foldchange: Computes Log2(A) - Log2(B) Fold Change of (non-) paired data.

In OmicFlow: Fast and Efficient (Automated) Analysis of Sparse Omics Data

Computes Log2(A) - Log2(B) Fold Change of (non-) paired data.

Description

Computes (non-)paired Log2(A) - Log2(B) Fold Change. This function is built into the class omics with method DFE() and inherited by other omics classes, such as; metagenomics and proteomics. The function handles zero's, and doesn't return +/- infinites.

Usage

foldchange(
  data,
  feature_rank,
  condition_A,
  condition_B,
  condition_labels,
  paired = FALSE
)

Arguments

data	A data.table.
feature_rank	A character variable of the feature level (e.g. "Genus" in taxonomy).
condition_A	A vector of categorical characters, it is possible to specify multiple labels.
condition_B	A vector of categorical characters, it is possible to specify multiple labels.
condition_labels	A vector character wherein condition_A and condition_B are present.
paired	A Boolean value to perform paired or non-paired test, see wilcox.test.

Value

A data.table

Examples

#-------------------------#
##      NON-PAIRED       ##
#-------------------------#
# Load required library
library(data.table)

# Define parameters and variables
sample_ids <- c("S1_A", "S2_A", "S3_A", "S4_B", "S5_B", "S6_B")
feature_ids <- c("Feature1", "Feature2", "Feature3")

# Simulated abundance matrix (features x samples)
abundances <- matrix(
  c(
    # Feature1 (e.g. GenusA)
    100, 120, 110,  55, 60, 65,
    # Feature2 (e.g. GenusB)
    50, 65, 60,    130, 120, 125,
    # Feature3 (e.g. GenusC)
    80, 85, 90,     80, 85, 90
  ),
  nrow = 3, byrow = TRUE,
  dimnames = list(feature_ids, sample_ids)
)

# A wide table with columns as samples, rows as features
# And an additional column as the feature_rank, a column for feature comparison.
mock_data <- data.table(
  Genus = feature_ids,  # feature_rank column (e.g. "Genus")
  S1_A = abundances[ , 1],
  S2_A = abundances[ , 2],
  S3_A = abundances[ , 3],
  S4_B = abundances[ , 4],
  S5_B = abundances[ , 5],
  S6_B = abundances[ , 6]
)
print(mock_data)

# It uses substring matching, and multiple conditions can be used
res <- foldchange(
  data = mock_data,
  feature_rank = "Genus",
  condition_A = c("_A", "_B"),
  condition_B = c("_B", "_A"),

  # This can also be a column wherein, conditions A and B are present
  condition_labels = sample_ids, 
  paired = FALSE
)
print(res)

#---------------------#
##      PAIRED       ##
#---------------------#
library(data.table)

# Define paired sample ids for 3 pairs:
paired_ids <- paste0("Pair", 1:3)

# Features:
feature_ids <- c("Feature1", "Feature2", "Feature3")

# Simulate abundances for each paired sample:
# For each pair, we have two samples: condition A and condition B.
# Make sure the length of condition A and condition B are the same!

# Construct the data.table with features as rows
mock_data_paired <- data.table(
  Genus = feature_ids,
  Pair1_A = c(100, 50, 80),
  Pair1_B = c(60, 130, 75),
  Pair2_A = c(120, 65, 85),
  Pair2_B = c(60, 120, 90),
  Pair3_A = c(110, 60, 90),
  Pair3_B = c(65, 125, 85) 
)

res <- foldchange(
  data = mock_data_paired,
  feature_rank = "Genus",
  condition_A = c("_A", "_B"),
  condition_B = c("_B", "_A"),

  # This can also be a column wherein, conditions A and B are present
  condition_labels = names(mock_data_paired)[-1], 
  paired = TRUE
)
print(res)

OmicFlow documentation built on Sept. 9, 2025, 5:24 p.m.