R/dbFD.R

`dbFD` <-
function(x, a , w, w.abun = TRUE, stand.x = TRUE, ord = c("podani", "metric"), asym.bin = NULL, corr = c("sqrt", "cailliez", "lingoes", "none"),  calc.FRic = TRUE, m = "max", stand.FRic = FALSE, scale.RaoQ = FALSE, calc.FGR = FALSE, clust.type = "ward", km.inf.gr = 2, km.sup.gr = nrow(x) - 1, km.iter = 100, km.crit = c("calinski", "ssi"),  calc.CWM = TRUE, CWM.type = c("dom", "all"), calc.FDiv = TRUE, dist.bin = 2, print.pco = FALSE, messages = TRUE){

# get tolerance value specifying small values that should be 0
tol <- .Machine$double.eps

# match and check arguments
corr <- match.arg(corr)
ord <- match.arg(ord)
CWM.type <- match.arg(CWM.type)
km.crit <- match.arg(km.crit)
if (!is.logical(messages) ) stop("'messages' must be TRUE or FALSE.", "\n")
if (!is.logical(stand.FRic) ) stop("'stand.FRic' must be TRUE or FALSE.","\n")
if (!is.logical(stand.x) ) stop("'stand.x' must be TRUE or FALSE.","\n")
if (!is.logical(w.abun) ) stop("'w.abun' must be TRUE or FALSE.","\n")
if (!is.logical(calc.FRic) ) stop("'calc.FRic' must be TRUE or FALSE.","\n")
if (!is.logical(calc.FDiv) ) stop("'calc.FDiv' must be TRUE or FALSE.","\n")
if (!is.logical(calc.FGR) ) stop("'calc.FGR' musts be TRUE or FALSE.","\n")
if (!is.logical(calc.CWM) ) stop("'calc.CWM' must be TRUE or FALSE.","\n")
if (!is.logical(scale.RaoQ) ) stop("'scale.RaoQ' must be TRUE or FALSE.","\n")
if (!is.logical(print.pco) ) stop("'print.pco' must be TRUE or FALSE.","\n")

# check coherence of number of species in 'x' and 'a'
if (is.matrix(x) | is.data.frame(x) )
{
	is.dist.x <- FALSE
	s.x <- dim(x)[1]
	t.x <- dim(x)[2]
	if (is.null(row.names(x) ) ) stop("'x' must have row names.","\n") else x.rn <- row.names(x)
}
if (is.vector(x) | is.factor(x) )
{	
	is.dist.x <- FALSE
	s.x <- length(x)
	t.x <- 1
	if (is.null(names(x) ) ) stop("'x' must have names.","\n") else x.rn <- names(x)
}
if (class(x)[1] == "dist" | class(x)[1] == "dissimilarity")
{
	is.dist.x <- TRUE
	s.x <- attr(x, "Size")
	t.x <- 1
	if (is.null(attr(x, "Labels") ) ) stop("'x' must have labels.","\n") else x.rn <- attr(x, "Labels")
}

# check if 'a' is present and create it if it is not
if (missing(a) )
{
	ab.names <- list("Community1", x.rn)
	a <- matrix(1, 1, s.x, dimnames = ab.names)
}
else
{
	if (is.matrix(a) | is.data.frame(a) )
	{
		s.a <- dim(a)[2]
		ab.t <- t(a)
		if (is.null(row.names(ab.t) ) ) stop("'a' must have column names.","\n") else ab.t.row <- row.names(ab.t)
		a <- as.matrix(a)
	}
	if (is.vector(a))
	{
		s.a <- length(a)
		if (is.null(names(a) ) ) stop("'a' must have names.","\n") else ab.t.row <- names(a)
		ab.names <- list("Community1", ab.t.row)
		a <- matrix(a, 1, s.a, dimnames = ab.names)
	}
	if (s.x != s.a) stop("Different number of species in 'x' and 'a'.","\n")
	if (any(ab.t.row != x.rn) ) stop("Species labels in 'x' and 'a' need to be identical and ordered alphabetically (or simply in the same order).","\n")
}

# replacement of NA in 'a' by 0
a <- as.matrix(a)
a[which(is.na(a) )] <- 0

# check if one community has total abundance of zero (no species)
abun.sum <- apply(a, 1, sum)
if (any(abun.sum == 0) ) stop("At least one community has zero-sum abundances (no species).","\n")

# check if one species has total abundance of zero (never occurs)
abun.sum2 <- apply(a, 2, sum)
if (any(abun.sum2 == 0) ) stop("At least one species does not occur in any community (zero total abundance across all communities).","\n")

# check for w
if (!missing(w) & is.dist.x) stop("When 'x' is a distance matrix, 'w' should be left missing.","\n")
if (!missing(w) & !is.dist.x){
				if (!is.numeric(w) | length(w) != t.x) stop("'w' should be a numeric vector of length = number of traits.","\n")
				else w <- w / sum(w)
				}
if (missing(w) ) w <- rep(1, t.x) / sum(rep(1, t.x) )

# check whether 'x' is a data frame, a matrix, a vector, or a distance matrix, and create x.dist
if (is.matrix(x) | is.data.frame(x) )
{
	x <- data.frame(x)
	if (t.x >= 2)
	{
		x.class <- sapply(x, data.class)
		if (any(x.class == "character") ) x[, x.class == "character"] <- as.factor(x[, x.class == "character"]) else x <- x
		# if all variables are numeric, have no NA, and have equal weights, distance measure = euclidean
		if (all(x.class == "numeric") & all(!is.na(x) ) )
		{
			if (length(unique(w)) == 1){
				x.s <- apply(x, 2, scale, center = TRUE, scale = stand.x)
				x.dist <- dist(x.s)
			}
			else{
				x.dist <- gowdis(x, w = w, ord = ord, asym.bin = asym.bin)
			}
		}
		else
		{
			x.dist <- gowdis(x, w = w, ord = ord, asym.bin = asym.bin)
		}
	}
	if (t.x == 1)
	{
		if (is.numeric(x[, 1]) )
		{
			if (all(!is.na(x) ) )
			{
				x.s <- apply(x, 2, scale, center = TRUE, scale = stand.x)
				x.dist <- dist(x.s)
			}
			if (any(is.na(x) ) )
			{
				pos.NA <- which(is.na(x), arr.ind = TRUE)
				x <- na.omit(x)
				x.s <- apply(x, 2, scale, center = TRUE, scale = stand.x)
				x.dist <- dist(x.s)
				row.excl.ab <- pos.NA[, 1]
				a <- a[, -row.excl.ab]
				if (messages) cat("Warning: Species with missing trait values have been excluded.","\n")
			}
		}
		if (is.factor(x[, 1]) | is.character(x[, 1]) )
		{
			if (is.ordered(x[,1])) x <- x else x[, 1] <- as.factor(x[, 1])
			if (any(is.na(x) ) ) 
			{
				pos.NA <- which(is.na(x), arr.ind = TRUE)
				x <- na.omit(x)
				row.excl.ab <- pos.NA[, 1]
				a <- a[, -row.excl.ab]
				x.rn <- x.rn[-pos.NA]
				if (messages) cat("Warning: Species with missing trait values have been excluded.","\n")
			}
			if (is.ordered(x[, 1]) )
			{
				x.s <- data.frame(rank(x[, 1]) )
				names(x.s) <- x.rn
				x.dist <- dist(x.s)
			}
			else
			{
				x.f <- as.factor(x[, 1])
				x.dummy <- diag(nlevels(x.f) )[x.f, ]
				x.dummy.df <- data.frame(x.dummy, row.names = x.rn)
				sequence <- 1:10
				if (all(dist.bin != sequence[any(sequence)] ) ) stop("'dist.bin' must be an integer between 1 and 10.","\n")
				x.dist <- dist.binary(x.dummy.df, method = dist.bin)
			}
		}

	}	
}

if (is.vector(x) & is.numeric(x) )
{
	if (any(is.na(x) ) )
	{
		pos.NA <- which(is.na(x) )
		x <- na.omit(x)
		a <- a[, -pos.NA]
		x.rn <- x.rn[-pos.NA]
		if (messages) cat("Warning: Species with missing trait values have been excluded.","\n")
	}
	else x <- x
	x.s <- scale(x, center=T, scale = stand.x)
	x.dist <- dist(x.s)
	x <- data.frame(x)
	dimnames(x) <- list(x.rn, "Trait")
}
if (is.vector(x) & is.character(x) )
{
	# convert to factor
	x <- as.factor(x)
	if (any(is.na(x) ) )
	{
		pos.NA <- which(is.na(x) )
		x <- na.omit(x)
		a <- a[, -pos.NA]
		x.rn <- x.rn[-pos.NA]
		if (messages) cat("Warning: Species with missing trait values have been excluded.","\n")
	}
	else x <- x
	dimnames(x) <- list(x.rn, "Trait")
	# create matrix of dummy variables
	x.dummy <- diag(nlevels(x) )[x, ]
	# keep species names
	x.dummy.df <- data.frame(x.dummy, row.names = x.rn)
	sequence <- 1:10
	if (all(dist.bin != sequence[any(sequence)]) ) stop("'dist.bin' must be an integer between 1 and 10.","\n")
	x <- data.frame(x)
	x.dist <- dist.binary(x.dummy.df, method = dist.bin)
	
}
if (is.ordered(x) )
{
	if (any(is.na(x) ) )
	{
		pos.NA <- which(is.na(x) )
		x <- na.omit(x)
		a <- a[, -pos.NA]
		x.rn <- x.rn[-pos.NA]
		cat("Warning: Species with missing trait values have been excluded.","\n")
	}
	else x <- x
		x <- data.frame(x)
		dimnames(x) <- list(x.rn, "Trait")
		x.dist <- gowdis(x, w = w, ord = ord, asym.bin = asym.bin)

}
if (is.factor(x) & !is.ordered(x) )
{
	if (any(is.na(x) )) 
	{
		pos.NA <- which(is.na(x) )
		x <- na.omit(x)
		a <- a[, -pos.NA]
		x.rn <- x.rn[-pos.NA]
		if (messages) cat("Warning: Species with missing trait values have been excluded.","\n")
	}
	else x <- x

	# create matrix of dummy variables
	x.dummy <- diag(nlevels(x) )[x, ]
	# keep species names
	x.dummy.df <- data.frame(x.dummy, row.names = x.rn)
	sequence <- 1:10
	if (all(dist.bin != sequence[any(sequence)] ) ) stop("'dist.bin' must be an integer between 1 and 10.","\n")
	x.dist <- dist.binary(x.dummy.df, method = dist.bin)
	x <- data.frame(x)
	dimnames(x) <- list(x.rn, "Trait")
}
if (class(x)[1] == "dist" | class(x)[1]=="dissimilarity")
{
	if (any(is.na(x) ) ) stop("When 'x' is a distance matrix, it cannot have missing values (NA).","\n")
	x.dist <- x
}
if (any(is.na(x.dist) ) ) stop("NA's in the distance matrix.","\n")


# check whether one or more species has no traits
if (!is.dist.x)
{
	no.traits <- apply(x, 1, function(v) length(v[!is.na(v)]) )
	if (any(no.traits == 0) ) stop("At least one species has no trait data.","\n")
}
# c = number of communities
c <- dim(a)[1]

# check for w.abun = TRUE or FALSE. If false, replace all abundances >0 by 1
if (!w.abun) for (h in 1:c) {abpos <- which(a[h, ] > 0) ; a[h, abpos] <- 1}
#
# check if x.dist is euclidean or not, and apply correction
attr(x.dist, "Labels" ) <- x.rn
if (is.euclid(x.dist) ) x.dist2 <- x.dist
if (!is.euclid(x.dist) )
{
	if (corr == "lingoes")
	{
		x.dist2 <- lingoes(x.dist)
		if (messages) cat("Species x species distance matrix was not Euclidean. Lingoes correction was applied.","\n")
	}
	if (corr == "cailliez")
	{
		x.dist2 <- cailliez(x.dist)
		if (messages) cat("Species x species distance matrix was not Euclidean. Cailliez correction was applied.","\n")
	}
	if (corr == "sqrt")
	{
		x.dist2 <- sqrt(x.dist)
		# check if sqrt correction actually worked
		if(!is.euclid(x.dist2) ) stop("Species x species distance matrix was still is not Euclidean after 'sqrt' correction. Use another correction method.","\n")
		if (is.euclid(x.dist2) ) if(messages) cat("Species x species distance matrix was not Euclidean. 'sqrt' correction was applied.","\n")
	}
	if (corr == "none")
	{
	x.dist2 <- quasieuclid(x.dist)
	if (messages) cat("Species x species distance was not Euclidean, but no correction was applied. Only the PCoA axes with positive eigenvalues were kept.","\n")
	}
}

# perform PCoA on x.dist2
x.pco <- dudi.pco(x.dist2, scannf = FALSE, full = TRUE)
traits <- round(x.pco$li, .Machine$double.exponent)

# calculate nb.sp to decide if dimensionality reduction is required for convex hull volume
# nb.sp = number of unique species in each community
nb.sp <- numeric(c)

for (i in 1:c){
		# selection of species present in each community
		sp.pres <- which(a[i, ] > 0)

		#  number of unique species in the community
                traits.sp.pres <- traits[sp.pres, , drop = F ]
                traits.sp.pres[traits.sp.pres != 0 & abs(traits.sp.pres) < tol] <- 0
		nb.sp[i] <- nrow(unique(traits.sp.pres ) )
	}

names(nb.sp) <- row.names(a)
# find number of unique species in community with the fewest unique species
min.nb.sp <- min(nb.sp)

if (min.nb.sp < 3) if (messages) cat("FEVe: Could not be calculated for communities with <3 functionally singular species.","\n")
if (min.nb.sp < 2) if (messages) cat("FDis: Equals 0 in communities with only one functionally singular species.","\n")
#
# pre-steps for calculating FRic	
if (calc.FRic)
{
	# find classes of traits in x
	# note: when 'x' is a distance or dissimilarity matrix, running sapply(x, data.class) will only return 'numeric', so that's ok
	x.class2 <- sapply(x, data.class)
	if (all(x.class2 == "factor" | x.class2 == "ordered") )
	{
		if (length(x.class2) == 1 & x.class2[1] == "ordered")
		{
			traits.FRic1 <- rank(x[, 1] ) ; names(traits.FRic1) <- x.rn ; traits.FRic <- data.frame(traits.FRic1)
			qual.FRic = 1
			if (messages) cat("FRic: Only one ordinal trait present in 'x'. FRic was measured as the range of the ranks, NOT as the convex hull volume.","\n")
			if (calc.FDiv)
			{	
				calc.FDiv <- FALSE
				if (messages) cat("FDiv: Cannot be computed when 'x' is a single ordinal trait.","\n")
			}
			if (stand.FRic)
			{
				traits.range <- range(traits.FRic[, 1] )
				FRic.all <- traits.range[2] - traits.range[1]
			}
		}	
		else
		{
			# when all traits are categorical or ordinal, FRic is not measured as the convex hull volume, but as the number of unique trait combinations
			traits.FRic <- x
			qual.FRic = 1
			if(messages) cat("FRic: Only categorical and/or ordinal trait(s) present in 'x'. FRic was measured as the number of unique trait combinations, NOT as the convex hull volume.","\n")
			if (stand.FRic) FRic.all <- nrow((unique(traits.FRic) ) )
			if (calc.FDiv)
			{	
				calc.FDiv <- FALSE
				if (messages) cat("FDiv: Cannot be computed when only categorical and/or ordinal trait(s) present in 'x'.","\n")
			}
		}
	}
	else
	{	
		# when there is only one continuous trait or one dimension (when 'x' is a distance matrix), the range is used instead of the convex hull volume
		if (x.pco$nf == 1)
		{
			traits.FRic <- x.pco$li
			qual.FRic = 1
			if (messages) cat("FRic: Only one continuous trait or dimension in 'x'. FRic was measured as the range, NOT as the convex hull volume.","\n")
			if (calc.FDiv)
			{	
				calc.FDiv <- FALSE
				if (messages) cat("FDiv: Cannot not be computed when 'x' contains one single continuous trait or dimension.","\n")
			}
			if (stand.FRic)
			{
				traits.range <- range(traits.FRic[, 1] )
				FRic.all <- traits.range[2] - traits.range[1]
			}
		}
		# when there is more than one trait or dimensions and that at least one of these traits is numeric, FRic is measured as the convex hull volume
		if (x.pco$nf > 1)
		{
			# calculate m.min and m.max
			# for the s >= 2^t condition to be respected, there needs to be at least 4 species (this gives two traits max)
			warning <- FALSE
			m.max <- min.nb.sp - 1
			if (m == "min")
			{
				warning <- TRUE
				if (min.nb.sp < 4)
				{
					nb.sp2 <- nb.sp[nb.sp > 3]
					m.min <- floor(log2(min(nb.sp2) ) )
					if (messages) cat("FRic: To respect s >= 2^t, FRic could not be calculated for communities with <4 functionally singular species.","\n")
				}
				else m.min <- floor(log2(min.nb.sp) )
			}
			else
			{
				if (min.nb.sp < 3)
				{
					nb.sp2 <- nb.sp[nb.sp > 2]
					m.max <- min(nb.sp2) - 1
					if (messages) cat("FRic: To respect s > t, FRic could not be calculated for communities with <3 functionally singular species.","\n")
				}
				else m.max <- m.max
			}
			
			# transform m
			if (is.numeric(m) & m <= 1) stop("When 'm' is an integer, it must be >1.","\n")
			if (is.numeric(m) & m > m.max) m <- m.max
			if (m == "min") m <- m.min
			if (m == "max") m <- m.max
			if (!is.numeric(m) & m != "min" & m != "max") stop("'m' must be an integer >1, 'min', or 'max'.","\n")
			#	
				# selection of traits
				if (m < x.pco$nf)
				{
					traits.FRic <- x.pco$li[, 1:m]
					if (x.pco$nf - m == 1) if (messages) cat("FRic: Dimensionality reduction was required. The last PCoA axis (out of",x.pco$nf,"in total) was removed.","\n")
					if (x.pco$nf - m > 1) if (messages) cat("FRic: Dimensionality reduction was required. The last",x.pco$nf-m,"PCoA axes (out of",x.pco$nf,"in total) were removed.","\n")
					if (is.euclid(x.dist) )
					{		
						qual.FRic <- sum(x.pco$eig[1:m]) / sum(x.pco$eig)
						if (messages) cat("FRic: Quality of the reduced-space representation =",qual.FRic,"\n")
					}
					if (!is.euclid(x.dist) & corr != "none")
					{
						qual.FRic <- sum(x.pco$eig[1:m]) / sum(x.pco$eig)
						if (messages) cat("FRic: Quality of the reduced-space representation (based on corrected distance matrix) =",qual.FRic,"\n")
					}
					if (!is.euclid(x.dist) & corr == "none")
					{		
						# compute the eigenvalues (including negative ones)
						delta <- -0.5*bicenter.wt(x.dist * x.dist)
						lambda <- eigen(delta, symmetric = TRUE, only.values = TRUE)$values
						# calculate corrected R^2-like ratio
						sum.m <- sum(lambda[1:m])
						sum.n <- sum(lambda)
						lambda.neg <- c(lambda[lambda < 0])
						max.neg <- abs(min(lambda.neg) )
						qual.FRic <- (sum.m + (length(lambda[1:m]) * max.neg) ) / (sum.n + ((length(lambda) - 1) * max.neg) )
						if (messages) cat("FRic: Quality of the reduced-space representation (taking into account the negative eigenvalues) =",qual.FRic,"\n")
					}
				}
				if (m >= x.pco$nf)
				{
					qual.FRic = 1
					traits.FRic <- x.pco$li
					if (x.pco$nf == 2) if (messages) cat("FRic: No dimensionality reduction was required. The 2 PCoA axes were kept as 'traits'.","\n")
					if (x.pco$nf > 2) if (messages) cat("FRic: No dimensionality reduction was required. All",x.pco$nf,"PCoA axes were kept as 'traits'.","\n")
				}
				#
				# calculate (or not) the global hull volume
				if (stand.FRic)
				{
					hull.all <- convhulln(traits.FRic, "FA")
					FRic.all <- hull.all$vol
				}
		} # end of if (x.pco$nf > 1)
	} # end of else	
} # end of pre-steps for FRic

# pre-steps for FDiv
# when calc.FDiv is TRUE, calc.FRic must also be true
if (!calc.FRic & calc.FDiv) cat("FDiv: Cannot be computed when 'calc.FRic' is FALSE.", "\n")
if (calc.FRic & calc.FDiv) if (min.nb.sp < 3) if (messages) cat("FDiv: Could not be calculated for communities with <3 functionally singular species.","\n")

# pre-steps for calculating FGR
if (calc.FGR)
{
	if (clust.type == "kmeans")
	{
		tr.clust <- cascadeKM(traits, km.inf.gr, km.sup.gr, km.iter, km.crit)
		cat("FGR: Summary of kmeans clustering\n")
		cat("\nPartition\n")
		print(tr.clust$partition)
		cat("\nResults\n")
		print(tr.clust$results)
		cat("\nSize\n")
		print(tr.clust$size)
		plot(tr.clust)
		part.names <- colnames(tr.clust$partition)
		part.names <- as.numeric(substr(part.names, 1, 1) )
		cat("\nFGR: How many groups?",'\n')
		cut.g <- toupper(scan(file="",what="character",nlines=1,quiet=T))
		cut.gr <- as.integer(cut.g)
		if (cut.gr < km.inf.gr | cut.gr > km.sup.gr) stop("You must type an integer between 'km.ing.gr' and 'km.sup.gr'.","\n")
		spfgr.all <- tr.clust$partition[, part.names == cut.gr]
		names(spfgr.all) <- x.rn
	}
	else
	{
		tr.clust <- hclust(x.dist, method = clust.type)
		plot(tr.clust, main = "Cluster dengrogram of species based on functional traits")
		cat("FGR: Do you want to cut the dendrogram from height or from the number of groups? Type 'h' for height, 'g' for groups.",'\n')
		cut <- toupper(scan(file = "", what = "character" , nlines = 1, quiet = T) )
		if(cut == "H")
		{
			cat("FGR: At what height do you want the dendrogram to be cut?",'\n')
			cut.d <- toupper(scan(file = "", what = "character", nlines = 1, quiet = T) )
			cut.dist <- as.numeric(cut.d)
			spfgr.all <- cutree(tr.clust, h = cut.dist)
		}
		if (cut == "G")
 		{
			cat("FGR: How many groups?",'\n')
			cut.g <- toupper(scan(file = "", what = "character", nlines = 1, quiet = T) )
			cut.gr <- as.integer(cut.g)
			spfgr.all <- cutree(tr.clust, k = cut.gr)
		}
		if (cut != "H" & cut != "G") stop("You must type 'h' or 'g'","\n")
	}

# compute abundances of individual functional groups
a.t <- t(a)
by.gr <- list(spfgr.all)
gr.abun <- aggregate(a.t, by.gr, sum)
lab <- paste("group", gr.abun[,1], sep="")
gr.abun <- data.frame(t(gr.abun[,-1])) ; colnames(gr.abun) <- lab ; rownames(gr.abun) <- rownames(a)


}
#
# pre-steps for calculating CWM
if (is.matrix(x) | is.data.frame(x) & calc.CWM)
{

#=================
# CWM
#=================

CWM <- functcomp(x, a, CWM.type = CWM.type)
}

if (calc.CWM & class(x)[1] == "dist" | class(x)[1] == "dissimilarity") if (messages) cat("CWM: When 'x' is a distance matrix, CWM cannot be calculated.","\n") 


#=================
# Rao's Q
#=================

# this is the div function of the ade4 package
# it is copied here only to remove the warning message about the distance matrix being not Euclidean.

divc <- function (df, dis = NULL, scale = FALSE) 
{
    if (!inherits(df, "data.frame")) 
        stop("Non convenient df")
    if (any(df < 0)) 
        stop("Negative value in df")
    if (!is.null(dis)) {
        if (!inherits(dis, "dist")) 
            stop("Object of class 'dist' expected for distance")
        dis <- as.matrix(dis)
        if (nrow(df) != nrow(dis)) 
            stop("Non convenient df")
        dis <- as.dist(dis)
    }
    if (is.null(dis)) dis <- as.dist((matrix(1, nrow(df), nrow(df)) - diag(rep(1, nrow(df)))) * sqrt(2))
    div <- as.data.frame(rep(0, ncol(df)))
    names(div) <- "diversity"
    rownames(div) <- names(df)
    for (i in 1:ncol(df)) {
        if (sum(df[, i]) < 1e-16) 
            div[i, ] <- 0
        else div[i, ] <- (t(df[, i]) %*% (as.matrix(dis)^2) %*% 
            df[, i])/2/(sum(df[, i])^2)
    }
    if (scale == TRUE) {
        divmax <- divcmax(dis)$value
        div <- div/divmax
    }
    return(div)
}

RaoQ <- divc(data.frame(t(a)), x.dist, scale = scale.RaoQ)
RaoQ <- RaoQ[, 1] ; names(RaoQ) <- rownames(a)

#=================
# FDis
#=================

disp <- fdisp(x.dist, a)
FDis <- disp$FDis

# calculate nbsp, FRic, FEve, FDis, and FGR
# definition of vector for results, with communities'names as given in 'a'
nbsp <- rep(NA, c) ; names(nbsp) <- row.names(a)
FRic <- rep(NA, c) ; names(FRic) <- row.names(a)
FEve <- rep(NA, c) ; names(FEve) <- row.names(a)
FGR <- rep(NA, c) ; names(FGR) <- row.names(a)
FDiv <- rep(NA, c) ; names(FDiv) <- row.names(a)

for (i in 1:c)
{
	# selection of species present in the community and filtering of trait matrix
	sppres <- which(a[i ,] > 0)
	# number of species in the community
	S <- length(sppres) ; nbsp[i] <- S
	# filter on 'traits' and 'abundances' to keep only values of species present in the community
	tr <- data.frame(traits[sppres, ])
	if (calc.FRic) tr.FRic <- data.frame(traits.FRic[sppres, ])
	ab <- as.matrix(a[i, sppres])
 	# scaling of abundances
	abundrel <- ab / sum(ab)
	
	#=================
	# FRic
	#=================
	if (calc.FRic)
	{	
		if (all(x.class2 == "factor" | x.class2 == "ordered"))
		{
			if (length(x.class2) == 1 & x.class2[1] == "ordered")
			{
				tr.range <- range(tr.FRic[, 1])
				t.range <- tr.range[2] - tr.range[1]
				if (!stand.FRic) FRic[i] <- t.range
				if (stand.FRic) FRic[i] <- t.range / FRic.all
			}
			else
			{
				if (!stand.FRic) FRic[i] <- nrow((unique(tr.FRic) ) )
				if (stand.FRic) FRic[i] <- nrow( (unique(tr.FRic) ) ) / FRic.all
			}
		}
		else
		{
			if (dim(tr.FRic)[2] > 1 & nb.sp[i] >= 3)
			{
				if (warning) thresh <- 4
				if (!warning) thresh<-3
				if (nb.sp[i] >= thresh)
				{
					convhull <- convhulln(tr.FRic, "FA")
					if (!stand.FRic) FRic[i] <- convhull$vol
					if (stand.FRic) FRic[i] <- convhull$vol / FRic.all
				}
				else{}
			}
			# when there is only one trait, FRic is the range instead of the convex hull volume
			if (dim(tr.FRic)[2] == 1)
			{
				tr.range <- range(tr.FRic[, 1])
				t.range <- tr.range[2] - tr.range[1]
				if (!stand.FRic) FRic[i] <- t.range
				if (stand.FRic) FRic[i] <- t.range / FRic.all
			}
		}
	} # end of if (calc.FRic)
	
	#=================
	# FEve
	#=================
	
        if (nb.sp[i] >= 3)
	{
		# computation of minimum spanning tree and conversion of the 'mst' matrix into 'dist' class
		tr.dist <- dist(tr)
		linkmst <- mst(tr.dist) ; mstvect <- as.dist(linkmst)
		# computation of the pairwise cumulative relative abundances and conversion into 'dist' class
		abund2 <- matrix(0, nrow = S, ncol = S)
		for (q in 1:S) for (r in 1:S) abund2[q, r] <- abundrel[q] + abundrel[r]
		abund2vect <- as.dist(abund2)
		# computation of EW for the (S - 1) segments to link the S points
		EW <- rep(0, S - 1)
		flag <- 1
		for (m in 1 : ((S - 1) * S / 2)) {if (mstvect[m] != 0) {EW[flag] <- tr.dist[m] / (abund2vect[m]) ; flag <- flag + 1}}
		# computation of the PEW and comparison with 1 / S - 1, finally computation of FEve
		minPEW <- rep(0, S - 1)  ;  OdSmO <- 1 / (S - 1)
		for (l in 1 : (S - 1) )
		minPEW[l] <- min((EW[l] / sum(EW)), OdSmO)
		FEve[i] <- ((sum(minPEW)) - OdSmO) / (1 - OdSmO)
	}
	

	#=================
	# FDiv
	#=================
	
	if (calc.FDiv & calc.FRic)
	{	
		if (any(x.class2 == "numeric") & dim(tr.FRic)[2] > 1 & nb.sp[i] >= 3)
		{
			# calculate vertices
			vert0 <- convhulln(tr.FRic, "Fx TO 'vert.txt'")
			vert1 <- scan("vert.txt", quiet = T)
			vert2 <- vert1 + 1
			vertices <- vert2[-1]
			# traits values of vertices
			trvertices <- tr.FRic[vertices, ]
			# coordinates of the center of gravity of the vertices (Gv)
			baryv <- apply(trvertices, 2, mean)
			# euclidian dstances to Gv (dB)
			distbaryv <- rep(0, S)
			for (j in 1:S) distbaryv[j] <- (sum((tr.FRic[j, ] - baryv)^2) ) ^0.5
			# mean of dB values
			meandB<-mean(distbaryv)
			# deviations to mean of db
			devdB <- distbaryv - meandB
			# relative abundances-weighted mean deviation
			abdev2 <- abundrel * devdB
			# relative abundances-weighted mean of absolute deviations
			ababsdev2 <- abundrel * abs(devdB)
			# computation of FDiv
			FDiv[i] <- (sum(abdev2) + meandB) / (sum(ababsdev2) + meandB)
		}
	}
		
	#=================
	# FGR
	#=================
	
	# computation of FGR
	if (calc.FGR) FGR[i] <- length(unique(spfgr.all[sppres]))
	
} # end of nbsp, FRic, FEve, FDis, FDiv, RaoQ, and FGR

res <- list()
res$nbsp <- nbsp
res$sing.sp <- nb.sp
if (calc.FRic) res$FRic <- FRic
if (calc.FRic) res$qual.FRic <- qual.FRic
res$FEve <- FEve
if (calc.FDiv) res$FDiv <- FDiv
res$FDis <- FDis
res$RaoQ <- RaoQ
if (calc.FGR)
{
	res$FGR <- FGR
	res$spfgr <- spfgr.all
	res$gr.abun <- gr.abun
}
if (is.matrix(x) | is.data.frame(x) & calc.CWM) res$CWM <- CWM
# return eigenvalues and PCoA axes
if (print.pco)
{
	res$x.values <- x.pco$eig
	res$x.axes <- x.pco$li
}
invisible(res)

}

Try the FD package in your browser

Any scripts or data that you put into this service are public.

FD documentation built on May 1, 2019, 9:27 p.m.