I am new to QIIME2. I have been using it to for microbiome analysis of raw goat milk and environmental samples from the milking environment from four farms. I would like to compare the relative abundance of the taxonomy groups in each farms milk. I also want to compare the relative abundance of taxonomy groups in the milk from each farm to different areas of the milking environment to determine the area that is most influential on the microbiota present in the milk.
I have read through past forum questions...though I am stuck and not sure which code to run in order to accomplish my goal.
Thank you for your help in advance!
@Alyssa_Thibodeau, this topic is surprisingly thorny, as sequencing data by definition is compositional and not count data. This means that you have to be very careful with your experimental design and the methods you use to analyze your results, as many common statistical tools do not work for this kind of data even though it seems like they should.
Here are a few resources that might help provide some good context and guide your choices with what tools to use going forward: Microbiome Datasets Are Compositional: And This Is Not Optional, an approachable discussion of the topic by @mortonjt , and a tutorial from a workshop that shows how to use some of the tools available for these kinds of questions in QIIME 2.
I hope these resources are helpful! I wish there was a more straightforward answer to your question and there has been active discussion amongst the developers about tools that we could implement to help with this sort of analysis. Unfortunately the timeline so far consists of "hopefully pretty soon" .
If after looking all of that over you have more questions, come back here and post those. There are a lot of people here who have much more expertise on these topics than I do. I also would encourage you to post your planned methods here for feedback before taking any steps in your experiment, that way we can help make sure that you will have the data you need to answer the questions you would like to.