Adding (or not) fatty acid compositions for bugphyzz signatures

[sdgamboa]

AFAIK, microbial fatty acids (FAs) compositions are interpreted as profiles (rather than individual classes/species of FAs). I think these FAs profiles are mostly used for taxonomic classification, basically like any other collection of characters or a DNA sequence:

"The fatty acid composition is probably the most important chemotaxonomic parameter used for the differentiation of prokaryote taxa at the species level because it correlates very well with the accepted DNA:DNA hybridization-based species concept of many groups of bacteria." Ref: https://doi.org/10.1016/B978-0-12-387730-7.00008-5

In fact, the references of this attribute seem to be all about taxonomy descriptions from the literature: https://github.com/bovee/fattyacids/blob/master/references.tsv

The information provided by these FA compositions is probably redundant since we already have the taxonomy of the microbes.

If individual FA classes are still of interest in creating signatures, two steps should be taken:

1. Convert the numeric values to TRUE/FALSE. This was suggested in Add rule to Fatty Acid Composition spreadsheets #27.
2. The names of the FAs are not very user-friendly. A higher class or group of FAs should replace FAs names. This was discussed but we didn't reach a solution.

I'm writing this issue now to document the removal of TODOs related to the FAs in the R code as requested during the feedback about the package submission to Bioconductor.