cytoDiv: Cytometric Diversity Indices

In armbrustlab/cytDiv: Cytometric diversity indices

Description

Calculate the various cytometric diversity indices described in Li, W. 1997 Cytometric diversity in marine ultraphytoplankton. Limnology and Oceanography 42: 874 - 880.

Usage

cytoDiv(df, para=c("fsc_small", "chl_small", "pe", "fsc_perp"), Ncat = 16, log=TRUE, do.plot=FALSE)

Arguments

df	flow cytometry data converted into a dataframe.
para	name(s) or indice(s) of the parameters . Default is c("fsc_small", "chl_small", "pe", "fsc_perp") from SeaFlow instrument. IMPORTANT: Max number of parameters limited to 4
Ncat	Number of categories per variable. Default is 16
do.plot	Option to plot the probability distribution of the bivariate data. Only for 1 and 2 parameters

Value

A data frame that contains the Richness index N0 (number of categories), the Shannon-Wiener Diversity index H', the reciprocal of Simpson's index N2, the Simpson's Index of Diversity D and the Evenness index J'. Indices calculation is based on Hill's method (Hill, M.O. 1973 Diversity and evenness: A unifying notation and its consequences. Ecology 54: 427 - 432).

Note

The highest number of distinct categories (Ncat) in cytometric classification depends on the resolution at which each variable is measured, which is controlled by the bit resolution (Nbit) of the Analog-to-Digital Converter. In the example below, measurements of single-cell forward light scatter and red fluorescence were collected using logarithmic amplification and recorded in relative units in a four-decade range spanned by 2^16 channels. For the calculation of the diversity indices, data resolution of each variable was reduced to 16 channels by successively binning the counts of 2^12 adjacent channels . In this case, the maximum number of possible categories in the light scatter-fluorescence domain was therefore 16x16 (=256).

Examples

## reading from a fcs dataframe

fcs.df <- system.file("extdata","fcs_dataframe.csv", 
				package="cytoDiv")
df <- read.csv(fcs.df)

## looking at the first rows of the data frame
head(df)

## plotting bivariate plot
cols <- colorRampPalette(c("blue4","royalblue4","deepskyblue3", "seagreen3", "yellow", "orangered2","darkred")) 

par(mfrow=c(3,1))
plot(df[,2], df[,3], col= densCols(df[,2], df[,3], colramp = cols))

# Remove the Internal Standard and noise before calculating the Cytometric Diversity Indices
cleaned.df <- subset(df, df[,4]  < df[,3] + 13000 & df[,3] > 5000)
plot(cleaned.df[,2], cleaned.df[,3], col= densCols(cleaned.df[,2], cleaned.df[,3], colramp = cols))

#Calculating the Cytometric Diversity Indices
div <- cytoDiv(cleaned.df, para=c("FSC","F692.4"), log=FALSE, do.plot=T)

print(div)

armbrustlab/cytDiv documentation built on May 10, 2019, 1:40 p.m.