R/QFeatures-aggregation.R, R/SummarizedExperiment-methods.R
QFeatures-aggregate.RdThis function aggregates the quantitative features of an assay,
applying a summarisation function (fun) to sets of features.
The fcol variable name points to a rowData column that defines
how to group the features during aggregate. This variable can
eigher be a vector (we then refer to an aggregation by vector)
or an adjacency matrix (aggregation by matrix).
The rowData of the aggregated SummarizedExperiment assay
contains a .n variable that provides the number of parent
features that were aggregated.
When aggregating with a vector, the newly aggregated
SummarizedExperiment assay also contains a new aggcounts assay
containing the aggregation counts matrix, i.e. the number of
features that were aggregated for each sample, which can be
accessed with the aggcounts() accessor.
# S4 method for QFeatures
aggregateFeatures(
object,
i,
fcol,
name = "newAssay",
fun = MsCoreUtils::robustSummary,
...
)
# S4 method for SummarizedExperiment
aggregateFeatures(object, fcol, fun = MsCoreUtils::robustSummary, ...)
# S4 method for QFeatures
adjacencyMatrix(object, i, adjName = "adjacencyMatrix")
adjacencyMatrix(object, i, adjName = "adjacencyMatrix") <- value
# S4 method for SummarizedExperiment
aggcounts(object, ...)An instance of class SummarizedExperiment or
QFeatures.
When adding an adjacency matrix to an assay of a
QFeatures object, the index or name of the assay the
adjacency matrix will be added to. Ignored when x is an
SummarizedExperiment.
A character(1) naming a rowdata variable (of assay
i in case of a QFeatures) defining how to aggregate the
features of the assay. This variable is either a character
or a (possibly sparse) matrix. See below for details.
A character(1) naming the new assay. Default is
newAssay. Note that the function will fail if there's
already an assay with name.
A function used for quantitative feature aggregation. See Details for examples.
Additional parameters passed the fun.
character(1) with the variable name containing
the adjacency matrix. Default is "adjacencyMatrix".
An adjacency matrix with row and column names. The
matrix will be coerced to compressed, column-oriented sparse
matrix (class dgCMatrix) as defined in the Matrix package,
as generaled by the sparseMatrix() constructor.
A QFeatures object with an additional assay or a
SummarizedExperiment object (or subclass thereof).
Aggregation is performed by a function that takes a matrix as
input and returns a vector of length equal to ncol(x). Examples
thereof are
MsCoreUtils::medianPolish() to fits an additive model (two way
decomposition) using Tukey's median polish_ procedure using
stats::medpolish();
MsCoreUtils::robustSummary() to calculate a robust aggregation
using MASS::rlm() (default);
base::colMeans() to use the mean of each column;
colMeansMat(x, MAT) to aggregate feature by the calculating
the mean of peptide intensities via an adjacency matrix. Shared
peptides are re-used multiple times.
matrixStats::colMedians() to use the median of each column.
base::colSums() to use the sum of each column;
colSumsMat(x, MAT) to aggregate feature by the summing the
peptide intensities for each protein via an adjacency
matrix. Shared peptides are re-used multiple times.
See MsCoreUtils::aggregate_by_vector() for more aggregation functions.
Missing quantitative values have different effects based on the aggregation method employed:
The aggregation functions should be able to deal with missing
values by either ignoring or propagating them. This is often
done with an na.rm argument, that can be passed with
.... For example, rowSums, rowMeans, rowMedians,
... will ignore NA values with na.rm = TRUE, as illustrated
below.
Missing values will result in an error when using medpolish,
unless na.rm = TRUE is used. Note that this option relies on
implicit assumptions and/or performes an implicit imputation:
when summing, the values are implicitly imputed by 0, assuming
that the NA represent a trully absent features; when
averaging, the assumption is that the NA represented a
genuinely missing value.
When using robust summarisation, individual missing values are
excluded prior to fitting the linear model by robust
regression. To remove all values in the feature containing the
missing values, use filterNA().
More generally, missing values often need dedicated handling such
as filtering (see filterNA()) or imputation (see impute()).
Missing values in the row data of an assay will also impact the
resulting (aggregated) assay row data, as illustrated in the
example below. Any feature variables (a column in the row data)
containing NA values will be dropped from the aggregated row
data. The reasons underlying this drop are detailed in the
reduceDataFrame() manual page: only invariant aggregated rows,
i.e. rows resulting from the aggregation from identical variables,
are preserved during aggregations.
The situation illustrated below should however only happen in rare
cases and should often be imputable using the value of the other
aggregation rows before aggregation to preserve the invariant
nature of that column. In cases where an NA is present in an
otherwise variant column, the column would be dropped anyway.
When considering non-unique peptides explicitly, i.e. peptides
that map to multiple proteins rather than as a protein group, it
is convenient to encode this ambiguity explicitly using a
peptide-by-proteins (sparse) adjacency matrix. This matrix is
typically stored in the rowdata and set/retrieved with the
adjacencyMatrix() function. It can be created manually (as
illustrated below) or using PSMatch::makeAdjacencyMatrix().
The QFeatures vignette provides an extended example and
the Processing vignette, for a complete quantitative
proteomics data processing pipeline. The
MsCoreUtils::aggregate_by_vector() manual page provides
further details.
## ---------------------------------------
## An example QFeatures with PSM-level data
## ---------------------------------------
data(feat1)
feat1
#> An instance of class QFeatures containing 1 assays:
#> [1] psms: SummarizedExperiment with 10 rows and 2 columns
## Aggregate PSMs into peptides
feat1 <- aggregateFeatures(feat1, "psms", "Sequence", name = "peptides")
feat1
#> An instance of class QFeatures containing 2 assays:
#> [1] psms: SummarizedExperiment with 10 rows and 2 columns
#> [2] peptides: SummarizedExperiment with 3 rows and 2 columns
## Aggregate peptides into proteins
feat1 <- aggregateFeatures(feat1, "peptides", "Protein", name = "proteins")
feat1
#> An instance of class QFeatures containing 3 assays:
#> [1] psms: SummarizedExperiment with 10 rows and 2 columns
#> [2] peptides: SummarizedExperiment with 3 rows and 2 columns
#> [3] proteins: SummarizedExperiment with 2 rows and 2 columns
assay(feat1[[1]])
#> S1 S2
#> PSM1 1 11
#> PSM2 2 12
#> PSM3 3 13
#> PSM4 4 14
#> PSM5 5 15
#> PSM6 6 16
#> PSM7 7 17
#> PSM8 8 18
#> PSM9 9 19
#> PSM10 10 20
assay(feat1[[2]])
#> S1 S2
#> ELGNDAYK 5.0 15.0
#> IAEESNFPFIK 8.5 18.5
#> SYGFNAAR 2.0 12.0
aggcounts(feat1[[2]])
#> S1 S2
#> ELGNDAYK 3 3
#> IAEESNFPFIK 4 4
#> SYGFNAAR 3 3
assay(feat1[[3]])
#> S1 S2
#> ProtA 3.5 13.5
#> ProtB 8.5 18.5
aggcounts(feat1[[3]])
#> S1 S2
#> ProtA 2 2
#> ProtB 1 1
## --------------------------------------------
## Aggregation with missing quantitative values
## --------------------------------------------
data(ft_na)
ft_na
#> An instance of class QFeatures containing 1 assays:
#> [1] na: SummarizedExperiment with 4 rows and 3 columns
assay(ft_na[[1]])
#> A B C
#> a NA 5 9
#> b 2 6 10
#> c 3 NA 11
#> d NA 8 12
rowData(ft_na[[1]])
#> DataFrame with 4 rows and 2 columns
#> X Y
#> <integer> <character>
#> a 1 A
#> b 2 B
#> c 1 A
#> d 2 B
## By default, missing values are propagated
ft2 <- aggregateFeatures(ft_na, 1, fcol = "X", fun = colSums)
#> Your quantitative data contain missing values. Please read the relevant
#> section(s) in the aggregateFeatures manual page regarding the effects
#> of missing values on data aggregation.
assay(ft2[[2]])
#> A B C
#> 1 NA NA 20
#> 2 NA 14 22
aggcounts(ft2[[2]])
#> A B C
#> 1 1 1 2
#> 2 1 2 2
## The rowData .n variable tallies number of initial rows that
## were aggregated (irrespective of NAs) for all the samples.
rowData(ft2[[2]])
#> DataFrame with 2 rows and 3 columns
#> X Y .n
#> <integer> <character> <integer>
#> 1 1 A 2
#> 2 2 B 2
## Ignored when setting na.rm = TRUE
ft3 <- aggregateFeatures(ft_na, 1, fcol = "X", fun = colSums, na.rm = TRUE)
#> Your quantitative data contain missing values. Please read the relevant
#> section(s) in the aggregateFeatures manual page regarding the effects
#> of missing values on data aggregation.
assay(ft3[[2]])
#> A B C
#> 1 3 5 20
#> 2 2 14 22
aggcounts(ft3[[2]])
#> A B C
#> 1 1 1 2
#> 2 1 2 2
## -----------------------------------------------
## Aggregation with missing values in the row data
## -----------------------------------------------
## Row data results without any NAs, which includes the
## Y variables
rowData(ft2[[2]])
#> DataFrame with 2 rows and 3 columns
#> X Y .n
#> <integer> <character> <integer>
#> 1 1 A 2
#> 2 2 B 2
## Missing value in the Y feature variable
rowData(ft_na[[1]])[1, "Y"] <- NA
rowData(ft_na[[1]])
#> DataFrame with 4 rows and 2 columns
#> X Y
#> <integer> <character>
#> a 1 NA
#> b 2 B
#> c 1 A
#> d 2 B
ft3 <- aggregateFeatures(ft_na, 1, fcol = "X", fun = colSums)
#> Your quantitative and row data contain missing values. Please read the
#> relevant section(s) in the aggregateFeatures manual page regarding the
#> effects of missing values on data aggregation.
## The Y feature variable has been dropped!
assay(ft3[[2]])
#> A B C
#> 1 NA NA 20
#> 2 NA 14 22
rowData(ft3[[2]])
#> DataFrame with 2 rows and 2 columns
#> X .n
#> <integer> <integer>
#> 1 1 2
#> 2 2 2
## --------------------------------------------
## Using a peptide-by-proteins adjacency matrix
## --------------------------------------------
## Let's use assay peptides from object feat1 and
## define that peptide SYGFNAAR maps to proteins
## Prot A and B
se <- feat1[["peptides"]]
rowData(se)$Protein[3] <- c("ProtA;ProtB")
rowData(se)
#> DataFrame with 3 rows and 4 columns
#> Sequence Protein location .n
#> <character> <character> <character> <integer>
#> ELGNDAYK ELGNDAYK ProtA Mitochondr... 3
#> IAEESNFPFIK IAEESNFPFI... ProtB unknown 4
#> SYGFNAAR SYGFNAAR ProtA;Prot... Mitochondr... 3
## This can also be defined using anadjacency matrix, manual
## encoding here. See PSMatch::makeAdjacencyMatrix() for a
## function that does it automatically.
adj <- matrix(0, nrow = 3, ncol = 2,
dimnames = list(rownames(se),
c("ProtA", "ProtB")))
adj[1, 1] <- adj[2, 2] <- adj[3, 1:2] <- 1
adj
#> ProtA ProtB
#> ELGNDAYK 1 0
#> IAEESNFPFIK 0 1
#> SYGFNAAR 1 1
adjacencyMatrix(se) <- adj
rowData(se)
#> DataFrame with 3 rows and 5 columns
#> Sequence Protein location .n adjacencyMatrix
#> <character> <character> <character> <integer> <dgCMatrix>
#> ELGNDAYK ELGNDAYK ProtA Mitochondr... 3 1:0
#> IAEESNFPFIK IAEESNFPFI... ProtB unknown 4 0:1
#> SYGFNAAR SYGFNAAR ProtA;Prot... Mitochondr... 3 1:1
adjacencyMatrix(se)
#> 3 x 2 sparse Matrix of class "dgCMatrix"
#> ProtA ProtB
#> ELGNDAYK 1 .
#> IAEESNFPFIK . 1
#> SYGFNAAR 1 1
## Aggregation using the adjacency matrix
se2 <- aggregateFeatures(se, fcol = "adjacencyMatrix",
fun = MsCoreUtils::colMeansMat)
## Peptide SYGFNAAR was taken into account in both ProtA and ProtB
## aggregations.
assay(se2)
#> S1 S2
#> ProtA 3.50 13.50
#> ProtB 5.25 15.25
## Aggregation by matrix on a QFeature object works as with a
## vector
ft <- QFeatures(list(peps = se))
ft <- aggregateFeatures(ft, "peps", "adjacencyMatrix", name = "protsByMat",
fun = MsCoreUtils::colMeansMat)
assay(ft[[2]])
#> S1 S2
#> ProtA 3.50 13.50
#> ProtB 5.25 15.25
rowData(ft[[2]])
#> DataFrame with 2 rows and 1 column
#> .n
#> <integer>
#> ProtA 2
#> ProtB 2