The most powerful means for detecting variation is sequencing, since every possible variant is identified and precisely mapped on the genome. There are several approaches that can be used. The purest and most expensive approach is to clone PCR products of the target(s) and sequence batches of individual clones. This also allows the relative frequencies of individual variants to be determined, even those variants occurring at very low frequencies. A second and cheaper approach is to sequence the PCR products directly and identify the polymorphisms at sites of ambiguity in the sequence (Forsgren et al., 2009; Fig. 7). Since such double peaks can also be the result of sequencing artefacts, each polymorphic site has to be confirmed by a matching pattern when sequencing the complementary strand. Only major polymorphisms can be identified and quantitation is moderate, similar to HRM. The new, high volume automated sequencing methods (Next Generation Sequencing, or NGS) have the capacity to directly analyse complex DNA and RNA mixtures through sequencing followed by automated similarity searches. These methods are rapidly becoming cheaper and more accurate, mostly through massive multiplexing of reactions and samples. They are increasingly being used as a one-step diagnostic method capturing millions of different targets, thus benefiting also from economy of scale in the data generated. They have recently also been used in honey bee pathology studies (Cox-Foster et al., 2007; Runckel et al., 2011) and are covered in detail in the BEEBOOK paper on molecular techniques (Evans et al., 2013).
Pros: Comprehensive; fast; flexible; accurate; sensitive; specific; low contamination risk; approximate quantification (NSG).
Cons : Expensive - precise - limited quantification (Sanger
sequencing); Very expensive -
less precise - approximate quantification (NGS).
Fig. 7. Mixed virus sequences as revealed in sequence electropherogram. The mixed sequence can be resolved into component sequences using specifically designed software. Adapted from Forsgren et al., 2009.