TABLE 1.

Applications, benefits and limitations of technologies used for the study of rumen viruses.

Method		Application and benefits	Limitations	Example publications
Microscopy	TEM	Visualization of viral particle morphology	Specialized equipment	Hoogenraad et al., 1967; Ritchie et al., 1970; Klieve and Bauchop, 1988
		Estimation of viral numbers	Sample preparation can bias enumeration
			Time consuming and expensive
			Cannot determine viral particle viability and biological attributes
Molecular Biology	PFGE	Provides snapshot of viral community and estimation of viral numbers	Cannot provide taxonomic and functional gene information	Klieve and Swain, 1993; Swain et al., 1996a
Isolation		Confirms viral particle viability	Requires availability of susceptible microbial host	Klieve, 2005; Gilbert and Klieve, 2015
		Allows viral cultivation and storage in reference collections	Bias from sample preparation methods (e.g. exclusion of large viral particles)
		Enables determination of biological parameters (host range, growth, replication and survival)
		Allows extraction and sequencing of virus-specific nucleic acids
		Allows viral protein purification
Sequencing	Viral fraction	Provides snapshot of viral community structure	Requires isolation of intact viral particles from environmental samples	Breitbart et al., 2002; Clokie et al., 2011; Berg Miller et al., 2012; Anderson et al., 2017
		Provides taxonomic and functional gene information	Bias from sample preparation methods (e.g. exclusion of large viral particles)
		Overcomes technical limitations from low concentrations of viral DNA in environmental samples	Bias from any DNA amplification steps
			High percentage of uncharacterized viral genes limits annotation of gene function and viral taxonomy
			Sequence assembly bias and challenges
			Cannot determine viral particle viability and biological attributes
			Difficult to identify viral lifecycles (e.g. lysogeny) and virus:host interactions
	Metagenomics	Provides snapshot of viral community structure	Virus sequence numbers relatively low	Dinsdale et al., 2008; Willner et al., 2009
		Captures sequences from intact viral particles and integrated prophages	High percentage of uncharacterized viral genes limits annotation of gene function and viral taxonomy
		Provides viral taxonomic and functional gene information	Difficult to identify viral lifecycles (e.g. lysogeny) and virus:host interactions
			Cannot determine viral particle viability and biological attributes
			Bias toward detection of double stranded DNA phages
	Transcriptomics	Identifies actively replicating viral genes	Virus sequence numbers relatively low	Hitch et al., 2019
		Allows detection of viruses with RNA genomes	Bias toward identification of over-expressed viral genes (e.g. structural proteins)
			High percentage of uncharacterized viral genes limits annotation of gene function and viral taxonomy
			Difficult to identify viral lifecycles (e.g. lysogeny) and virus:host interactions
			Cannot determine viral particle viability and biological attributes
	Whole genome sequencing	Provides complete viral genome sequences	For lytic viruses requires pure, viable virus isolates	Leahy et al., 2010; Kelly et al., 2014; Gilbert et al., 2017
		Viral reference sequences increase the accuracy of sequence analysis	For lysogenic viruses requires viable prokaryote hosts containing intact, integrated prophage/s
		Provides structural and functional viral protein information	High percentage of uncharacterized viral genes limits annotation of gene function
		Indicates mechanisms of virus:host interactions and viral replication
		Enables assignment of taxonomy and phylogenetic comparison
Proteomics		Detects proteins produced by actively replicating viruses	Virus proteins found in relatively low concentrations	Solden et al., 2018
			Difficult to identify viral lifecycles (e.g. lysogeny) and virus:host interactions
			High percentage of uncharacterized viral proteins limits functional annotation
			Bias toward identification of over-produced viral proteins (e.g. structural proteins)