Microbial Diversity
We know there are tons of microbes in a sample… so how do we go about quantifying them for analysis? Diversity is the keyword we’re dealing with, specifically alpha and beta diversity. Alpha diversity is straightforward – how many organisms are within the sample you took and how abundant they are. Beta diversity deals with how similar or dissimilar two samples are.
Specifically with alpha diversity… there are two characteristics that are useful when describing a sample: richness and evenness. Richness is the number of organisms within your collected sample, and evenness is how similar an abundance of the microbes within your sample is. The diversity of a sample with respect to alpha diversity doesn’t consider microbes on the individual level, it’s more of a read on the overall characteristics of richness and evenness as a benchmark. This means you can look at the alpha levels of two communities that have similar richness and evenness, but at the taxa level, they contain drastically different microbes.
Photo by Nathan Anderson on Unsplash
Beta diversity works to describe the change between environments. In an ecological lens, the higher the diversity the less similar two communities will be. Quantitative data points are important in beta diversity, meaning how many organisms within two compared samples are shared. If there are more shared between both samples, the beta diversity is lower.
So what do we do with all these markers for diversity? As with most data, we can find ways to plot, compare and analyze. The difficulty arises with larger dimensions in our data sets. It can be straightforward to compare data with three different data points, using X, Y and Z planes. As you increase the dimensions, visualization becomes murkier or even impossible, and while mathematical formulas like Euclidian distance can still aid in data analysis, finding ways to continue visualizing the data remains important. This is where dimensionality reduction can be used. Finding ways to sift through the patterns within the data by limiting, ignoring, or removing dimensions within the set.
Join me in learning about the microbiome!