In meireles/evolspec: Evolution of leaf spectra

Installing `evolspec`

The easiest way to install evolspec directly from github using devtools

# If you don't already have `devtools`
install.packages("devtools")

library("devtools")

# Install `evolspec`
devtools::install_github("meireles/evolspec")

Using `evolspec`

Datasets

Spectra

We provide the spectral dataset of ASD measurements of Oaks from Cavender-Bares et al. 2016 as an example dataset.

library(evolspec)

## The oak_spec_dataet has "species" as the first column.
oak_spec_jcb2016[ 1:3, 1:5 ]

The oak_spec_jcb2016 has many individuals measured per species, hence many more rows than species names. If you want to aggregate data by species, the function aggregate_by_species will return aggregated means $mean, standard deviation $sd and standard errors $se for each wavength.

oak_spec_by_sp = evolspec::aggregate_by_species(oak_spec_jcb2016)
names(oak_spec_by_sp)

You may want to manipulate spectral data with base R too, for instance, to get wavelength labels or resample the spectra.

## To get a matrix with spectra only and get the wavelength labels, do
oak_spec_only = oak_spec_jcb2016[ , -1]
wlavelengths  = colnames(oak_spec_only)

## You can also resample the original oak data at a certain resolution,
## for instance, 10 nm
wl_subset_10nm     = as.character( seq(400, 2400, by = 10) )
oak_spec_10nm      = oak_spec_jcb2016[ , c("species", wl_subset_10nm)]
oak_spec_only_10nm = oak_spec_only[ , wl_subset_10nm ]

## Finally you can also aggregate by species the resamples spectra at 10nm resolution
oak_spec_by_sp_10nm_mu =  evolspec::aggregate_by_species(oak_spec_10nm)$mean

Phylogenetic tree

We also provide the Oak phylogenetic tree oak_tree that matches the example spectral data.

library("evolspec")

plot(oak_tree, cex = 0.5)

Phylogenetic signal

spec_phylosig_lambda = evolspec::phylo_signal(oak_tree,
                                              oak_spec_by_sp$mean,
                                              method = "lambda")

spec_phylosig_k      = evolspec::phylo_signal(oak_tree,
                                              oak_spec_by_sp$mean,
                                              method = "K")

par(mfrow = c(1, 2))

plot(x = wlavelengths,
     y = spec_phylosig_lambda$lambda,
     ylab = "Pagel's lambda", cex = 0.5,
     col = ifelse(spec_phylosig_lambda$pval <= 0.05, "red", "black"))

plot(x = wlavelengths,
     y = spec_phylosig_k$K,
     ylab = "Bloombergs's K", cex = 0.5,
     col = ifelse(spec_phylosig_k$pval <= 0.05, "red", "black"))