ing generated.” Another key challenge is the lack of communication and collaboration among experts in microbiology, nutrition, and other relevant fields. She encouraged more dialogue across disciplinary groups and more industry participation in the dialogue.
Challenges with Experimental Design
During the question-and-answer period following Nelson’s talk, most of the discussion revolved around experimental design, including tissue sampling, sample size, and the definition of “normal.” An audience member suggested that sampling from the small intestine would be more informative than sampling from fecal samples or from the esophagus. Nelson agreed that the field needs to move in a direction where scientists are sampling from other parts of the GI tract, but researchers have done as well as they have been able in the early stages. Sampling from the small intestine would be more invasive than fecal sampling. She suggested that more cross-disciplinary dialogue, in this case with medical researchers, could help move the field in that direction. Another question was raised about the value of a study based on a sample size of 300, let alone 2, individuals (the person asking the question was referring to Nelson’s mention of the N = 300 sample size of the HMP reference database and the N = 2 sample size of the Gill et al., 2006, study). Nelson agreed that the field needs to move in that direction, that is, toward sampling large host populations, but again she said, “We did what we could do at that point in time.” Another audience member asked how HMP investigators define “normal.” Lita Proctor agreed that the question is “nontrivial.” Defining normal was a struggle. They finally decided that for the purpose of the HMP, “normal,” or “healthy,” meant verification of no overt disease based on clinical examination.
The microbiome is integral to human physiology, health, and disease.