CHNOSZ (version 0.9-1)

revisit: Diversity Calculations for Chemical Species

Description

Calculate species richness, or standard deviation, coefficient of variation or Shannon diversity index of activities or logarithms of activities of chemical species, and plot the results.

Usage

revisit(logact, which = "cv", as.is = FALSE)
  richness(logact, logactmin = -4, mult = 1, as.is = FALSE)
  extremes(z, what = "minimum")
  where.extreme(z, what= "minimum")
  draw.diversity(d, which = "cv", logactmin = -4, col = par("fg"),
    as.is = FALSE, yline = 2, ylim = NULL, ispecies = NULL,
    add = FALSE, cex = par("cex"), lwd = par("lwd"), mar = par("mar"),
    mult = 1, side = 1:4)

Arguments

logact

list, logarithms of activities of species.

d

list, output from diagram.

which

character, type of diversity index to calculate.

as.is

logical, do not convert logarithms of activities to activities?

logactmin

numeric, cutoff value for a species to be counted in richness calculations.

z

numeric, matrix of values.

what

character, minimum or maximum.

col

character, color to use for line plots.

yline

numeric, margin line for y-axis label.

ylim

numeric, limits of y axis.

ispecies

numeric, which species to consider.

add

logical, add to an existing plot?

cex

numeric, character expansion factor.

lwd

numeric, line width.

mar

numeric, plot margin specifications.

mult

numeric, multiply richness counts by this factor.

side

numeric, which sides of plot to draw axes.

Value

• diversity and richness return a numeric (vector or matrix) that has the same dimensions as a single element of logact. extremes and where.extreme both return a list of $x$ and $y$ values. draw.diversity returnes a list with elements H (the result of either diversity or richness), ixmin and iymin (the column and row numbers where the maximum or minimum value is found), xmin and ymin (the values of the x and y variables where the maximum or minimum is found) and extval (the value of H at the maximum or minimum).

Details

revisit and richness both take as input a list of logarithms of activities of chemical species calculated using diagram. The calculations available in revisit are the Shannon index, standard deviation, and coefficient of variation, which are specified in which by shannon, sd and cv, respectively. By default the input is assumed to contain logarithms of activities which are exponentiated so that activities of species are used in the calculations, but if as.is is set to TRUE, the values are not changed.

richness simply counts the numbers of species whose logarithms of activities are above logactmin. The species can be weighted differently by supplying values of mult other than unity. If logactmin is any character value, its values are calculated as the mean of activities at each point (or mean of logarithms of activities if as.is is TRUE).

Given a matrix of numeric values in z, extremes locates the maximum or minimum values in both dimensions. That is, the $x$ values that are returned are the column numbers where the extreme is found for each row, and the $y$ values that are returned are the row numbers where the extreme is found for each column. where.extreme takes a matrix and returns the row and column numbers where the overall maximum or minimum value is found. If the extreme value is repeated, the row and column numbers for all instances are returned.

draw.diversity takes output from diagram, calls either diversity or richness, and plots the results. For systems of two variables where.extreme (and extremes) are used to indicate the location of the maximum Shannon index or richness (and the ridges around it), or minimum standard deviation or coefficient of variation (and the valleys around it). The ridges and valleys are plotted as dashed lines and are colored green for the $x$ values returned by extremes and blue for the $y$ values returned by extremes.

The name was changed from diversity to revisit in CHNOSZ-0.9 because there is a function named diversity in the vegan package. There is some overlap in the related functions from the different packages. The major difference in the interfaces is that diversity takes a matrix with species on the columns, while revisit takes a list with species as the elements of the list. See the examples for a calculation that uses revisit, as well as radfit from vegan.

Examples

data(thermo)
  ## carboxylases from different organisms
    # ribulose phosphate carboxylase/oxygenase
    rubisco <- c("Q6JAI0_9RHOD","A5CKC7_9CHRO","RBL_SYNJA","A1E8R4_9CHLO",
      "A8C9T6_9MYCO","A3EQE1_9BACT","A6YF84_9PROT", "RBL_BRAJA",
      "A1RZJ5_THEPD","RBL_METJA","A3DND9_STAMF","RBL_PYRHO") 
    # acetyl-coenzyme A carboxylase
    accoaco <- c("Q05KD0_HYDTH","Q9F7M8_PRB01","A6CDM2_9PLAN","A0GZU2_9CHLR",
      "ACCA_DEIRA","A1VC70_DESVV","A7WGI1_9AQUI","Q2JSS7_SYNJA",
      "A4AGS7_9ACTN","ACCA_AQUAE","ACCA_CAUCR","A6VIX9_METM7")  
    # calculate affinities as a function of T with
    # buffered logfH2 and CO2.. adjust the glutathione buffer 
    # (more oxidizing than default)
    mod.buffer("GSH-GSSG",c("GSH","GSSG"),logact=c(-3,-7))
    # add a CO2 gas saturation buffer
    mod.buffer("CO2","CO2","gas",-1)
    basis(c("CO2","H2O","NH4+","SO4-2","H2","H+"),
      c("CO2",0,-4,-4,"GSH-GSSG",-7))
    species(c(rubisco,accoaco))
    a <- affinity(T=c(0,160))
    # create speciation diagram with colors
    col <- c(rep("red",12),rep("blue",12))
    d <- diagram(a,residue=TRUE,color=col,ylim=c(-6,-1),legend.x=NULL)
    legend("topleft",col=c("red","blue"),lty=1,
      legend=c("ribulose phosphate carboxylase",
      "acetyl-coenzyme A carboxylase"))
    title(main=paste("Calculated relative abundances of",
      "carboxylases from different organisms",sep="\n"))
    # ... now on to a species richness diagram
    # all the proteins, then rubisco and accoaco
    draw.diversity(d,"richness",logactmin=-3.6)
    draw.diversity(d,"richness",logactmin=-3.6,
      ispecies=1:12,col="red",add=TRUE)
    draw.diversity(d,"richness",logactmin=-3.6,
      ispecies=13:24,col="blue",add=TRUE)
    legend("bottomleft",col=c("red","blue","black"),lty=1,
      legend=c("ribulose phosphate carboxylase",
      "acetyl-coenzyme A carboxylase","all"))
    title(main=paste("Carboxylases with activities",
      "greater than 10^(-3.6)",sep="\n"))

    ## continuing from above ... make a rank-abundance
    ## diagram and fit with a lognormal distribution
    if(require(vegan)) {
      basis("H2",-4)
      a <- affinity()
      logact <- diagram(a,residue=TRUE,do.plot=FALSE)$logact
      act <- 10^as.numeric(logact)
      # we use family=Gamma because our species have activities
      # (i.e., proportional abundances) and not integer counts
      mod <- rad.lognormal(act,family=Gamma)
      plot(mod,main=paste("Relative abundances of carboxylases",
        "fit with lognormal distribution",sep="\n"))
      # calculate Shannon diversity index
      # using revisit (CHNOSZ)
      H1 <- revisit(act,"shannon",as.is=TRUE)
      legend("topright",legend=paste("H =",round(H1,2)),pch=1)
      # using diversity (vegan)
      H2 <- diversity(matrix(act,nrow=1))
      stopifnot(isTRUE(all.equal(H1,H2)))
    }

    ### using grep.file, read.fasta, add.protein
    # calculations for Pelagibacter ubique
    f <- system.file("HTCC1062.faa",package="CHNOSZ")
    # what types of proteins we want (set to "" for all proteins)
    w <- "transcription"
    # locate entries whose names contain what we want
    j <- grep.file(f,w)
    # get the amino acid compositions of these protein
    p <- read.fasta(f,j)
    # add these proteins to CHNOSZ's inventory
    i <- add.protein(p)
    # set up a thermodynamic system
    basis("CHNOS+")
    # calculate affinities in logfO2-logaH2O space
    a <- affinity(O2=c(-120,-40),H2O=c(-40,40),iprotein=i)
    # calculate the logarithms of activities
    d <- diagram(a,do.plot=FALSE,mam=FALSE)
    # show the coefficient of variation of logarithms of activities
    draw.diversity(d,"cv")
    # make a title
    expr <- as.expression(substitute(x~y~"proteins in"~
      italic("P. ubique"),list(x=length(j),y=w)))
    mtitle(c("Coefficient of variation of log activities of",expr))