In W. P. Hanage’s article, he discusses the importance of five key questions when interpreting scientific literature:
- Can experiments detect differences that matter?
W.P. Hanage first talked about profiling a microbiome which is able to categorize at the level of phyla, species or genes. The disadvantage is from criterion which researchers used the different ratio of bacteria to distinguish microbes. For example, if an experiment was used to characterize animal communities, based on the ratio between an aviary of 100 birds and 25 snails and the ratio between an aquarium with 8 fish and 2 squids, both an aviary of 100 birds and 25 snails were identical with an aquarium with 8 fish and 2 squids. However, it was not right because of the differences in stains in the genes which each species contain. if the networks are not characterized well, the outcome to any specific object is hard. In addition, the ability to identify functional differences in related genes play the important role to understand the genes or networks. An experiment can detect differences or not play a significant role for researchers to understand fully.
- Does the study show causation or correlation?
This section discusses how an experiment was conducted in two directions, inverse and reverse directions in order to know if the study showed causation or just correlation. A study about gut microbiomes and diet in a 2012 article which was proposed a causal relationship after conducting a study between the gut microbiomes of old people living in care homes and old people living in the community. Though the data and proposal were fit together, the reverse causality, the potential for poor health to alter the gut microbiome, was not investigated. The less active immune system and differences in the digestion of frailer people could lead changes in the microbiome. The conclusion about the causal relationship was incorrect.
- What is the mechanism?
This question allows researchers to understand the subject on a deeper level. All scientists are taught that correlation is not causation, but correlation almost always implies some sort of causal relationship. Researchers just dont know what it is, and must determine it with careful experiment. Experiments can be designed to precisely define actions of components of microbiomes, for example by reconstituting communities but leaving out specific taxa. The experiment can be done to define actions of elements in microbiome related to the biochemical activity. This contributes to making a study become more convincing.
- How much do experiments reflect reality?
Focuses on examination the reflection of experiments to the reality. For example, the study about gut flora and weight gain. The researchers did the experiment on germ-free mice which did not represent the natural state of animals and do not have healthy owing. Meaning, the study did not include the responses in animals with flouring microbiomes related to different adaption between mice and their microbiomes and human. This question provides researchers about their subject choices and evaluation if the result from their study is suitable.
- Could anything else explain the results?
When researches want to experiment on a specific subject they should think about a variety of variables that could also contribute to results. That there are other potential factors which could affect the results of their study and the way to analyze the data, make generate the hypotheses and evaluate the conclusion. For example, bacteria affects human but whether or not there are possible factors contributing to these effects.
When it comes to which is most important when discussing controversies I would have to say defining the mechanism because this allows the researchers to understand what they experimenting on a deeper level, become more interested and knowledgeable on the subject they are studying. This allows for the researcher to go farther and when they get different results multiple times, they will be willing to do the experiment over again to get results that are similar to each other.
