A guide to statistical analysis in microbial ecology: a community-focused, living review of multivariate data analyses

Abstract

The application of multivariate statistical analyses has become a consistent feature in microbial ecology. However, many microbial ecologists are still in the process of developing a deep understanding of these methods and appreciating their limitations. As a consequence, staying abreast of progress and debate in this arena poses an additional challenge to many microbial ecologists. To address these issues, we present the GUide to STatistical Analysis in Microbial Ecology (GUSTA ME): a dynamic, web-based resource providing accessible descriptions of numerous multivariate techniques relevant to microbial ecologists. A combination of interactive elements allows users to discover and navigate between methods relevant to their needs and examine how they have been used by others in the field. We have designed GUSTA ME to become a community-led and -curated service, which we hope will provide a common reference and forum to discuss and disseminate analytical techniques relevant to the microbial ecology community.

Geochemical characterization of groundwater discharging from springs north of the Grand Canyon, Arizona, 2009–2016

“…11) which falls between a fair and suspect result (Buttigieg and Ramette, 2014). Elements associated near uranium were molybdenum, strontium, selenium and lithium, and springs with higher uranium loading were farther from springs influenced by arsenic, cadmium, copper, lead, and manganese ( fig.…”

“…An analysis of variance (ANOVA) was run for the elements without censored values and the TukeyHSD function was used to determine which groups were statistically different. Non-metric multidimensional scaling (NMDS) was used to reduce the complex data structure (many samples and many elements) to represent the pairwise dissimilarity between objects in a low-dimensional space (Buttigieg and Ramette, 2014). The following elements, As, Cd, Cu, Li, Mn, Mo, Pb, Se, Sr, U, V, and Zn were used for multivariate comparison based on the distinction between breccia pipe uranium mining sediment leachate and spring chemistry from Beisner and others (2017).…”

“…The NMDS was performed on the uscores using the metaMDS function from the vegan package in R (Oksanen and others, 2016) using euclidean distance, zerodist = add and autotransform = false (Helsel, 2012). NMDS stress values less than or equal to 0.1 are considered fair, while values equal to or below 0.05 indicate good fit and values near or greater than 0.2 are deemed suspect (Buttigieg and Ramette, 2014).…”

Exploring Long Term Spatial Vegetation Trends in Taiwan from AVHRR NDVI3g Dataset Using RDA and HCA Analyses

“…Redundancy is often synonymous with explained variance [69]. The RDA is a direct extension of multiple linear regression by allowing the regression of multiple response variables (RVs) on multiple explanatory variables (EVs) [70,71]. In addition, an RDA may also be seen as an extension of a principal component analysis because the canonical ordination vectors are linear combinations of the RVs.…”

“…In addition, an RDA may also be seen as an extension of a principal component analysis because the canonical ordination vectors are linear combinations of the RVs. In general, RDA is a method based on canonical multivariate analyses by assuming a linear response between variables [72][73][74] to extract and summarize the variation in a set of RVs that can be explained by a set of EVs [71]. The assumed linear response between variables can be expressed by the Eigen analysis equation as follows, and be decomposed using a standard eigenvalue-eigenvector algorithm.´S…”