Viruses are identified by their shape (rods, spheres), size, nucleic acid type (RNA, DNA), size, single or double stranded, number of nucleic acid pieces, type of vector for spread and serological characteristics of the coat or capsid protein. These characteristics are used because they tend to be conserved characteristics for each virus type. However, they are characteristics that often require considerable cost, skill, time and equipment to observe for purposes of virus identification. Therefore, new viruses can take months to a year to identify.
It is very important to know the correct identity of the problem virus. If we know identify of the virus, much will be known about where the virus comes from, how it spreads and what can be done about it. Known viruses often can be quickly identified (several days) by serology or by a new technique called polymerase chain reaction (PCR) which targets and amplifies specific portions of the viral genetic material that are recognizable. Serology is very dependable and easy to run but there is considerable cost to acquisition or manufacture of the antiserum specific to each virus. PCR is very sensitive, relatively inexpensive once developed but is based on knowledge of the nucleotide sequence of the virus genetic material. The sequences of many viruses are now available from database collections such as GENBANK. The exception has been dahlia mosaic virus. Recently, we cloned and nucleotide sequenced multiple strains of the virus and have developed a very sensitive PCR test for the virus. Because there are many strains of dahlia mosaic virus we do not yet know if we can detect all strains of the virus.
Symptoms and host range which are very important in a practical sense are of little use for virus identification. Because of the mutable nature of viruses, there are usually many strains of each virus. Differences in symptoms and host range are usually linked to very minor genetic sequences.