Integrating Multiple Datasets in R
Edward Curry · 2020
It really is a useful skill to be able to manipulate datasets so they can be loaded into R, then explore structures and patterns in the data, relating different entities to each other. Having said that, often the real power in analyzing a dataset is to look for associations with characteristics of a different dataset. For example, it is interesting to see the samples in a molecular profiling dataset segregate into clearly distinct clusters, but it adds a lot to our interpretation of these clusters if the samples in different clusters tend to have different phenotypic traits. As another common example, we may wish to find proteins or genes whose expression levels in tumours are associated with the patients’ clinical outcomes: this typically necessitates integration of data from multiple sources.