Package 'specmine'

Description

Aggregate samples according to an aggregate function like mean, median, etc. This can be used to merge replicates.

Usage

aggregate_samples(dataset, indexes, aggreg.fn = "mean", meta.to.remove = c())

Arguments

Value

A dataset object in which samples have been aggregated according to 'indexes' and 'aggreg.fn', with updated 'data', 'metadata', and preserved dataset-level fields such as labels, type, and description.

Examples

data <- matrix(
  c(1, 2, 10, 12,
    3, 4, 14, 16),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3", "s4"))
)
metadata <- data.frame(
  class = factor(c("A", "A", "B", "B")),
  batch = c(1, 1, 2, 2),
  row.names = c("s1", "s2", "s3", "s4")
)
dataset <- list(
  data = data,
  metadata = metadata,
  labels = list(),
  type = "example",
  description = "toy dataset"
)
aggregate_samples(dataset, c(1, 1, 2, 2), "mean")

Description

Applies adaptive iteratively reweighted penalized least squares (airPLS) baseline correction to Raman or surface-enhanced Raman spectroscopy (SERS) spectra.

Usage

airPLS_fast_dataset(dataset, lambda = 1e+05, max_iter = 50)

Arguments

Details

The dataset is expected to contain spectral variables or wavenumbers in rows and samples in columns. Each sample is baseline-corrected independently. Corrected spectra are shifted so that their minimum intensity is zero.

Value

A dataset equivalent to dataset, with baseline-corrected values stored in dataset$data. Details of the correction are stored in dataset$baseline.

References

Zhang, Z.-M., Chen, S., and Liang, Y.-Z. (2010). Baseline correction using adaptive iteratively reweighted penalized least squares. Analyst, 135(5), 1138-1146. doi:10.1039/B922045C.

Examples

set.seed(123)

dataset = list(
  data = matrix(
    rnorm(100, mean = 10, sd = 2),
    nrow = 10,
    dimnames = list(
      paste0("wavenumber", 1:10),
      paste0("sample", 1:10)
    )
  )
)

corrected = airPLS_fast_dataset(
  dataset,
  lambda = 1e4,
  max_iter = 5
)

range(corrected$data)

Description

Performs one-way ANOVA for all variables in a dataset.

Usage

aov_all_vars(
  dataset,
  column.class,
  doTukey = TRUE,
  write.file = FALSE,
  file.out = NULL
)

Arguments

Value

A data frame with ANOVA results for all variables.

Examples

datamatrix <- matrix(
  c(10, 11, 20, 21,
    5,  6,  5,  6),
  nrow = 2,
  byrow = TRUE,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3", "s4"))
)
metadata <- data.frame(
  group = factor(c("A", "A", "B", "B")),
  row.names = c("s1", "s2", "s3", "s4")
)
dataset <- list(data = datamatrix, metadata = metadata)
aov_all_vars(dataset, "group", doTukey = FALSE)

Description

Performs one-way ANOVA for a single variable in a dataset.

Usage

aov_one_var(dataset, x.val, groups, doTukey = TRUE)

Arguments

Value

A list with p-value, -log10(p), FDR, and optional Tukey results.

Examples

datamatrix <- matrix(
  c(10, 11, 20, 21,
    5,  6,  5,  6),
  nrow = 2,
  byrow = TRUE,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3", "s4"))
)
metadata <- data.frame(
  group = factor(c("A", "A", "B", "B")),
  row.names = c("s1", "s2", "s3", "s4")
)
dataset <- list(data = datamatrix, metadata = metadata)
aov_one_var(dataset, "x1", metadata$group, doTukey = FALSE)

Description

Applies a function to the variables of samples belonging to a given group.

Usage

apply_by_group(dataset, fn.to.apply, metadata.var, var.value)

Arguments

Value

A named vector or matrix, depending on the output of 'fn.to.apply', containing the values obtained by applying the function to each variable across the samples belonging to the selected group.

Examples

data <- matrix(
  c(1, 2, 3, 4, 5, 6),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3"))
)
metadata <- data.frame(
  group = c("control", "control", "case"),
  row.names = c("s1", "s2", "s3")
)
dataset <- list(data = data, metadata = metadata)
apply_by_group(dataset, mean, "group", "control")

Description

Applies a function to groups defined by a metadata variable.

Usage

apply_by_groups(
  dataset,
  metadata.var,
  fn.to.apply = "mean",
  variables = NULL,
  variable.bounds = NULL
)

Arguments

Value

A matrix-like object with one row per selected variable and one column per group defined by 'metadata.var'. Each cell contains the result of applying 'fn.to.apply' to the values of that variable within the corresponding group.

Examples

data <- matrix(
  c(1, 2, 3, 4, 5, 6),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3"))
)
metadata <- data.frame(
  group = c("control", "control", "case"),
  row.names = c("s1", "s2", "s3")
)
dataset <- list(data = data, metadata = metadata)
apply_by_groups(dataset, "group", mean)

Description

Get kegg codes from hmdb codes:

Usage

convert_hmdb_to_kegg(hmdb_codes)

Arguments

Value

A named character vector with the KEGG compound identifiers corresponding to the input HMDB codes. The vector values are KEGG compound codes prefixed with "cpd:", and the element names are the matched compound names.

Examples

## Not run: 
convert_hmdb_to_kegg(c("HMDB0000122"))

## End(Not run)

Description

Convert KEGGPathway object to graph object

Usage

convert_keggpathway_2_reactiongraph(pathObj)

Arguments

Value

A graph object representing the reactions in the input KEGGPathway object. The returned object is suitable for graph-based inspection or for use in downstream visualization functions.

Examples

## Not run: 
path <- get_MetabolitePath("hsa00010")
graph <- convert_keggpathway_2_reactiongraph(path)
class(graph)

## End(Not run)

Description

Get kegg codes from spcmnm codes:

Usage

convert_multiple_spcmnm_to_kegg(spcmnm_codes)

Arguments

Value

A named character vector with the KEGG compound identifiers corresponding to the input SPCMNM codes. The vector values are KEGG compound codes prefixed with "cpd:", and the element names are the matched compound names.

Examples

## Not run: 
convert_multiple_spcmnm_to_kegg(c("SPCM00001"))

## End(Not run)

Description

Returns the total number of missing values in the dataset.

Usage

count_missing_values(dataset)

Arguments

Value

A single numeric value giving the total count of 'NA' entries present in 'dataset$data'.

Examples

data <- matrix(
  c(1, NA, 3, 4),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2"))
)
metadata <- data.frame(class = c("A", "B"), row.names = c("s1", "s2"))
dataset <- list(data = data, metadata = metadata)
count_missing_values(dataset)

Description

Returns the number of missing values per sample.

Usage

count_missing_values_per_sample(dataset, remove.zero = TRUE)

Arguments

Value

A named numeric vector giving the number of missing values for each sample in 'dataset$data'. If 'remove.zero' is 'TRUE', only samples with at least one missing value are returned.

Examples

data <- matrix(
  c(1, NA, 3,
    4,  5, NA),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3"))
)
metadata <- data.frame(class = c("A", "B", "A"), row.names = c("s1", "s2", "s3"))
dataset <- list(data = data, metadata = metadata)
count_missing_values_per_sample(dataset)

Description

Returns the number of missing values per variable.

Usage

count_missing_values_per_variable(dataset, remove.zero = TRUE)

Arguments

Value

A named numeric vector giving the number of missing values for each variable in 'dataset$data'. If 'remove.zero' is 'TRUE', only variables with at least one missing value are returned.

Examples

data <- matrix(
  c(1, 2, NA,
    4, 5, NA),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3"))
)
metadata <- data.frame(class = c("A", "B", "A"), row.names = c("s1", "s2", "s3"))
dataset <- list(data = data, metadata = metadata)
count_missing_values_per_variable(dataset)

Description

Creates a dataset from a numeric matrix.

Usage

create_dataset(
  datamatrix,
  type = "undefined",
  metadata = NULL,
  description = "",
  sample.names = NULL,
  x.axis.values = NULL,
  label.x = NULL,
  label.values = NULL,
  xSet = NULL
)

Arguments

Value

A list representing a specmine dataset. The returned object contains at least the elements data, a numeric matrix with variables in rows and samples in columns; type, a character string identifying the dataset type; description, a character string describing the dataset; metadata, a data frame with one row per sample when available; labels, a list with axis and value labels when provided; and xSet, an optional object associated with LC-MS processing. This object is the standard input structure used by downstream specmine analysis functions.

Examples

datamatrix <- matrix(
  c(1.1, 2.2, 3.3, 4.4, 5.5, 6.6),
  nrow = 2,
  dimnames = list(NULL, NULL)
)
metadata <- data.frame(
  class = c("A", "B", "A"),
  row.names = c("s1", "s2", "s3")
)
create_dataset(
  datamatrix,
  type = "concentrations",
  metadata = metadata,
  sample.names = c("s1", "s2", "s3"),
  x.axis.values = c("v1", "v2"),
  label.x = "variable",
  label.values = "intensity"
)

Description

Creates the pathway, with reactions included in the nodes

Usage

create_pathway_with_reactions(
  path,
  path.name,
  identified_cpds,
  nodeNames = "kegg",
  nodeTooltip = FALSE,
  map.zoom = FALSE,
  map.layout = "preset",
  map.width = NULL,
  map.height = NULL
)

Arguments

Value

A cyjShiny widget representing the pathway with reactions included in the nodes. The widget contains the pathway graph ready for interactive visualization, with identified compounds highlighted when they are present in the pathway.

Examples

## Not run: 
path <- get_MetabolitePath("hsa00010")
create_pathway_with_reactions(
  path = path,
  path.name = "hsa00010",
  identified_cpds = c("cpd:C00031", "cpd:C00022")
)

## End(Not run)

Description

Merges equivalent peaks across samples, extracts intensities, and creates a dataset object in standard package format.

Usage

dataset_from_peaks(
  sample.list,
  metadata = NULL,
  description = "",
  type = "nmr-peaks"
)

Arguments

Value

A dataset object created from the input peak lists.

Examples

sample.list <- list(
  s1 = data.frame(ppm = c(1.0, 2.0), int = c(10, 20)),
  s2 = data.frame(ppm = c(1.0, 3.0), int = c(15, 30))
)
metadata <- data.frame(class = c("A", "B"), row.names = c("s1", "s2"))
dataset <- dataset_from_peaks(sample.list, metadata = metadata)
class(dataset)

Description

Performs feature selection using univariate filters.

Usage

filter_feature_selection(
  datamat,
  samples.class,
  functions = caret::rfSBF,
  method = "cv",
  repeats = 5
)

Arguments

Value

An sbf object.

Examples

datamat <- matrix(
  rnorm(20),
  nrow = 4,
  dimnames = list(paste0("x", 1:4), paste0("s", 1:5))
)
classes <- factor(c("A", "A", "B", "B", "A"))
filter_feature_selection(datamat, classes, caret::rfSBF)

Description

Performs a flat pattern filter over the dataset.

Usage

flat_pattern_filter(
  dataset,
  filter.function = "iqr",
  by.percent = TRUE,
  by.threshold = FALSE,
  red.value = 0
)

Arguments

Value

A dataset object with the same overall structure as the input, where 'dataset$data' has been filtered according to the selected flat-pattern criterion. Variables failing the selection rule are removed, the remaining dataset components are preserved, and 'dataset$description' is updated to record the filtering step.

Examples

if (requireNamespace("specmine.datasets", quietly = TRUE)) {
  data(propolis, package = "specmine.datasets")
  propolis_proc = missingvalues_imputation(propolis)
  propolis_proc = flat_pattern_filter(propolis_proc, "iqr",
    by.percent = TRUE, red.value = 75)
}

Description

Returns compound names associated with KEGG compound identifiers.

Usage

get_cpd_names(kegg_codes)

Arguments

Value

A named character vector in which the values are KEGG compound identifiers and the names are the corresponding compound names. Each element of the returned vector represents one KEGG compound code matched to its primary compound name.

Examples

## Not run: 
get_cpd_names(c("C00031", "C00022"))

## End(Not run)

Description

Downloads files and metadata for every assay in a public MetaboLights study.

Usage

get_metabolights_study(studyID, directory, verbose = TRUE)

Arguments

Value

Invisibly returns a list with one entry per assay containing downloaded files, metadata, and sample-file mappings.

Examples

## Not run: 
tmpdir <- tempdir()
get_metabolights_study("MTBLS1", directory = tmpdir, verbose = FALSE)

## End(Not run)

Description

Downloads the data files associated with one assay of a MetaboLights study.

Usage

get_metabolights_study_files_assay(studyID, assay, directory, verbose = TRUE)

Arguments

Value

Invisibly returns a character vector with downloaded file paths.

Examples

## Not run: 
tmpdir <- tempdir()
get_metabolights_study_files_assay("MTBLS1", assay = 1, directory = tmpdir, verbose = FALSE)

## End(Not run)

Description

Returns the metadata associated with the samples in one assay of a MetaboLights study and can optionally save it as a CSV file.

Usage

get_metabolights_study_metadata_assay(studyID, assay, directory = NULL)

Arguments

Value

Invisibly returns a data.frame with sample metadata for the selected assay.

Examples

## Not run: 
md <- get_metabolights_study_metadata_assay("MTBLS1", assay = 1)
head(md)

## End(Not run)

Description

Returns the sample-to-file mapping for one assay in a MetaboLights study and can optionally save it as a CSV file.

Usage

get_metabolights_study_samples_files(studyID, assay, directory = NULL)

Arguments

Value

Invisibly returns a two-column object with samples and corresponding files.

Examples

## Not run: 
get_metabolights_study_samples_files("MTBLS1", assay = 1)

## End(Not run)

Description

Returns an object of KEGGPathway of the pathway especified in pathcode

Usage

get_MetabolitePath(pathcode)

Arguments

Value

A KEGGPathway object corresponding to the pathway specified by pathcode. This object contains the parsed KEGG pathway structure and can be used as input to downstream graph conversion and visualization functions.

Examples

## Not run: 
path <- get_MetabolitePath("hsa00010")
class(path)

## End(Not run)

Description

Get vector with paths numbers that occur in the given organism, named with the full path name:

Usage

get_metabPaths_org(org_code)

Arguments

Value

A named character vector with the pathway identifiers available for the specified organism, where each value is an organism-specific pathway code and each name is the corresponding KEGG pathway name.

Examples

## Not run: 
head(get_metabPaths_org("hsa"))

## End(Not run)

Description

Get code, t number, full name and phylogeny of all organisms in KEGG:

Usage

get_OrganismsCodes()

Value

A data.frame with the KEGG T number, organism code, species names, and phylogeny for all organisms available in KEGG. Each row represents one organism and the columns are Tnumber, organismCode, speciesNames, and phylogeny.

Examples

## Not run: 
head(get_OrganismsCodes())

## End(Not run)

Description

Get only the paths of the organism that contain given compounds:

Usage

get_paths_with_cpds_org(organism_code, compounds, full.result = TRUE)

Arguments

Value

A data.frame with the pathways containing the input compounds. Each row represents one matched pathway. When full.result = TRUE, the returned data frame includes the columns pathways, ratio, compounds, and compounds_names; otherwise it contains only pathways and ratio. Row names correspond to pathway names.

Examples

## Not run: 
cpds <- c(glucose = "cpd:C00031", pyruvate = "cpd:C00022")
head(get_paths_with_cpds_org("hsa", cpds, full.result = FALSE))

## End(Not run)

Description

Returns the x-axis label associated with the dataset.

Usage

get_x_label(dataset)

Arguments

Value

A character string giving the label of the x axis stored in 'dataset$labels$x'. If no x-axis label has been defined, the function returns an empty string.

Examples

datamatrix <- matrix(
  c(1, 2, 3, 4),
  nrow = 2,
  dimnames = list(c("10.5", "11.0"), c("s1", "s2"))
)
metadata <- data.frame(class = c("A", "B"), row.names = c("s1", "s2"))
dataset <- create_dataset(
  datamatrix,
  type = "concentrations",
  metadata = metadata,
  label.x = "ppm",
  label.values = "intensity"
)
get_x_label(dataset)

Description

Returns the x values of a dataset as text.

Usage

get_x_values_as_text(dataset)

Arguments

Value

A character vector containing the variable identifiers stored in 'rownames(dataset$data)'. Each element corresponds to one row of the data matrix and represents the x-axis value or variable label associated with that feature.

Examples

datamatrix <- matrix(
  c(1, 2, 3, 4),
  nrow = 2,
  dimnames = list(c("10.5", "11.0"), c("s1", "s2"))
)
metadata <- data.frame(class = c("A", "B"), row.names = c("s1", "s2"))
dataset <- create_dataset(
  datamatrix,
  type = "concentrations",
  metadata = metadata
)
get_x_values_as_text(dataset)

Description

Performs Independent Component Analysis (ICA) on a specmine dataset using the FastICA algorithm.

Usage

ica_analysis_dataset(
  dataset,
  n_components = 2,
  alg.typ = "parallel",
  fun = "logcosh",
  maxit = 200,
  tol = 1e-04,
  scale = FALSE,
  seed = 42,
  write.file = FALSE,
  file.out = NULL,
  ...
)

Arguments

Value

A list containing:

• embedding: the independent component scores;

• S: the estimated source matrix;

• A: the estimated mixing matrix;

• K: the whitening matrix;

• W: the estimated unmixing matrix;

• loadings: the component loadings;

• params: the analysis parameters.

Examples

if (requireNamespace("fastICA", quietly = TRUE)) {
  datamat <- matrix(
    rnorm(240),
    nrow = 8,
    dimnames = list(
      paste0("feature", 1:8),
      paste0("sample", 1:30)
    )
  )

  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 15))
    )
  )

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 2,
    maxit = 100,
    seed = 42
  )

  dim(ica.result$embedding)
}

Description

Creates a two-dimensional ICA plot coloured by k-means clusters.

Usage

ica_kmeans_plot2D(
  dataset,
  ica.result,
  num.clusters = 3,
  dims = c(1, 2),
  kmeans.result = NULL,
  use.embedding = TRUE,
  labels = FALSE,
  bw = FALSE,
  ellipses = FALSE,
  leg.pos = "right",
  xlim = NULL,
  ylim = NULL
)

Arguments

Value

A ggplot object.

Examples

if (requireNamespace("fastICA", quietly = TRUE)) {
  datamat <- matrix(rnorm(240), nrow = 8)
  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 15))
    )
  )

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 2,
    maxit = 100,
    seed = 42
  )

  kmeans.result <- list(
    cluster = rep(1:2, each = 15)
  )

  ica_kmeans_plot2D(
    dataset,
    ica.result,
    num.clusters = 2,
    kmeans.result = kmeans.result
  )
}

Description

Creates an interactive three-dimensional ICA plot coloured by k-means clusters.

Usage

ica_kmeans_plot3D(
  dataset,
  ica.result,
  num.clusters = 3,
  dims = c(1, 2, 3),
  kmeans.result = NULL,
  use.embedding = TRUE,
  title = "ICA 3D K-means Plot"
)

Arguments

Value

A plotly htmlwidget.

Examples

if (requireNamespace("fastICA", quietly = TRUE) &&
    requireNamespace("plotly", quietly = TRUE)) {
  datamat <- matrix(rnorm(240), nrow = 8)
  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 15))
    )
  )

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 3,
    maxit = 100,
    seed = 42
  )

  kmeans.result <- list(
    cluster = rep(1:2, each = 15)
  )

  ica_kmeans_plot3D(
    dataset,
    ica.result,
    num.clusters = 2,
    kmeans.result = kmeans.result
  )
}

Description

Creates a two-dimensional plot of the loadings associated with two independent components.

Usage

ica_loadingsplot(ica.result, dims = c(1, 2), top.n = NULL, labels = FALSE)

Arguments

Value

A ggplot object.

Examples

if (requireNamespace("fastICA", quietly = TRUE)) {
  datamat <- matrix(
    rnorm(240),
    nrow = 8,
    dimnames = list(
      paste0("feature", 1:8),
      paste0("sample", 1:30)
    )
  )

  dataset <- list(data = datamat)

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 2,
    maxit = 100,
    seed = 42
  )

  ica_loadingsplot(
    ica.result,
    dims = c(1, 2),
    labels = TRUE
  )
}

Description

Creates a pairs plot for independent components coloured by k-means clusters.

Usage

ica_pairs_kmeans_plot(
  dataset,
  ica.result,
  num.clusters = 3,
  kmeans.result = NULL,
  use.embedding = TRUE,
  dims = 1:min(5, ncol(ica.result$embedding))
)

Arguments

Value

A GGally pairs plot object.

Examples

if (requireNamespace("fastICA", quietly = TRUE) &&
    requireNamespace("GGally", quietly = TRUE)) {
  datamat <- matrix(rnorm(240), nrow = 8)
  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 15))
    )
  )

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 3,
    maxit = 100,
    seed = 42
  )

  kmeans.result <- list(
    cluster = rep(1:2, each = 15)
  )

  ica_pairs_kmeans_plot(
    dataset,
    ica.result,
    num.clusters = 2,
    kmeans.result = kmeans.result,
    dims = 1:3
  )
}

Description

Creates a pairs plot for selected independent components.

Usage

ica_pairs_plot(
  dataset,
  ica.result,
  column.class = NULL,
  dims = 1:min(5, ncol(ica.result$embedding)),
  ...
)

Arguments

Value

A GGally pairs plot object.

Examples

if (requireNamespace("fastICA", quietly = TRUE) &&
    requireNamespace("GGally", quietly = TRUE)) {
  datamat <- matrix(rnorm(240), nrow = 8)
  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 15))
    )
  )

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 3,
    maxit = 100,
    seed = 42
  )

  ica_pairs_plot(
    dataset,
    ica.result,
    column.class = "class",
    dims = 1:3
  )
}

Description

Creates a two-dimensional scores plot from an ICA result.

Usage

ica_scoresplot2D(
  dataset,
  ica.result,
  column.class = NULL,
  dims = c(1, 2),
  labels = FALSE,
  ellipses = FALSE,
  bw = FALSE,
  palette = 2,
  leg.pos = "right",
  xlim = NULL,
  ylim = NULL
)

Arguments

Value

A ggplot object.

Examples

if (requireNamespace("fastICA", quietly = TRUE)) {
  datamat <- matrix(rnorm(240), nrow = 8)
  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 15))
    )
  )

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 2,
    maxit = 100,
    seed = 42
  )

  ica_scoresplot2D(
    dataset,
    ica.result,
    column.class = "class"
  )
}

Description

Creates an interactive three-dimensional scores plot from an ICA result.

Usage

ica_scoresplot3D(
  dataset,
  ica.result,
  column.class = NULL,
  dims = c(1, 2, 3),
  title = "ICA 3D Scores Plot"
)

Arguments

Value

A plotly htmlwidget.

Examples

if (requireNamespace("fastICA", quietly = TRUE) &&
    requireNamespace("plotly", quietly = TRUE)) {
  datamat <- matrix(rnorm(240), nrow = 8)
  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 15))
    )
  )

  ica.result <- ica_analysis_dataset(
    dataset,
    n_components = 3,
    maxit = 100,
    seed = 42
  )

  ica_scoresplot3D(
    dataset,
    ica.result,
    column.class = "class"
  )
}

Description

Replace missing values using k-nearest neighbors imputation.

Usage

impute_nas_knn(dataset, k = 10, ...)

Arguments

Value

A dataset object with the same structure as the input, where missing values in 'dataset$data' have been imputed using the k-nearest neighbors method implemented in 'impute::impute.knn()'. The returned object preserves the remaining dataset components unchanged.

Examples

data <- matrix(
  c(1,  2, NA, 4,
    2,  3,  4, 5,
    3, NA,  5, 6,
    4,  5,  6, 7),
  nrow = 4,
  byrow = TRUE,
  dimnames = list(c("x1", "x2", "x3", "x4"), c("s1", "s2", "s3", "s4"))
)
dataset <- list(data = data)
impute_nas_knn(dataset, k = 2)

Description

Replace missing values in each variable by the mean of the observed values for that variable.

Usage

impute_nas_mean(dataset)

Arguments

Value

A dataset object with the same structure as the input, where missing values in 'dataset$data' have been replaced by the mean of the corresponding variable calculated with 'na.rm = TRUE'.

Examples

data <- matrix(
  c(1, NA, 3, 4),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2"))
)
dataset <- list(data = data)
impute_nas_mean(dataset)

Description

Replace missing values in each variable by the median of the observed values for that variable.

Usage

impute_nas_median(dataset)

Arguments

Value

A dataset object with the same structure as the input, where missing values in 'dataset$data' have been replaced by the median of the corresponding variable calculated with 'na.rm = TRUE'.

Examples

data <- matrix(
  c(1, NA, 5, 4),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2"))
)
dataset <- list(data = data)
impute_nas_median(dataset)

Description

Replace all 'NA' values in 'dataset$data' by a user-defined constant.

Usage

impute_nas_value(dataset, value)

Arguments

Value

A dataset object with the same structure as the input, where all missing values in 'dataset$data' have been replaced by 'value'. Other components of the dataset are preserved unchanged.

Examples

data <- matrix(
  c(1, NA, 3, 4),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2"))
)
dataset <- list(data = data)
impute_nas_value(dataset, 0)

Description

Merges data and metadata into a data frame.

Usage

merge_data_metadata(
  dataset,
  samples = NULL,
  metadata.vars = NULL,
  x.values = NULL,
  by.index = FALSE
)

Arguments

Value

A 'data.frame' containing the selected data matrix, transposed so that samples are rows, combined with the selected metadata columns.

Examples

data <- matrix(
  1:6,
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3"))
)
metadata <- data.frame(
  class = c("A", "B", "A"),
  batch = c(1, 1, 2),
  row.names = c("s1", "s2", "s3")
)
dataset <- list(data = data, metadata = metadata)
merge_data_metadata(dataset, metadata.vars = "class")

Description

Returns the identifiers of public studies available in MetaboLights.

Usage

metabolights_studies_list()

Value

A character vector with public MetaboLights study identifiers.

Examples

## Not run: 
head(metabolights_studies_list())

## End(Not run)

Description

Impute missing values in a dataset using different methods.

Usage

missingvalues_imputation(dataset, method = "value", value = 5e-04, k = 5)

Arguments

Value

A dataset object with the same overall structure as the input, in which missing values in 'dataset$data' have been imputed according to the selected method. The returned object preserves the dataset components and updates the description to record the imputation step.

Examples

data <- matrix(
  c(1, NA, 3, 4, 5, NA),
  nrow = 2,
  dimnames = list(c("x1", "x2"), c("s1", "s2", "s3"))
)
dataset <- list(data = data, description = "toy dataset")
missingvalues_imputation(dataset, method = "value", value = 0)

Description

Compute overall and class-averaged performance metrics for multiclass classification models, including ROC AUC and log-loss when class probabilities are available.

Usage

multiClassSummary(data, lev = NULL, model = NULL)

Arguments

Value

A named numeric vector with overall and class-averaged classification statistics.

Examples

data <- data.frame(
  pred = factor(c("A", "B", "A", "B"), levels = c("A", "B")),
  obs = factor(c("A", "B", "B", "B"), levels = c("A", "B")),
  A = c(0.8, 0.2, 0.7, 0.3),
  B = c(0.2, 0.8, 0.3, 0.7)
)
multiClassSummary(data)

Description

Creates the pathway wanted. If any of the given compounds is present in the pathway, it is coloured differently.

Usage

pathway_analysis(
  compounds,
  pathway,
  nodeNames = "kegg",
  nodeTooltip = FALSE,
  map.zoom = FALSE,
  map.layout = "preset",
  map.width = NULL,
  map.height = NULL,
  verbose = TRUE
)

Arguments

Value

A cyjShiny widget representing the selected pathway with the input compounds highlighted when present. The returned widget can be printed or embedded in an interactive R session to inspect the pathway graph.

Examples

## Not run: 
pathway_analysis(
  compounds = c(glucose = "cpd:C00031", pyruvate = "cpd:C00022"),
  pathway = "hsa00010",
  verbose = FALSE
)

## End(Not run)

Description

Performs classical Principal Component Analysis (PCA) on a specmine dataset using stats::prcomp().

Usage

pca_analysis_dataset(
  dataset,
  scale = TRUE,
  center = TRUE,
  write.file = FALSE,
  file.out = NULL,
  ...
)

Arguments

Details

The dataset is expected to contain variables in rows and samples in columns. The data matrix is transposed internally so that samples are treated as observations and variables as features.

Value

An object of class prcomp. The returned object contains:

• x: a matrix of PCA scores, with one row per sample;

• rotation: a matrix of PCA loadings, with one row per variable;

• sdev: the standard deviations of the principal components;

• center: the centring values used by the PCA;

• scale: the scaling values used by the PCA when scale = TRUE;

• call: the matched function call.

The row names of x correspond to sample names when the columns of dataset$data are named. The row names of rotation correspond to variable names when the rows of dataset$data are named.

Examples

datamat <- matrix(
  rnorm(40),
  nrow = 4,
  ncol = 10,
  dimnames = list(
    paste0("feature", 1:4),
    paste0("sample", 1:10)
  )
)

dataset <- list(data = datamat)

pca.result <- pca_analysis_dataset(
  dataset,
  scale = TRUE,
  center = TRUE
)

dim(pca.result$x)
dim(pca.result$rotation)


file.prefix <- file.path(tempdir(), "pca_example")

pca_analysis_dataset(
  dataset,
  write.file = TRUE,
  file.out = file.prefix
)

file.exists(paste0(file.prefix, "_scores.csv"))
file.exists(paste0(file.prefix, "_loadings.csv"))

Description

Draws a PCA biplot for a prcomp or princomp result.

Usage

pca_biplot(
  dataset,
  pca.result,
  cex = 0.8,
  legend.cex = 0.8,
  x.colors = 1,
  inset = c(0, 0),
  legend.place = "topright",
  ...
)

Arguments

Value

No return value, called for its side effect of plotting.

Examples

datamat <- matrix(
  rnorm(20),
  nrow = 4,
  dimnames = list(paste0("x", 1:4), paste0("s", 1:5))
)
metadata <- data.frame(class = factor(c("A", "A", "B", "B", "A")))
dataset <- list(data = datamat, metadata = metadata)
pca_res <- pca_analysis_dataset(dataset)
pca_biplot(dataset, pca_res)

Description

Creates an interactive 3D PCA biplot using rgl.

Usage

pca_biplot3D(dataset, pca.result, column.class = NULL, pcas = c(1, 2, 3))

Arguments

Value

No return value, called for its side effect of plotting.

Examples

datamat <- matrix(
  rnorm(20),
  nrow = 4,
  dimnames = list(paste0("x", 1:4), paste0("s", 1:5))
)
metadata <- data.frame(class = factor(c("A", "A", "B", "B", "A")))
dataset <- list(data = datamat, metadata = metadata)
pca_res <- pca_analysis_dataset(dataset)
pca_biplot3D(dataset, pca_res, column.class = "class")

Description

Calculates the standard deviation, proportion of variance and cumulative proportion of variance explained by selected principal components.

Usage

pca_importance(
  pca.res,
  pcs = 1:length(pca.res$sdev),
  sd = TRUE,
  prop = TRUE,
  cumul = TRUE,
  min.cum = NULL
)

Arguments

Details

The function accepts PCA results generated by pca_analysis_dataset(), stats::prcomp() or stats::princomp().

Value

A numeric matrix with one row for each requested measure: Standard deviation, Proportion of Variance and/or Cumulative Proportion. Columns correspond to the selected principal components.

For a prcomp object, the values are calculated from its sdev component. For a princomp object, the same calculation is applied to its sdev component.

Examples

datamat <- matrix(
  rnorm(40),
  nrow = 4,
  ncol = 10,
  dimnames = list(
    paste0("feature", 1:4),
    paste0("sample", 1:10)
  )
)

dataset <- list(data = datamat)
pca.result <- pca_analysis_dataset(dataset)

pca_importance(
  pca.result,
  pcs = 1:3,
  sd = TRUE,
  prop = TRUE,
  cumul = TRUE
)

pca_importance(
  pca.result,
  min.cum = 0.90,
  sd = FALSE,
  prop = TRUE,
  cumul = TRUE
)

Description

Plots PCA scores in 2D coloured by k-means cluster assignment.

Usage

pca_kmeans_plot2D(
  dataset,
  pca.result,
  num.clusters = 3,
  pcas = c(1, 2),
  kmeans.result = NULL,
  labels = FALSE,
  bw = FALSE,
  ellipses = FALSE,
  leg.pos = "right",
  xlim = NULL,
  ylim = NULL
)

Arguments

Value

A ggplot object.

Examples

datamat <- matrix(
  rnorm(20),
  nrow = 4,
  dimnames = list(paste0("x", 1:4), paste0("s", 1:5))
)
metadata <- data.frame(class = factor(c("A", "A", "B", "B", "A")))
dataset <- list(data = datamat, metadata = metadata)
pca_res <- pca_analysis_dataset(dataset)
pca_kmeans_plot2D(dataset, pca_res, num.clusters = 2)

Description

Plots PCA scores in 3D coloured by k-means cluster assignment.

Usage

pca_kmeans_plot3D(
  dataset,
  pca.result,
  num.clusters = 3,
  pcas = c(1, 2, 3),
  kmeans.result = NULL,
  labels = FALSE,
  size = 1,
  ...
)

Arguments

Value

No return value, called for its side effect of plotting.

Examples

datamat <- matrix(
  rnorm(20),
  nrow = 4,
  dimnames = list(paste0("x", 1:4), paste0("s", 1:5))
)
metadata <- data.frame(class = factor(c("A", "A", "B", "B", "A")))
dataset <- list(data = datamat, metadata = metadata)
pca_res <- pca_analysis_dataset(dataset)
pca_kmeans_plot3D(dataset, pca_res, num.clusters = 2)

Description

Creates a pairs plot of PCA scores coloured by k-means cluster assignment.

Usage

pca_pairs_kmeans_plot(
  dataset,
  pca.result,
  num.clusters = 3,
  kmeans.result = NULL,
  pcas = c(1, 2, 3, 4, 5)
)

Arguments

Value

A ggmatrix object.

Examples

datamat <- matrix(
  rnorm(30),
  nrow = 6,
  dimnames = list(paste0("x", 1:6), paste0("s", 1:5))
)
metadata <- data.frame(class = factor(c("A", "A", "B", "B", "A")))
dataset <- list(data = datamat, metadata = metadata)
pca_res <- pca_analysis_dataset(dataset)
pca_pairs_kmeans_plot(dataset, pca_res, num.clusters = 2, pcas = 1:3)

Description

Creates a pairs plot of PCA scores.

Usage

pca_pairs_plot(
  dataset,
  pca.result,
  column.class = NULL,
  pcas = c(1, 2, 3, 4, 5),
  ...
)

Arguments

Value

A ggmatrix object.

Examples

datamat <- matrix(
  rnorm(30),
  nrow = 6,
  dimnames = list(paste0("x", 1:6), paste0("s", 1:5))
)
metadata <- data.frame(class = factor(c("A", "A", "B", "B", "A")))
dataset <- list(data = datamat, metadata = metadata)
pca_res <- pca_analysis_dataset(dataset)
pca_pairs_plot(dataset, pca_res, column.class = "class", pcas = 1:3)

Description

Performs robust Principal Component Analysis using pcaPP::PCAgrid().

Usage

pca_robust(
  dataset,
  center = "median",
  scale = "mad",
  k = 10,
  write.file = FALSE,
  file.out = NULL,
  ...
)

Arguments

Details

The dataset is expected to contain variables in rows and samples in columns. The data matrix is transposed internally so that samples are treated as observations and variables as features.

Value

The object returned by pcaPP::PCAgrid(). Depending on the version of pcaPP, the result contains robust PCA scores, loadings, component deviations and the centring and scaling information used in the analysis. When scores are available, they can be accessed with pca.result$scores; loadings can be accessed with pca.result$loadings.

Examples

if (requireNamespace("pcaPP", quietly = TRUE)) {
  datamat <- matrix(
    rnorm(40),
    nrow = 4,
    ncol = 10,
    dimnames = list(
      paste0("feature", 1:4),
      paste0("sample", 1:10)
    )
  )

  dataset <- list(data = datamat)

  robust.result <- pca_robust(
    dataset,
    center = "median",
    scale = "mad",
    k = 2
  )

  is.list(robust.result)
}


if (requireNamespace("pcaPP", quietly = TRUE)) {
  datamat <- matrix(
    rnorm(40),
    nrow = 4,
    ncol = 10,
    dimnames = list(
      paste0("feature", 1:4),
      paste0("sample", 1:10)
    )
  )

  dataset <- list(data = datamat)
  file.prefix <- file.path(tempdir(), "robust_pca_example")

  pca_robust(
    dataset,
    k = 2,
    write.file = TRUE,
    file.out = file.prefix
  )

  file.exists(paste0(file.prefix, "_scores.csv"))
  file.exists(paste0(file.prefix, "_loadings.csv"))
}

Description

Creates a two-dimensional PCA scores plot.

Usage

pca_scoresplot2D(
  dataset,
  pca.result,
  column.class = NULL,
  pcas = c(1, 2),
  labels = FALSE,
  ellipses = FALSE,
  bw = FALSE,
  pallette = 2,
  leg.pos = "right",
  xlim = NULL,
  ylim = NULL
)

Arguments

Details

The function accepts results from pca_analysis_dataset(), stats::prcomp() or stats::princomp().

Value

A ggplot object containing the PCA scores plot. The plotted points represent samples, and their coordinates are the selected PCA scores. When column.class is supplied, point colours or shapes represent the corresponding metadata groups.

Examples

datamat <- matrix(
  rnorm(40),
  nrow = 4,
  ncol = 10,
  dimnames = list(
    paste0("feature", 1:4),
    paste0("sample", 1:10)
  )
)

dataset <- list(
  data = datamat,
  metadata = data.frame(
    class = factor(rep(c("A", "B"), each = 5))
  )
)

pca.result <- pca_analysis_dataset(dataset)

pca_scoresplot2D(
  dataset,
  pca.result,
  column.class = "class",
  pcas = c(1, 2),
  labels = TRUE
)

Description

Creates an interactive three-dimensional scores plot from a PCA result.

Usage

pca_scoresplot3D(
  dataset,
  pca.result,
  column.class = NULL,
  pcas = c(1, 2, 3),
  title = "PCA 3D Scores Plot"
)

Arguments

Value

A plotly htmlwidget.

Examples

if (requireNamespace("plotly", quietly = TRUE)) {
  datamat <- matrix(
    rnorm(40),
    nrow = 5,
    ncol = 8
  )

  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), length.out = 8))
    )
  )

  result <- pca_analysis_dataset(dataset)

  pca_scoresplot3D(
    dataset,
    result,
    column.class = "class",
    pcas = c(1, 2, 3)
  )
}

Description

Creates an interactive three-dimensional PCA scores plot using rgl.

Usage

pca_scoresplot3D_rgl(
  dataset,
  pca.result,
  column.class = NULL,
  pcas = c(1, 2, 3),
  size = 1,
  labels = FALSE
)

Arguments

Details

Samples are represented as points in the selected principal component space. When column.class is supplied, point colours represent the corresponding metadata groups.

Value

Invisibly returns the object produced by rgl::plot3d(); the three-dimensional plot is also drawn as a side effect.

Examples

if (requireNamespace("rgl", quietly = TRUE)) {
  options(rgl.useNULL = TRUE)

  datamat <- matrix(
    rnorm(40),
    nrow = 4,
    ncol = 10,
    dimnames = list(
      paste0("feature", 1:4),
      paste0("sample", 1:10)
    )
  )

  dataset <- list(
    data = datamat,
    metadata = data.frame(
      class = factor(rep(c("A", "B"), each = 5))
    )
  )

  pca.result <- pca_analysis_dataset(dataset)

  pca_scoresplot3D_rgl(
    dataset,
    pca.result,
    column.class = "class",
    pcas = c(1, 2, 3),
    labels = TRUE
  )
}

Description

Draws a scree plot showing the individual and cumulative percentage of variance explained by the principal components.

Usage

pca_screeplot(
  pca.result,
  num.pcs = NULL,
  cex.leg = 0.8,
  leg.pos = "right",
  lab.text = c("individual percent", "cumulative percent"),
  fill.col = c("blue", "red"),
  ylab = "Percentage",
  xlab = "Principal components",
  ...
)

Arguments

Value

No return value; the scree plot is drawn as a side effect.

Examples

datamat <- matrix(
  rnorm(40),
  nrow = 4,
  ncol = 10,
  dimnames = list(
    paste0("feature", 1:4),
    paste0("sample", 1:10)
  )
)

dataset <- list(data = datamat)
pca.result <- pca_analysis_dataset(dataset)

pca_screeplot(
  pca.result,
  num.pcs = 4
)

Description

Detects local maxima in each spectrum of a 2D NMR dataset and converts the detected peaks into a 1D specmine dataset suitable for downstream analysis.

Usage

peak_detection2d(
  specmine_2d_dataset,
  baseline_thresh = NULL,
  noiseFilt = 0,
  negatives = FALSE,
  verbose = TRUE
)

Arguments

Value

A specmine dataset object as a list. The data element is a numeric matrix where rows correspond to detected 2D peak coordinates encoded as combined F1/F2 positions and columns correspond to samples. Matrix entries contain peak intensities, with missing values indicating that a given peak was not detected in a sample. The returned object also includes the original sample metadata, the dataset description, the dataset type set to "nmr-peaks", and labels describing the x-axis and data values. This output represents a reduced 1D peak table derived from the input 2D NMR spectra.

Examples

m <- matrix(c(0, 5, 0, 0), nrow = 2)
rownames(m) <- c("1.0", "1.1")
colnames(m) <- c("2.0", "2.1")
x <- list(data = list(s1 = m), metadata = data.frame(), description = "Toy")
try(peak_detection2d(x, baseline_thresh = 1, negatives = TRUE, verbose = FALSE))

Description

Groups peaks detected in multiple spectra into wavenumber bins defined by a tolerance. The resulting feature matrix can be used in downstream analyses such as PCA, clustering or classification.

Usage

raman_align_peaks(
  peak_results,
  tolerance = 4,
  reference_wavenumbers = NULL,
  feature_method = "max_height"
)

Arguments

Details

Missing peaks are represented by zero in the returned feature matrix.

Value

Data frame with one row per reference wavenumber and one column per sample. The first column, wavenumber, contains bin reference positions.

Examples

wn = seq(400, 1800, by = 1)

x1 = dnorm(wn, mean = 1000, sd = 20) +
  0.01 * rnorm(length(wn))

x2 = dnorm(wn, mean = 1002, sd = 20) +
  0.01 * rnorm(length(wn))

p1 = raman_find_peaks(x1, height = 0.01, distance = 5)
p2 = raman_find_peaks(x2, height = 0.01, distance = 5)

peak_results = list(
  list(
    sample_id = "sample1",
    wavenumbers = wn,
    peaks = p1$peaks,
    properties = p1$properties
  ),
  list(
    sample_id = "sample2",
    wavenumbers = wn,
    peaks = p2$peaks,
    properties = p2$properties
  )
)

features = raman_align_peaks(
  peak_results,
  tolerance = 4,
  feature_method = "max_height"
)

head(features)

Description

Restricts a Raman or surface-enhanced Raman spectroscopy (SERS) dataset to an inclusive wavenumber interval.

Usage

raman_crop_spectra(dataset, min_wn, max_wn)

Arguments

Details

The dataset must contain wavenumbers in rows and samples in columns. Wavenumbers are obtained from dataset$wavenumbers when available. Otherwise, numeric row names in dataset$data are used.

Value

A dataset equivalent to dataset, with dataset$data restricted to wavenumbers between min_wn and max_wn. The dataset$wavenumbers element is updated or created to match the cropped data.

Examples

dataset = list(
  data = matrix(
    seq_len(30),
    nrow = 6,
    dimnames = list(
      as.character(seq(400, 900, by = 100)),
      paste0("sample", 1:5)
    )
  ),
  wavenumbers = seq(400, 900, by = 100)
)

cropped = raman_crop_spectra(
  dataset,
  min_wn = 500,
  max_wn = 800
)

cropped$wavenumbers

Description

Detects and replaces intensity spikes in Raman or surface-enhanced Raman spectroscopy (SERS) spectra using modified Z-scores based on the median absolute deviation (MAD).

Usage

raman_despike(dataset, ma = 10, threshold = 7)

Arguments

Details

The dataset must contain wavenumbers in rows and samples in columns. Each sample spectrum is processed independently. Detected spikes are replaced by the mean intensity of neighbouring non-spike points.

Value

A dataset equivalent to dataset, with despiked spectra stored in dataset$data. Information about the operation is stored in dataset$despike.

Examples

dataset = list(
  data = matrix(
    c(
      1, 1.1, 1.2, 20, 1.1, 1.2,
      2, 2.1, 2.2, 30, 2.1, 2.2
    ),
    nrow = 6,
    ncol = 2,
    dimnames = list(
      paste0("wavenumber", 1:6),
      paste0("sample", 1:2)
    )
  )
)

despiked = raman_despike(
  dataset,
  ma = 1,
  threshold = 3
)

despiked$despike$n_spikes

Description

Finds local maxima in one-dimensional spectra according to peak properties such as height, threshold, distance, prominence, width and plateau size.

Usage

raman_find_peaks(
  x,
  height = NULL,
  threshold = NULL,
  distance = NULL,
  prominence = NULL,
  width = NULL,
  plateau_size = NULL,
  wlen = NULL,
  rel_height = 0.5
)

Arguments

Details

The interface and peak-property terminology are inspired by scipy.signal.find_peaks(). Prominence and width calculations use an R implementation tailored to Raman/SERS spectra and are not intended to reproduce SciPy results exactly.

Value

A list with two elements:

peaks

Integer vector of one-based indices of peaks in x satisfying the requested conditions.

properties

Data frame containing properties of the returned peaks, including peak_index and peak_height. Additional columns are included when the respective properties are calculated.

Examples

x = c(0, 1, 0, 2, 1, 0, 3, 2, 0, 4, 1)

res = raman_find_peaks(
  x,
  height = 1,
  distance = 2
)

res$peaks
res$properties

Description

Normalizes Raman or surface-enhanced Raman spectroscopy (SERS) spectra using min-max normalization, L2 vector normalization, or normalization by the maximum intensity near a target silicon peak.

Usage

raman_normalize(
  dataset,
  method = "min-max",
  target_peak = 520,
  search_window = 20
)

Arguments

Details

The dataset must contain wavenumbers in rows and samples in columns. Each sample spectrum is normalized independently.

Value

A dataset equivalent to dataset, with normalized spectra stored in dataset$data. Information about the selected normalization method and its parameters is stored in dataset$normalization.

Examples

dataset = list(
  data = matrix(
    c(
      1, 2, 3, 4, 5,
      2, 4, 6, 8, 10
    ),
    nrow = 5,
    ncol = 2,
    dimnames = list(
      as.character(seq(500, 900, by = 100)),
      paste0("sample", 1:2)
    )
  ),
  wavenumbers = seq(500, 900, by = 100)
)

normalized = raman_normalize(
  dataset,
  method = "min-max"
)

range(normalized$data[, 1])

Description

Normalizes a peak-feature matrix returned by raman_align_peaks().

Usage

raman_normalize_peak_features(features, method = "min-max")

Arguments

Details

With method = "min-max", normalization is carried out row-wise across samples. Rows containing identical values for every sample are set to zero. With method = "L2", normalization is performed column-wise for each sample. Columns with zero L2 norm are retained as zero.

Value

Data frame with the same structure as features, containing normalized sample-feature values.

Examples

features = data.frame(
  wavenumber = c(1000, 1200, 1600),
  sample1 = c(1, 2, 3),
  sample2 = c(2, 4, 6)
)

raman_normalize_peak_features(features, method = "L2")

Description

Applies a Savitzky-Golay smoothing filter and computes the second derivative of Raman or surface-enhanced Raman spectroscopy (SERS) spectra.

Usage

raman_sgolay_derivative(dataset, window_size = 7, polynomial_order = 2)

Arguments

Details

This function is a wrapper around savitzky_golay() specialized for Raman/SERS datasets. It fixes the differentiation order to 2 and provides argument names and validation tailored to spectroscopic workflows.

The dataset must contain wavenumbers in rows and samples in columns. Each sample spectrum is processed independently. The second derivative is computed using polynomial fitting within a moving window.

Value

A dataset equivalent to dataset, with second derivative spectra stored in dataset$data. The original wavenumbers are stored in dataset$original_wavenumbers, when available. Information about the transform is stored in dataset$derivative.

Examples

wavenumbers = seq(400, 1000, length.out = 21)

dataset = list(
  data = cbind(
    sample1 = sin(seq(0, 2 * pi, length.out = 21)),
    sample2 = cos(seq(0, 2 * pi, length.out = 21))
  ),
  wavenumbers = wavenumbers
)

deriv2 = raman_sgolay_derivative(
  dataset,
  window_size = 7,
  polynomial_order = 2
)

dim(deriv2$data)

Description

Computes one-sided Fourier power spectra for each Raman or surface-enhanced Raman spectroscopy (SERS) sample.

Usage

raman_transform_fourier(dataset, remove_dc = FALSE)

Arguments

Details

This function follows the specmine Raman workflow by preserving the original spectrum in dataset$data. Fourier features are stored in dataset$fourier$data; their normalized frequencies are stored in dataset$fourier$frequencies.

The input data matrix must have wavenumbers in rows and samples in columns. Each sample is transformed independently.

Value

The input dataset, with a fourier element containing:

data

Matrix of one-sided Fourier power features, with frequencies in rows and samples in columns.

frequencies

Numeric vector of normalized frequencies.

remove_dc

Logical value indicating whether the DC component was removed.

n_wavenumbers

Number of original spectral variables.

Examples

wavenumbers = seq(400, 1000, length.out = 16)

dataset = list(
  data = cbind(
    sample1 = sin(seq(0, 2 * pi, length.out = 16)),
    sample2 = cos(seq(0, 2 * pi, length.out = 16))
  ),
  wavenumbers = wavenumbers
)

transformed = raman_transform_fourier(dataset)

dim(transformed$fourier$data)

Description

Computes a multilevel discrete Haar wavelet transform for each Raman or surface-enhanced Raman spectroscopy (SERS) sample.

Usage

raman_transform_wavelet(dataset, level = 1)

Arguments

Details

This function preserves the original data in dataset$data, following the specmine pipeline. Wavelet coefficients are stored in dataset$wavelet$data.

The input data matrix must have wavenumbers in rows and samples in columns. If needed, each spectrum is extended by repeating its final intensity value until its length is a power of two.

Value

The input dataset, with a wavelet element containing:

data

Matrix of Haar wavelet coefficients, with coefficients in rows and samples in columns.

filter

The character string "haar".

level

Number of decomposition levels.

original_length

Number of original wavenumbers.

padded_length

Length used after optional padding.

Examples

wavenumbers = seq(400, 1100, length.out = 8)

dataset = list(
  data = cbind(
    sample1 = sin(seq(0, 2 * pi, length.out = 8)),
    sample2 = cos(seq(0, 2 * pi, length.out = 8))
  ),
  wavenumbers = wavenumbers
)

transformed = raman_transform_wavelet(
  dataset,
  level = 2
)

dim(transformed$wavelet$data)

Description

Lists CSV files in a folder and reads them into a named list of data frames.

Usage

read_csvs_folder(foldername, verbose = TRUE, ...)

Arguments

Value

A named list of data frames, one per CSV file found in the folder.

Examples

folder <- file.path(tempdir(), "peak_csvs")
dir.create(folder, showWarnings = FALSE)
utils::write.csv(data.frame(ppm = c(1, 2), int = c(10, 20)),
                 file.path(folder, "sample1.csv"), row.names = FALSE)
utils::write.csv(data.frame(ppm = c(1, 3), int = c(15, 30)),
                 file.path(folder, "sample2.csv"), row.names = FALSE)
read_csvs_folder(folder, verbose = FALSE)