extrapolate: Extrapolate nematode community composition from a subsample...

In amynang/marcel: Calculates nematode community indices

Extrapolate nematode community composition from a subsample of identified individuals

Usage

extrapolate(df, counts)
extrapolate.robust(df, counts, n = 100)

Arguments

df	A sites by taxa matrix or dataframe, containing the numbers (or frequencies) of identified nematodes. Either way, it will be standardised by dividing with row totals.
counts	A vector with the number of counted nematodes in each site.
n	number of iterations

Details

• extrapolate() simply multiplies the frequency of each taxon in the subsample with the total number of nematodes in each site. This implicitly assumes that the composition of the identified subsample is perfectly representative of the total.

• extrapolate.robust() is a bootstraping approach (I think). It treats the frequencies of taxa in the identified subsample as probabilities. It then draws randomly, with replacement, from the possible taxa, a number equal to the counted nematodes, where the probability of each taxon being drawn is its frequency in the subsample. The process is iterated n times yielding n versions of each site.

Value

extrapolate()	returns a matrix of the same dimensions as the df.
extrapolate.robust()	returns a list of n such matrices.

Note

What you should NOT do with the output of .robust (or any downstream product such as the Enrichment Index), is treat the iterations as levels of a random effect in a regression; this would artificially inflate the number of observations, likely making any difference, however small and bogus, highly significant!

What you should do is open to discussion; one simple idea is to fit n individual regressions and examine the distribution of the lower and upper bounds of their confidence intervals.

Author(s)

Angelos Amyntas

Examples

# community matrix with frequencies
example1 = matrix(c(.1,0,.2,.7,0,0,.4,.6,.8,.1,.1,0),
                  byrow = T, nrow = 3, ncol = 4,
                  dimnames = list(c("site1","site2","site3"),
                                  c("sp1","sp2","sp3","sp4")))

# community matrix with raw counts
example2 = matrix(c(10,0,20,70,0,0,40,60,80,10,10,0),
                  byrow = T, nrow = 3, ncol = 4,
                  dimnames = list(c("site1","site2","site3"),
                                  c("sp1","sp2","sp3","sp4")))

# counted nematodes
counts = c(132,312,213)

test1 = extrapolate.robust(example1,counts)
test2 = extrapolate.robust(example2,counts)

amynang/marcel documentation built on June 24, 2024, 10 p.m.