Package 'RBPEqBind'

Call Peaks from Binding Simulation

Description

Identifies peak regions based on binding occupancy. If the transcript headers contain genomic coordinates (e.g. "chr1:1000-2000(+)"), it appends translated genomic coordinate columns (chrom, genomic_start, genomic_end, strand).

Usage

callPeaks(
  results,
  rbp_primary,
  rbp_reference = NULL,
  method = c("competitive", "local", "global"),
  threshold = 2,
  min_width = 3,
  window_size = 50
)

Arguments

results	data.table of simulation results.
rbp_primary	Name of the primary RBP.
rbp_reference	Name of the reference RBP (for competitive advantage) (default NULL).
method	Method: "competitive" (default), "local", or "global".
threshold	Threshold (ratio for competitive, occupancy for others) (default 2.0).
min_width	Minimum width of peak to report (default 3).
window_size	Window size for local background (default 50).

Value

A data.table of peaks with start, end, scores, and optionally genomic coordinates.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBinding("ACGUACGUACGUUUUUACGUACGU", rbp_models, c(HH = 100, HL = 100))
results$transcript <- "test_transcript"
peaks <- callPeaks(results, rbp_primary = "HH", method = "global", threshold = 1.1)
head(peaks)

---

Export BED format

Description

Simple BED export for ranges exceeding threshold. If the transcript headers contain genomic coordinates (e.g. "chr1:1000-2000(+)"), it translates transcript-relative peak coordinates into genomic coordinates.

Usage

exportBed(results, output_file, rbp, threshold = 0)

Arguments

results	data.table.
output_file	File path.
rbp	RBP name to export.
threshold	Threshold for score/occupancy.

Value

Invisibly returns NULL. Called for side effect of writing BED file.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBinding("ACGUACGU", rbp_models, c(HH = 100, HL = 100))
results$transcript <- "test_transcript"
tmp_bed <- tempfile(fileext = ".bed")
unlink(tmp_bed)

---

Export Results

Description

Exports simulation results to JSON or CSV.

Usage

exportResults(
  results,
  output_file,
  format = c("json", "csv"),
  include_sequence = TRUE
)

Arguments

results	data.table of results.
output_file	Output file path.
format	"json" or "csv".
include_sequence	Logical, include sequence in output (default TRUE).

Value

Invisibly returns NULL. Called for side effect of writing file.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBinding("ACGUACGU", rbp_models, c(HH = 100, HL = 100))

# Export JSON
tmp_json <- tempfile(fileext = ".json")
exportResults(results, tmp_json, format = "json")
unlink(tmp_json)

# Export CSV
tmp_csv <- tempfile(fileext = ".csv")
exportResults(results, tmp_csv, format = "csv")
unlink(tmp_csv)

---

Generate Random RNA Sequence

Description

Generates a random RNA sequence with specified nucleotide frequencies.

Usage

generateRNA(L, A = 0.25, G = 0.25, C = 0.25, U = 0.25)

Arguments

L	Length of the sequence.
A	Frequency of Adenine.
G	Frequency of Guanine.
C	Frequency of Cytosine.
U	Frequency of Uracil.

Value

A character string representing the RNA sequence.

Examples

set.seed(42)
generateRNA(100)

---

Import Results

Description

Reads previously exported simulation results from a JSON or CSV file.

Usage

importResults(input_file)

Arguments

input_file

Path to JSON or CSV file (exported by exportResults).

Value

A data.table with the imported results.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBinding("ACGUACGU", rbp_models, c(HH = 100, HL = 100))

# JSON Example
tmp_json <- tempfile(fileext = ".json")
exportResults(results, tmp_json, format = "json")
results_json <- importResults(tmp_json)
unlink(tmp_json)

# CSV Example
tmp_csv <- tempfile(fileext = ".csv")
exportResults(results, tmp_csv, format = "csv")
results_csv <- importResults(tmp_csv)
unlink(tmp_csv)

---

Load RBP Model from CSV

Description

Reads a CSV file containing RBP binding affinities/scores. The file should have a 'Motif' column and one or more RBP score columns.

Usage

loadModel(file, k = NULL, rbp = NULL)

Arguments

file	Path to the CSV file.
k	Optional integer. If provided, filters for k-mers of this length.
rbp	Optional character vector of RBP names to load. If NULL, loads all.

Value

A named list of data.tables, each with 'motif' and 'score' columns.

Examples

# Load sample model from package
model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
raw_models <- loadModel(model_file)
names(raw_models)

# Load specific RBPs
raw_models <- loadModel(model_file, rbp = c("HH", "HL"))

---

Create SummarizedExperiment from Results

Description

Converts simulation results to a SummarizedExperiment object.

Usage

makeSE(results, rbp_models = NULL)

Arguments

results	data.table from simulateBinding or simulateBindingFasta.
rbp_models	Optional named list of RBP models (for colData metadata).

Value

A SummarizedExperiment with assays and row/col metadata.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBinding("ACGUACGU", rbp_models, c(HH = 100, HL = 100))

# Create density columns for the mock results to enable makeSE
results[, HH_density := HH / sum(HH)]
results[, HL_density := HL / sum(HL)]
results[, HH_density_fc := HH_density * 8]
results[, HL_density_fc := HL_density * 8]
results[, HH_occupancy_fc := HH / mean(HH)]
results[, HL_occupancy_fc := HL / mean(HL)]

se <- makeSE(results, rbp_models)
se

---

Parse Genomic Coordinates from FASTA Header

Description

Extracts genomic coordinates from a UCSC-style header string, e.g. "chr1:1000-2000" or "chr1:1000-2000(-)".

Usage

parseGenomicHeader(h)

Arguments

h	Character string.

Value

A list with chrom, start, end, and strand, or NULL.

---

Plot Binding Profile

Description

Plots the binding profile for one or more RBPs.

Usage

plotBinding(
  results,
  rbp,
  transcript = NULL,
  metric = c("occupancy", "density", "density_fc", "occupancy_fc"),
  ylim = NULL,
  xlim = NULL,
  xaxis_type = c("nucleotide", "position", "both"),
  rna_conc = NULL,
  protein_conc = NULL,
  window = NULL
)

Arguments

results	data.table of simulation results.
rbp	Character vector of RBP names to plot.
transcript	Optional transcript name to filter (default NULL).
metric	Type of metric: "occupancy" (default), "density", "density_fc", or "occupancy_fc".
ylim	Optional y-axis limits (default NULL).
xlim	Optional x-axis limits (default NULL).
xaxis_type	Type of x-axis labels: "nucleotide" (default), "position", or "both".
rna_conc	Optional numeric. Filter by RNA concentration (default NULL).
protein_conc	Optional named numeric vector. Filter by protein concentration (default NULL).
window	Optional integer. Applies K-mer sliding window average (default NULL).

Value

A ggplot object.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBinding("ACGUACGUACGU", rbp_models, c(HH = 100, HL = 100))
results$transcript <- "test_transcript"
plotBinding(results, rbp = c("HH", "HL"))

---

Plot Competition Grid

Description

Visualizes binding across a concentration grid.

Usage

plotGrid(
  results,
  rbp1,
  rbp2,
  roi_range,
  metric = c("occupancy", "density", "density_fc", "occupancy_fc"),
  rbp1_concs = NULL
)

Arguments

results	data.table from simulateGrid.
rbp1	Name of first RBP (subdivides each cell).
rbp2	Name of second RBP (X-axis).
roi_range	Numeric vector (start, end) for region of interest.
metric	"occupancy" (default), "density", "density_fc", or "occupancy_fc".
rbp1_concs	Optional vector of RBP1 concentrations to include (default NULL).

Value

A ggplot object.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateGrid(
  sequence = "ACGUACGU",
  rbp_models = rbp_models,
  protein_conc_grid = list(HH = c(10, 100), HL = c(10, 100)),
  rna_conc_grid = 10
)
plotGrid(results, rbp1 = "HH", rbp2 = "HL", roi_range = c(1, 8))

---

Plot Binding Heatmap

Description

Plots a heatmap of binding for multiple RBPs.

Usage

plotHeatmap(
  results,
  transcript,
  rbps = NULL,
  xlim = NULL,
  xaxis_type = c("nucleotide", "position", "both"),
  metric = c("occupancy", "density", "density_fc", "occupancy_fc"),
  window = NULL,
  zlim = NULL
)

Arguments

results	data.table of simulation results.
transcript	Transcript name to filter (required).
rbps	Optional vector of RBP names to include (default NULL, infers all).
xlim	Optional x-axis limits (default NULL).
xaxis_type	Type of x-axis labels: "nucleotide" (default), "position", or "both".
metric	Type of metric: "occupancy" (default), "density", "density_fc", or "occupancy_fc".
window	Optional integer for K-mer sliding window average (default NULL).
zlim	Optional numeric vector for color scale limits (default NULL).

Value

A ggplot object.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBinding("ACGUACGUACGU", rbp_models, c(HH = 100, HL = 100))
results$transcript <- "test_transcript"
plotHeatmap(results, transcript = "test_transcript")

---

Set Model Affinity (Score to Kd Conversion)

Description

Converts a model's normalized scores to Kd values using affinity scaling. Score is assumed to be proportional to affinity (Ka), so Kd = 1/Ka.

Usage

setModel(models, max_affinity = 100, min_affinity = 1e-05)

Arguments

models	Named list of models (from loadModel), each with 'motif' and 'score'.
max_affinity	Named numeric vector of max Ka per RBP, or single default (100).
min_affinity	Named numeric vector of min Ka per RBP, or single default (0.00001).

Value

A named list of data.tables, each with 'motif' and 'Kd' columns.

Examples

# Load and process model
model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
raw_models <- loadModel(model_file)
rbp_models <- setModel(raw_models, max_affinity = 100)

# Check Kd values
head(rbp_models[[1]])

---

Simulate Competitive Binding

Description

Simulates competitive binding of multiple RBPs to a single RNA sequence.

Usage

simulateBinding(sequence, rbp_models, protein_concs, rna_conc = 10, k = 5)

Arguments

sequence	RNA sequence string.
rbp_models	Named list of data.frames/data.tables with 'motif' and 'Kd'.
protein_concs	Named numeric vector of total protein concentrations.
rna_conc	Total RNA concentration (default 10).
k	K-mer size (default 5).

Value

A data.table containing per-position binding probabilities/occupancies.

Examples

# Load and process model
model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
raw_models <- loadModel(model_file, rbp = c("HH", "HL"))
rbp_models <- setModel(raw_models, max_affinity = 100)

# Simulate binding on a short sequence
seq <- "ACGUACGUACGUACGUACGU"
results <- simulateBinding(seq, rbp_models, c(HH = 100, HL = 100), rna_conc = 10)
head(results)

---

Simulate Binding from FASTA File(s)

Description

Simulates competitive binding for all sequences in one or more FASTA files. Concurrency is controlled by the user's active plan (default is sequential).

Usage

simulateBindingFasta(
  fasta_file,
  rbp_models,
  protein_concs,
  rna_conc = 10,
  k = 5
)

Arguments

fasta_file	Character vector of path(s) to FASTA files.
rbp_models	Named list of RBP models.
protein_concs	Named numeric vector of protein concentrations.
rna_conc	RNA concentration (default 10).
k	K-mer size (default 5).

Value

A data.table with combined results, including a transcript column.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
fasta_file <- system.file("extdata", "test_transcripts.fa", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateBindingFasta(fasta_file, rbp_models, c(HH = 100, HL = 100))
head(results)

---

Simulate Concentration Grid

Description

Runs simulations across a grid of protein/RNA concentrations. Concurrency is controlled by the user's active plan (default is sequential).

Usage

simulateGrid(sequence, rbp_models, protein_conc_grid, rna_conc_grid, k = 5)

Arguments

sequence	RNA sequence string.
rbp_models	Named list of RBP models.
protein_conc_grid	List of numeric vectors for each RBP concentration.
rna_conc_grid	Numeric vector of RNA concentrations to sweep.
k	K-mer size (default 5).

Value

A data.table with results for all grid combinations.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
grid <- simulateGrid("ACGUACGU", rbp_models,
  protein_conc_grid = list(HH = c(10, 100), HL = c(10, 100)),
  rna_conc_grid = 10)
head(grid)

---

Simulate Grid from FASTA File(s)

Description

Simulates competitive binding for all sequences in one or more FASTA files across a concentration grid. Concurrency is controlled by the user's active plan (default is sequential).

Usage

simulateGridFasta(
  fasta_file,
  rbp_models,
  protein_conc_grid,
  rna_conc_grid,
  k = 5
)

Arguments

fasta_file	Character vector of path(s) to FASTA files.
rbp_models	Named list of RBP models.
protein_conc_grid	List of numeric vectors for each RBP concentration.
rna_conc_grid	Numeric vector of RNA concentrations to sweep.
k	K-mer size (default 5).

Value

A data.table with combined results, including a transcript column.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
fasta_file <- system.file("extdata", "test_transcripts.fa", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file, rbp = c("HH", "HL")))
results <- simulateGridFasta(
  fasta_file = fasta_file,
  rbp_models = rbp_models,
  protein_conc_grid = list(HH = c(10, 100), HL = c(10, 100)),
  rna_conc_grid = 10,
  k = 5
)
head(results)

---

View Model Affinity Distributions

Description

Creates a histogram of Kd or Ka distributions for RBP models.

Usage

viewModel(
  rbp_models,
  rbp = NULL,
  metric = c("Kd", "Ka"),
  bins = 50,
  colors = NULL,
  alpha = 0.5,
  log_scale = FALSE
)

Arguments

rbp_models	Named list of RBP model data.tables (from setModel).
rbp	Character vector of RBP names to plot (default NULL, plots all).
metric	Display metric: "Kd" (default) or "Ka" (affinity = 1/Kd).
bins	Number of histogram bins (default 50).
colors	Named vector of colors per RBP (default NULL, uses Okabe-Ito palette).
alpha	Transparency of histogram bars (default 0.5).
log_scale	Logical, use log10 scale for x-axis (default FALSE).

Value

A ggplot object.

Examples

model_file <- system.file("extdata", "model_RBP.csv", package = "RBPEqBind")
rbp_models <- setModel(loadModel(model_file))
viewModel(rbp_models, metric = "Kd")

---