To query a SummarizedExperiment for interesting features, several
functions are available.
summarizeDominance(x, group = NULL, name = "dominant_taxa", ...)
getUnique(x, ...)
getTop(
x,
top = 5L,
method = c("mean", "sum", "median", "sd", "cv", "prevalence"),
assay.type = assay_name,
assay_name = "counts",
na.rm = TRUE,
...
)
# S4 method for class 'SummarizedExperiment'
getTop(
x,
top = 5L,
method = c("mean", "sum", "median", "sd", "cv", "prevalence"),
assay.type = assay_name,
assay_name = "counts",
na.rm = TRUE,
...
)
# S4 method for class 'SummarizedExperiment'
getUnique(x, rank = NULL, ...)
# S4 method for class 'SummarizedExperiment'
summarizeDominance(x, group = NULL, name = "dominant_taxa", ...)
# S4 method for class 'SummarizedExperiment'
summary(object, assay.type = assay_name, assay_name = "counts")With group, it is possible to group the observations in an
overview. Must be one of the column names of colData.
Character scalar. A name for the column of the
colData where results will be stored.
(Default: "dominant_taxa")
Additional arguments passed on to agglomerateByRank() when
rank is specified for summarizeDominance.
Numeric scalar. Determines how many top taxa to return.
Default is to return top five taxa. (Default: 5)
Character scalar. Specify the method to determine top
taxa. Either sum, mean, median, sd, cv or prevalence. With "sd" or
"cv", the features with the highest standard deviation or
coefficient of variation (standard deviation divided by mean) are returned.
Features present in only one sample have the highest possible coefficient
of variation, so rare features should be filtered out first, e.g., with
subsetByPrevalent().
(Default: "mean")
Character scalar. Specifies the name of the
assay used in calculation. (Default: "counts")
Deprecated. Use assay.type instead.
Logical scalar. Should NA values be omitted?
(Default: TRUE)
Character scalar. Defines a taxonomic rank. Must be a
value of the output of taxonomyRanks(). (Default: NULl)
A
SummarizedExperiment
object.
The getTop returns a vector of the most top abundant
“FeatureID”s
The getUnique returns a vector of unique taxa present at a
particular rank
The summarizeDominance returns an overview in a tibble. It contains
dominant taxa in a column named *name* and its abundance in the data
set.
The summary returns a list with two tibbles
The getTop extracts the most top abundant “FeatureID”s
in a
SummarizedExperiment
object.
The getUnique is a basic function to access different taxa at a
particular taxonomic rank.
summarizeDominance returns information about most dominant
taxa in a tibble. Information includes their absolute and relative
abundances in whole data set.
The summary will return a summary of counts for all samples and
features in
SummarizedExperiment
object.
data(GlobalPatterns)
top_taxa <- getTop(
GlobalPatterns,
method = "mean",
top = 5,
assay.type = "counts"
)
top_taxa
#> [1] "549656" "331820" "279599" "360229" "317182"
# Most variable taxa
getTop(GlobalPatterns, method = "sd", top = 5)
#> [1] "549656" "279599" "331820" "360229" "94166"
# Taxa with the highest coefficient of variation among prevalent taxa
tse <- subsetByPrevalent(GlobalPatterns, prevalence = 0.5)
getTop(tse, method = "cv", top = 5)
#> [1] "278873" "211720" "263681" "244304" "249661"
# Use 'detection' to select detection threshold when using prevalence method
top_taxa <- getTop(
GlobalPatterns,
method = "prevalence",
top = 5,
assay.type = "counts",
detection = 100
)
top_taxa
#> [1] "549656" "331820" "94166" "317182" "279599"
# Top taxa in specific rank
getTop(GlobalPatterns, rank = "Genus", na.rm = TRUE)
#> Duplicated labels were made unique.
#> [1] "Bacteroides" "Dolichospermum" "Faecalibacterium" "Neisseria"
#> [5] "Haemophilus"
# Gets the overview of dominant taxa
dominant_taxa <- summarizeDominance(GlobalPatterns,
rank = "Genus")
#> Duplicated labels were made unique.
dominant_taxa
#> # A tibble: 17 × 3
#> dominant_taxa n rel_freq
#> <chr> <int> <dbl>
#> 1 Bacteroides 5 0.192
#> 2 Crenothrix 3 0.115
#> 3 Faecalibacterium 2 0.0769
#> 4 Prochlorococcus 2 0.0769
#> 5 Streptococcus 2 0.0769
#> 6 CandidatusNitrososphaera 1 0.0385
#> 7 CandidatusPortiera 1 0.0385
#> 8 CandidatusSolibacter 1 0.0385
#> 9 Corynebacterium 1 0.0385
#> 10 Desulfuromonas 1 0.0385
#> 11 Dolichospermum 1 0.0385
#> 12 Luteolibacter 1 0.0385
#> 13 MC18 1 0.0385
#> 14 Neisseria 1 0.0385
#> 15 Nitrosopumilus 1 0.0385
#> 16 Polaribacter 1 0.0385
#> 17 Veillonella 1 0.0385
# With group, it is possible to group observations based on specified groups
# Gets the overview of dominant taxa
dominant_taxa <- summarizeDominance(
GlobalPatterns,
rank = "Genus",
group = "SampleType",
na.rm = TRUE
)
#> Duplicated labels were made unique.
dominant_taxa
#> # A tibble: 20 × 4
#> # Groups: SampleType [9]
#> SampleType dominant_taxa n rel_freq
#> <fct> <chr> <int> <dbl>
#> 1 Mock Bacteroides 3 1
#> 2 Feces Bacteroides 2 0.5
#> 3 Feces Faecalibacterium 2 0.5
#> 4 Freshwater (creek) Crenothrix 2 0.667
#> 5 Skin Streptococcus 2 0.667
#> 6 Freshwater Dolichospermum 1 0.5
#> 7 Freshwater Prochlorococcus 1 0.5
#> 8 Freshwater (creek) Luteolibacter 1 0.333
#> 9 Ocean CandidatusPortiera 1 0.333
#> 10 Ocean Polaribacter 1 0.333
#> 11 Ocean Prochlorococcus 1 0.333
#> 12 Sediment (estuary) Crenothrix 1 0.333
#> 13 Sediment (estuary) Desulfuromonas 1 0.333
#> 14 Sediment (estuary) Nitrosopumilus 1 0.333
#> 15 Skin Corynebacterium 1 0.333
#> 16 Soil CandidatusNitrososphaera 1 0.333
#> 17 Soil CandidatusSolibacter 1 0.333
#> 18 Soil MC18 1 0.333
#> 19 Tongue Neisseria 1 0.5
#> 20 Tongue Veillonella 1 0.5
# Get an overview of sample and taxa counts
summary(GlobalPatterns, assay.type= "counts")
#> $samples
#> # A tibble: 1 × 6
#> total_counts min_counts max_counts median_counts mean_counts stdev_counts
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 28216678 58688 2357181 1106849 1085257. 650145.
#>
#> $features
#> # A tibble: 1 × 3
#> total singletons per_sample_avg
#> <int> <int> <dbl>
#> 1 19216 2134 4022.
#>
# Get unique taxa at a particular taxonomic rank
# sort = TRUE means that output is sorted in alphabetical order
# With na.rm = TRUE, it is possible to remove NAs
# sort and na.rm can also be used in function getTop
getUnique(GlobalPatterns, "Phylum", sort = TRUE)
#> [1] "ABY1_OD1" "AC1" "AD3" "Acidobacteria"
#> [5] "Actinobacteria" "Armatimonadetes" "BRC1" "Bacteroidetes"
#> [9] "CCM11b" "Caldiserica" "Caldithrix" "Chlamydiae"
#> [13] "Chlorobi" "Chloroflexi" "Crenarchaeota" "Cyanobacteria"
#> [17] "Elusimicrobia" "Euryarchaeota" "Fibrobacteres" "Firmicutes"
#> [21] "Fusobacteria" "GAL15" "GN02" "GN04"
#> [25] "GN06" "GN12" "GOUTA4" "Gemmatimonadetes"
#> [29] "Hyd24-12" "KSB1" "LCP-89" "LD1"
#> [33] "Lentisphaerae" "MVP-15" "NC10" "NKB19"
#> [37] "Nitrospirae" "OP11" "OP3" "OP8"
#> [41] "OP9" "PAUC34f" "Planctomycetes" "Proteobacteria"
#> [45] "SAR406" "SBR1093" "SC3" "SC4"
#> [49] "SM2F11" "SPAM" "SR1" "Spirochaetes"
#> [53] "Synergistetes" "TG3" "TM6" "TM7"
#> [57] "Tenericutes" "Thermi" "Thermotogae" "Verrucomicrobia"
#> [61] "WPS-2" "WS1" "WS2" "WS3"
#> [65] "ZB2" "ZB3" NA