1 retains the header, and $3 ~ /RNA/ selects records whose third field contains RNA.

2.8.3 3. Count genome categories

cut -f3 viral_genome_types.tsv | tail -n +2 | sort | uniq -c

The pipeline extracts the third column, removes the header, sorts identical values together, and counts them. Because this teaching dataset contains one example per Baltimore group, each category should have a count of one.

TipCheck your understanding

Modify the awk command to select bacteriophages using the fourth column:

awk -F'\t' 'NR==1 || $4 == "Bacteriophage"' viral_genome_types.tsv

NoteExercise: build an environments table

Create virome_environments.tsv with columns Environment, Main_viruses, Main_hosts, and Bioinformatics_challenge for five environments (human gut, soil, marine, wastewater, plant), then print only the rows where the challenge mentions “diversity”.

One solution:

cat > virome_environments.tsv << 'EOF'
Environment Main_viruses    Main_hosts  Bioinformatics_challenge
Human_gut   Bacteriophages  Gut_bacteria    high_individual_variation
Soil    Phages_and_plant_viruses    Soil_microbes   inhibitors_and_high_diversity
Marine  Marine_phages   Marine_bacteria huge_diversity_and_novelty
Wastewater  Mixed_viruses   Mixed_hosts mixed_sources_contamination
Plant   Plant_RNA_viruses   Plant_cells host_background_signal
EOF

awk -F'\t' 'NR==1 || $4 ~ /diversity/' virome_environments.tsv

Estimated resources: one CPU thread, less than 100 MB RAM, less than 1 MB storage, and under one minute.

WarningCommon beginner mistakes

• Trusting one tool. A single positive call is a hypothesis, not a conclusion — require agreement across evidence.
• Ignoring CheckV quality. Reporting fragments as if they were complete genomes overstates your results.
• Forcing species-level names. For novel viruses, family-level or higher assignments are usually more honest.

2.9 Key takeaways

• Viromics targets the viral population of a sample, sitting between single-virus virology and whole-community metagenomics, and every reported virome is an operational, protocol-dependent measurement rather than a complete inventory.
• Viral genomes span DNA and RNA, single- and double-stranded, linear, circular, and segmented forms; the Baltimore system organizes them by their route to translatable mRNA and complements formal taxonomy (Baltimore 1971).
• Viruses lack a universal marker gene like 16S rRNA, so identification must combine several imperfect signals — hallmark genes, gene density, composition, coverage, and host association — rather than relying on any single line of evidence (Minot et al. 2011).
• Lytic and lysogenic lifestyles leave distinct fingerprints in sequence data, which is why prophages appear embedded in host contigs and why proviral host flanks must be trimmed before analysis.
• Much of the virosphere is uncharacterized “viral dark matter,” so predictions from tools such as VirSorter2 remain hypotheses whose confidence rises with contig length, database coverage, and agreement across methods (Guo et al. 2021; Nayfach et al. 2021).

Guo, Jiarong, Benjamin Bolduc, Ahmed A. Zayed, Arvind Varsani, Gabriela Dominguez-Huerta, Tom O. Delmont, Akbar A. Pratama, et al. 2021. “VirSorter2: A Multi-Classifier, Expert-Guided Approach to Detect Diverse DNA and RNA Viruses.” Microbiome 9: 37. https://doi.org/10.1186/s40168-020-00990-y.

2.10 Further reading

• Baltimore (1971) introduces the classification of viruses by their strategy for producing mRNA, the conceptual backbone of the Baltimore groups discussed here.
• Minot et al. (2011) is a foundational human gut virome study illustrating why viral identification depends on multiple, imperfect signals.
• Nayfach et al. (2021) describes CheckV, the standard for estimating completeness and contamination of assembled viral genomes and for trimming prophage host flanks.
• The ICTV maintains the authoritative framework for virus taxonomy, complementing Baltimore groups (International Committee on Taxonomy of Viruses 2026); see the official site at https://ictv.global.

Baltimore, David. 1971. “Expression of Animal Virus Genomes.” Bacteriological Reviews 35 (3): 235–41. https://doi.org/10.1128/br.35.3.235-241.1971.

Minot, Samuel, Rohini Sinha, Jun Chen, Hongzhe Li, Sue A. Keilbaugh, Gary D. Wu, James D. Lewis, and Frederic D. Bushman. 2011. “The Human Gut Virome: Inter-Individual Variation and Dynamic Response to Diet.” Genome Research 21 (10): 1616–25. https://doi.org/10.1101/gr.122705.111.

Nayfach, Stephen, Antonio Pedro Camargo, Frederik Schulz, Emiley Eloe-Fadrosh, Simon Roux, and Nikos C. Kyrpides. 2021. “CheckV Assesses the Quality and Completeness of Metagenome-Assembled Viral Genomes.” Nature Biotechnology 39: 578–85. https://doi.org/10.1038/s41587-020-00774-7.

International Committee on Taxonomy of Viruses. 2026. “ICTV Taxonomy.” https://ictv.global/taxonomy.

2.11 Chapter figure

Figure 2.4: Viromics places viruses inside a microbial ecosystem and feeds the resulting sequences into a Linux-based bioinformatics pipeline.

Tip🎨 Generate this figure

Save as: images/ch01-viromics-ecosystem.png · Aspect ratio: 16:9 · Style: clean flat vector infographic, Codanics palette (teal #008b8b, navy #05043b, white background), no photorealism.

Prompt: Create a clean educational infographic showing viromics as a viral population inside a microbial ecosystem. On the left, show a stylized ecosystem containing bacteria, archaea, fungi, plant cells, and animal cells, with bacteriophages, RNA viruses, and DNA viruses highlighted among them. In the center, show viral particles being enriched and their genomes extracted. On the right, show viral contigs flowing into a Linux terminal representing a bioinformatics pipeline. Use a modern scientific style with teal and navy Codanics branding and clear labels.

2.12 Quiz: Fundamentals

Q1. What is a virome?

A. all bacteria in a sample B. all viruses in a sample or ecosystem C. only disease-causing viruses D. only RNA viruses

Note

Answer: B. A virome is the viral community associated with a sample or environment, although an experiment generally detects only a protocol-dependent subset.

Q2. Why is viromics difficult compared with 16S profiling?

A. viruses have no universal marker gene B. viruses are always larger than bacteria C. viral genomes are all identical D. viruses cannot be assembled

Note

Answer: A. There is no universal viral barcode equivalent to 16S rRNA.

Q3. What is a prophage?

A. a temperate phage genome persisting in or alongside a host genome B. a sequencing adapter C. a protein domain D. a host ribosome

Note

Answer: A. A prophage persists in a lysogen, commonly integrated into the host chromosome but sometimes maintained extrachromosomally.

Q4. Which Baltimore group includes +ssRNA viruses?

A. I B. II C. IV D. VII

Note

Answer: C. Group IV includes positive-sense single-stranded RNA viruses.

Q5. What is viral dark matter?

A. viral sequences with no close reference match B. contaminated adapters C. low-quality bases only D. a type of culture medium

Note

Answer: A. Many viral sequences remain poorly represented in reference databases.

2.13 Interactive quiz: Fundamentals

How to use this quiz: Select one option, click Check answer, and read the explanation. Use the reset button if you want to try again.

1. What is the best description of a virome?

A. All proteins in a sample B. All viruses in a sample or ecosystem C. Only viruses that cause disease D. Only bacteriophages

A virome refers to the viral community associated with a sample or environment. It can include bacteriophages, animal viruses, plant viruses, and other viral groups.

2. Why is viromics harder than bacterial 16S profiling?

A. Viruses cannot be sequenced B. Viruses are always rare C. Viruses lack a universal marker gene like 16S rRNA D. All viral genomes are circular

There is no universal viral barcode equivalent to bacterial 16S rRNA. Viromics therefore depends on multiple signals such as hallmark genes, genome structure, sequence composition, and reference comparison.

3. Which statement best contrasts a virus genome with a bacterial genome?

A. Viruses contain ribosomes and bacteria do not B. Bacteria usually have larger genomes and can live independently C. Viruses always have DNA and bacteria always have RNA D. Bacterial genomes cannot be circular

Bacterial genomes are usually much larger and encode the core functions needed for independent life. Viral genomes are compact and rely on host cells for replication and translation.

4. What is a prophage?

A. A host ribosome B. A low-quality read C. A temperate phage genome persisting in or alongside a host genome D. A type of sequencing adapter

A prophage is the persistent form of a temperate phage in a lysogen. It is commonly integrated into the host chromosome but can also be maintained extrachromosomally until induction initiates productive infection.

5. Which Baltimore group includes positive-sense single-stranded RNA viruses?

A. Group I B. Group III C. Group IV D. Group VII

Group IV viruses have positive-sense single-stranded RNA genomes. Their genomes can function directly as mRNA.