scoreLDA: Scores LDA

In qzhang503/proteoQ: Processing and Informatic Analysis of Mass Spectrometrirc Data

Scores LDA

Description

Scores LDA

Usage

scoreLDA(
  df,
  id,
  label_scheme_sub,
  anal_type,
  scale_log2r,
  center_features,
  scale_features,
  choice = "lda",
  method = "moment",
  type,
  col_group = "Group",
  folds,
  out_file,
  ...
)

Arguments

df	The name of a primary data file. By default, it will be determined automatically after matching the types of data and analysis with an id among c("pep_seq", "pep_seq_mod", "prot_acc", "gene"). A primary file contains normalized peptide or protein data and is among c("Peptide.txt", "Peptide_pVal.txt", "Peptide_impNA_pVal.txt", "Protein.txt", "Protein_pVal.txt", "protein_impNA_pVal.txt"). For analyses require the fields of significance p-values, the df will be one of c("Peptide_pVal.txt", "Peptide_impNA_pVal.txt", "Protein_pVal.txt", "protein_impNA_pVal.txt").
id	Character string; one of pep_seq, pep_seq_mod, prot_acc and gene.
label_scheme_sub	A data frame. Subset entries from label_scheme for selected samples.
anal_type	Character string; the type of analysis that are preset for method dispatch in function factories. The value will be determined automatically. Exemplary values include anal_type = c("PCA", "Corrplot", "EucDist", "GSPA", "Heatmap", "Histogram", "MDS", "Model", "NMF", "Purge", "Trend", "LDA", ...).
scale_log2r	Logical; if TRUE, adjusts log2FC to the same scale of standard deviation across all samples. The default is TRUE. At scale_log2r = NA, the raw log2FC without normalization will be used.
center_features	Logical; if TRUE, adjusts log2FC to center zero by features (proteins or peptides). The default is TRUE. Note the difference to data alignment with method_align in standPrn or standPep where log2FC are aligned by observations (samples).
scale_features	Logical; if TRUE, adjusts log2FC to the same scale of variance by features (protein or peptide entries). The default is TRUE. Note the difference to data scaling with scale_log2r where log2FC are scaled by observations (samples).
choice	Character string; the PCA method in c("prcomp"). The default is "prcomp".
method	Character string; the distance measure in one of c("euclidean", "maximum", "manhattan", "canberra", "binary") for dist. The default method is "euclidean".
type	Character string indicating the type of PCA by either observations or features. At the type = obs default, observations (samples) are in rows and features (peptides or proteins) in columns for prcomp. The principal components are then plotted by observations. Alternatively at type = feats, features (peptides or proteins) are in rows and observations (samples) are in columns. The principal components are then plotted by features.
col_group	Character string to a column key in expt_smry.xlsx. Samples corresponding to non-empty entries under col_group will be used for sample grouping in the indicated analysis. At the NULL default, the column key Group will be used. No data annotation by groups will be performed if the fields under the indicated group column is empty.
folds	Not currently used. Integer; the degree of folding data into subsets. The default is one without data folding.
out_file	A file path object to an output file.
...	filter_: Variable argument statements for the row filtration against data in a primary file linked to df. See also normPSM for the format of filter_ statements. Arguments passed to prcomp: rank., tol etc. At type = obs, argument scale becomes scale_features and center matches center_features. At type = feats, the setting of scale_log2r will be applied for data scaling and data centering be automated by standPep or standPrn. Additional arguments for ggsave: width, the width of plot; height, the height of plot ...

qzhang503/proteoQ documentation built on April 13, 2025, 8:31 a.m.