Discussion

New and emerging (including re-emerging) diseases are mostly caused by microbes and there again mainly by viruses, which through their need to replicate in cells and their size traditionally pose the biggest challenge to identify. Before microbe hunting (Lipkin 2009) can begin, however, the clinical recognition is core to the process and once a problem has been recognised the appropriate set of samples is crucial to the success.
Traditional microbiology is based on the isolation of pathogens, which for viruses first required animals then organ cultures and at last tissue cultures. While virus isolation still resembles the gold standard of virus detection, it is full of pitfalls, laborious and time consuming. More so, it often fails, as the different viruses have very specific culture conditions, difficult or impossible to mimic in vitro. Ultimately, however, it is the best way for further analysis, starting with Koch's postulates, to demonstrate the causative nature of the pathogen.

In recent times, molecular methods have become more standard tools in diagnostic labs and for known pathogens, techniques like PCR or real-time PCR (qPCR) have become standard. Together with quality control schemes to limit false negative and positive results they are state of the art, safe and cost-effective (if not cheap) tools. As they build on existing knowledge about the viral genomes, they have limits though, when it comes to the detection of new and emerging diseases. Also, to start an investigation by PCR (or related) technologies, one needs to have a suspicion to pursue, derived from a tentative, clinical (or post mortem) diagnosis. If this initial diagnosis proves to be wrong or the disease symptoms are fairly unspecific, even multiple tests might not provide a conclusive result. Furthermore, double infections are often neglected, from the suspicion to the diagnostic result.

Most recently more complex technologies, able to identify multiple (thousands) of pathogens in parallel, including such that were never detected before have been established. Pathogen microarrays provide us with the ability to detect literally all known viruses in one reaction. Here, thousands of probes are spotted on a slide and reactors are detected through hybridisation. We have successfully employed this technology several times now to detect new and emerging diseases, including a double infection and the equine encephalosis virus in Israel (Mildenberg et al. 2009; Barlow et al. 2012).
Array technologies, however, much like PCR technologies, are only as good as the existing knowledge and come to limits when a pathogen is really new (or not well characterised). Here deep sequencing, also known as Next Generation Sequencing (NGS) is the latest, exciting technology in the toolbox. This technology (or better technologies) make use of the vast progress in sequencing technology, originally initiated by the genome projects, which also benefited the horse (Wade et al. 2009). In essence, primary tissue samples are completely sequenced, including their genome and/or transcriptome. Subsequently, bioinformatics methods are used to filter out known sequences from the host (e.g. horses) and obtain such that could resemble new pathogens/viruses. These suspect sequences need to be further analysed. Processes like this have been successfully used to detect new animal viruses including the Schmallenberg virus or to rule out infectious processes such as in the case of bovine neonatal pancytopenia (Willoughby et al. 2010; Kapoor et al. 2011; Hoffmann et al. 2012).