Compute Summary Statistics of Data Subsets

Compute summary statistics (sums, means) of data subsets.

Usage

groupSums(object, ...)
# S4 method for class 'matrix'
groupSums(object, MARGIN, by, na.rm = FALSE, ...)
# S4 method for class 'sparseMatrix'
groupSums(object, MARGIN, by, na.rm = FALSE, ...)
# S4 method for class 'Mefa'
groupSums(object, MARGIN, by, replace, na.rm = FALSE, ...)

groupMeans(object, ...)
# S4 method for class 'matrix'
groupMeans(object, MARGIN, by, na.rm = FALSE, ...)
# S4 method for class 'sparseMatrix'
groupMeans(object, MARGIN, by, na.rm = FALSE, ...)
# S4 method for class 'Mefa'
groupMeans(object, MARGIN, by, replace, na.rm = FALSE, ...)

sum_by(x, by)

Arguments

object: an object.
x: a vector.
MARGIN: numeric, 1 indicates rows are to be summed/averaged, 2 indicates columns are to be summed/averaged. c(1, 2) is not yet implemented, but can be calculated calling the function twice.
by: a vector of grouping elements corresponding to dimensions of object and MARGIN.
replace: a data frame to be used when applying the method on a "Mefa" object. The attribute table corresponding to MARGIN is dropped (NULL), replacement table can be specified via this argument.
na.rm: logical. Should missing values be removed? Sum is calculated by zeroing out NA values, mean is calculated as dividing the sum by the number of non-NA values when collapsing.
...: other argument, currently not implemented.

Details

The method sums/averages cells in a matrix. The functions behind these methods use sparse matrices, so calculations can be more efficient compared to using aggregate.

Value

An object similar to the input one.

Author

Peter Solymos <solymos@ualberta.ca>

Examples

x <- data.frame(
    sample = paste("Sample", c(1,1,2,2,3,4), sep="."),
    species = c(paste("Species", c(1,1,1,2,3), sep="."),
    "zero.pseudo"), count = c(1,2,10,3,4,0),
    stringsAsFactors = TRUE)
samp <- data.frame(samples=levels(x$sample), var1=1:2,
    stringsAsFactors = TRUE)
taxa <- data.frame(specnames=levels(x$species), var2=c("b","a"),
    stringsAsFactors = TRUE)
rownames(samp) <- samp$samples
rownames(taxa) <- taxa$specnames
x2 <- Xtab(count ~ sample + species, x, cdrop=FALSE,rdrop=TRUE)
x5 <- Mefa(x2, samp, taxa, join="inner")

groupSums(as.matrix(x2), 1, c(1,1,2))
#>   Species.1 Species.2 Species.3 zero.pseudo
#> 1        13         3         0           0
#> 2         0         0         4           0
groupSums(as.matrix(x2), 2, c(1,1,2,2))
#>           1 2
#> Sample.1  3 0
#> Sample.2 13 0
#> Sample.3  0 4
groupSums(x2, 1, c(1,1,2))
#> 2 x 4 sparse Matrix of class "dgCMatrix"
#>   Species.1 Species.2 Species.3 zero.pseudo
#> 1        13         3         .           .
#> 2         .         .         4           .
groupSums(x2, 2, c(1,1,2,2))
#> 3 x 2 sparse Matrix of class "dgCMatrix"
#>           1 2
#> Sample.1  3 .
#> Sample.2 13 .
#> Sample.3  . 4
groupSums(x5, 1, c(1,1,2))
#> Object of class "Mefa"
#>   ..@ xtab: 2 x 4 sparse Matrix
#>   ..@ samp: NULL
#>   ..@ taxa: data frame with 2 variables
#>   ..@ join: inner 
groupSums(x5, 2, c(1,1,2,2))
#> Object of class "Mefa"
#>   ..@ xtab: 3 x 2 sparse Matrix
#>   ..@ samp: data frame with 2 variables
#>   ..@ taxa: NULL
#>   ..@ join: inner 

groupMeans(as.matrix(x2), 1, c(1,1,2))
#>   Species.1 Species.2 Species.3 zero.pseudo
#> 1       6.5       1.5         0           0
#> 2       0.0       0.0         4           0
groupMeans(as.matrix(x2), 2, c(1,1,2,2))
#>            1 2
#> Sample.1 1.5 0
#> Sample.2 6.5 0
#> Sample.3 0.0 2
groupMeans(x2, 1, c(1,1,2))
#> 2 x 4 sparse Matrix of class "dgCMatrix"
#>   Species.1 Species.2 Species.3 zero.pseudo
#> 1       6.5       1.5         .           .
#> 2       .         .           4           .
groupMeans(x2, 2, c(1,1,2,2))
#> 3 x 2 sparse Matrix of class "dgCMatrix"
#>            1 2
#> Sample.1 1.5 .
#> Sample.2 6.5 .
#> Sample.3 .   2
groupMeans(x5, 1, c(1,1,2))
#> Object of class "Mefa"
#>   ..@ xtab: 2 x 4 sparse Matrix
#>   ..@ samp: NULL
#>   ..@ taxa: data frame with 2 variables
#>   ..@ join: inner 
groupMeans(x5, 2, c(1,1,2,2))
#> Object of class "Mefa"
#>   ..@ xtab: 3 x 2 sparse Matrix
#>   ..@ samp: data frame with 2 variables
#>   ..@ taxa: NULL
#>   ..@ join: inner 

sum_by(runif(100, 0, 1), sample(LETTERS[1:4], 100, replace=TRUE))
#>           x by
#> C 17.099100 29
#> A  7.992116 18
#> D 14.054393 29
#> B 12.379071 24