Shulou(Shulou.com)06/01 Report--

This article mainly introduces how to use R language to analyze the relevant knowledge of Hellinger transformation problems, the content is detailed and easy to understand, the operation is simple and fast, and has a certain reference value. I believe you will gain something after reading this article on how to analyze Hellinger transformation problems in R language. Let's take a look.

On the conversion of species composition data Hellinger

Generally speaking, linear models such as PCA are rarely used for species composition data, and PCoA or NMDS are recommended.

If we still want to use PCA, in fact, it does not mean that species abundance data have to be transformed. If the species abundance data are uniform and contain few zeros (which is the most important, avoiding the Euclidean distance sensitivity to double zeros), it is also possible not to convert.

For example, the example data "phylum_table.txt", because it is a gate-level (a large category) group statistics, its own value of 0 is less, in fact, it does not have to perform Hellinger conversion. However, the difference in abundance between high-abundance and low-abundance populations may be very large, and the data are not "uniform". In this case, the sensitivity of Euclidean distance to high values can be reduced by using "scale=TRUE" to eliminate the difference in importance between high-abundance and rare species when running rda ().

However, if the value of 0 is too high, Hellinger conversion is still recommended when performing PCA. Let's change the data to explain it.

The net disk file "otu_table.txt" is a species abundance table at the OTU level, which contains a lot of zero values, and then we compare the PCA differences before and after Hellinger transformation.

# compare the differences of species data before and after Hellinger transformation, otu

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.