Figure 2: Possible series of genomic steps in pathogen genomic epidemiological studies.

First, genomic data are collected. To be useful, those data must contain variation that has accumulated on an epidemiological relevant scale, e.g. due to mutation or due to recombination, processes that data summaries generated in step two often reflect. Step three involves a suite of descriptive analyses; for example, the computation of descriptive statistics of pathogen diversity and differentiation, population assignment and clustering analyses. A model that connects the genomic processes to epidemiological ones is formulated in the forth step, typically using mathematics to articulate hypotheses concretely. This process incites clarification and thus is valuable in and of itself. An arrow connects data to the fifth step, because data are used to infer the parameters of the model. This step also links to epidemiological data; however, these are not shown. Above and below the steps, examples are provided for viral and malaria genomic epidemiology, respectively. Those that do not currently exist are highlighted in italic and bold.