Linking questions to practices in the study of microbial pathogens: sampling bias and typing methods.

The importance of understanding the population genetics and evolution of microbial pathogens is increasing as a result of the spread and re-emergence of many infectious diseases and their impact for public health. In the last few years, the development of high throughput multi-gene sequence methodologies has opened new opportunities for studying pathogen populations, providing reliable and robust means for both epidemiological and evolutionary investigations. For instance, for many pathogens, multilocus sequence typing has become the "gold standard" in molecular epidemiology, allowing strain identification and discovery. However, there is a huge gap between typing a clinical collection of isolates and making inferences about their evolutionary history and population genetics. Critical issues for studying microbial pathogens such as an adequate sampling design and the appropriate selection of the genetic technique are also required, and will rely on the scale of study and the characteristics of the biological system (e.g., multi- vs. single-host pathogens and vector vs. food or air-borne pathogens). My aim here is to discuss some of these issues in more detail and illustrate how these aspects are often overlooked and easily neglected in the field. Finally, given the rapid accumulation of complete genome sequences and the increasing effort on microbiology research, it is clear that now more than ever integrative approaches bringing together epidemiology and evolutionary biology are needed for understanding the diversity of microbial pathogens.