Package 'simPIC'

Add feature statistics

Description

Add additional feature statistics to a SingleCellExperiment object

Usage

addFeatureStats(
  sce,
  value = "counts",
  log = FALSE,
  offset = 1,
  no.zeros = FALSE
)

Arguments

sce	SingleCellExperiment to add feature statistics to.
value	the count value to calculate statistics.
log	logical. Whether to take log2 before calculating statistics.
offset	offset to add to avoid taking log of zero.
no.zeros	logical. Whether to remove all zeros from each feature before calculating statistics.

Details

Currently adds the following statistics: mean and variance. Statistics are added to the rowData slot and are named Stat[Log]Value[No0] where Log and No0 are added if those arguments are true.

Value

SingleCellExperiment with additional feature statistics

---

Convert Sparse Matrix to SingleCellExperiment object

Description

This function converts a sparse matrix into a SingleCellExperiment(SCE) object.

Usage

convert_to_SCE(sparse_data)

Arguments

sparse_data

A sparse matrix containing count data, where rows are peaks and columns represent cells.

Value

A SingleCellExperiment(SCE) object with the sparse matrix stored in the "counts" assay.

---

Get counts from Single Cell Experiment object

Description

Get counts matrix from a SingleCellExperiment object. If counts is missing a warning is issued and the first assay is returned.

Usage

getCounts(sce)

Arguments

sce	SingleCellExperiment object

Value

counts matrix

---

simPIC: Simulate single-cell ATAC-seq data

Description

simPIC: Simulate single-cell ATAC-seq data

Value

globalvariables

---

newsimPICcount

Description

Create a newsimPICcount object to store parameters.

Usage

newsimPICcount(...)

Arguments

...	Variables to set newsimPICcount object parameters.

Details

This function creates the object variable which is passed in all functions.

Value

new object from class simPICcount.

Examples

object <- newsimPICcount()

---

Custom theme for ggplot2

Description

This function defines a custom theme for ggplot2 to ensure consistent visual appearance across multiple plots.

Usage

plot_theme()

Value

A ggplot2 theme object with predefined settings.

---

Bind rows (matched)

Description

Bind the rows of two data frames, keeping only the columns that are common to both.

Usage

rbindMatched(df1, df2)

Arguments

df1	first data.frame to bind.
df2	second data.frame to bind.

Value

data.frame containing rows from df1 and df2 but only common columns.

---

Select fit

Description

Trying two fitting methods and selecting the best one.

Usage

selectFit(data, distr, verbose = TRUE)

Arguments

data	The data to fit.
distr	Name of the distribution to fit.
verbose	logical. To print messages or not.

Details

The distribution is fitted to the data using each of the fitdist fitting methods. The fit with the smallest Cramer-von Mises statistic is selected.

Value

The selected fit object

---

Set simPIC parameters

Description

Set input parameters of the simPICcount object.

Usage

setsimPICparameters(object, update = NULL, ...)

Arguments

object	input simPICcount object.
update	new parameters.
...	set new parameters for simPICcount object.

Value

simPICcount object with updated parameters.

Examples

object <- newsimPICcount()
object <- setsimPICparameters(object, nCells = 200, nPeaks = 500)

---

Compare SingleCellExperiment objects

Description

Combine data from several SingleCellExperiment objects and produce some basic plots comparing them.

Usage

simPICcompare(
  sces,
  point.size = 0.2,
  point.alpha = 0.1,
  fits = TRUE,
  colours = NULL
)

Arguments

sces	named list of SingleCellExperiment objects to combine and compare.
point.size	size of points in scatter plots.
point.alpha	opacity of points in scatter plots.
fits	whether to include fits in scatter plots.
colours	vector of colours to use for each dataset.

Details

The returned list has three items:

RowData

Combined row data from the provided SingleCellExperiments.

ColData

Combined column data from the provided SingleCellExperiments.

Plots

Comparison plots

Means

Boxplot of mean distribution.

Variances

Boxplot of variance distribution.

MeanVar

Scatter plot with fitted lines showing the mean-variance relationship.

LibrarySizes

Boxplot of the library size distribution.

ZerosPeak

Boxplot of the percentage of each peak that is zero.

ZerosCell

Boxplot of the percentage of each cell that is zero.

MeanZeros

Scatter plot with fitted lines showing the mean-zeros relationship.

The plots returned by this function are created using ggplot and are only a sample of the kind of plots you might like to consider. The data used to create these plots is also returned and should be in the correct format to allow you to create further plots using ggplot.

Value

List containing the combined datasets and plots.

Examples

sim1 <- simPICsimulate(
    nPeaks = 1000, nCells = 500,
    pm.distr = "weibull", seed = 7856
)
sim2 <- simPICsimulate(
    nPeaks = 1000, nCells = 500,
    pm.distr = "gamma", seed = 4234
)
comparison <- simPICcompare(list(weibull = sim1, gamma = sim2))
names(comparison)
names(comparison$Plots)

---

The simPICcount class

Description

S4 class that holds parameters for simPIC simulation.

Value

a simPIC class object. The parameters not shown in brackets can be estimated from real data using simPICestimate. For details of the simPIC simulation see simPICsimulate. The default parameters are based on PBMC10k dataset and can be reproduced using test data and script provided in inst/script

Parameters

simPIC simulation parameters:

nPeaks

The number of peaks to simulate.

nCells

The number of cells to simulate.

[seed]

Seed to use for generating random numbers.

[default]

The logical variable whether to use default parameters (TRUE) or learn from data (FALSE)

Library size parameters

lib.size.meanlog

meanlog (location) parameter for the library size log-normal distribution.

lib.size.sdlog

sdlog (scale) parameter for the library size log-normal distribution.

Peak mean parameters

mean.scale

scale parameter for the mean weibull distribution.

mean.shape

shape parameter for the mean weibull distribution.

Cell sparsity parameters

sparsity

probability of openness to be multiplied to the input of poisson distribution to generate final simulated matrix.

---

Estimate simPIC Biological Coefficient of Variation parameters

Description

Parameters are estimated using the estimateDisp function in the edgeR package.

Usage

simPICEstBCV(counts, object, verbose)

Arguments

counts	counts matrix to estimate parameters from.
object	simPICcount object to store estimated values in.
verbose	logical. To print progress messages or not.

Details

The estimateDisp function is used to estimate the common dispersion and prior degrees of freedom. See estimateDisp for details. When estimating parameters on simulated data we found a broadly linear relationship between the true underlying common dispersion and the edgR estimate, therefore we apply a small correction, disp = -0.3 + 0.15 * edgeR.disp.

Value

simPICcount object with estimated values.

---

Estimate simPIC simulation parameters

Description

Estimate simulation parameters for library size, peak means, and sparsity for simPIC simulation from a real peak by cell input matrix

Usage

simPICestimate(
  counts,
  object = newsimPICcount(),
  pm.distr = c("gamma", "weibull", "pareto", "lngamma"),
  method = c("single", "groups"),
  verbose = TRUE
)

## S3 method for class 'SingleCellExperiment'
simPICestimate(
  counts,
  object = newsimPICcount(),
  pm.distr = "weibull",
  method = "single",
  verbose = TRUE
)

## S3 method for class 'dgCMatrix'
simPICestimate(
  counts,
  object = newsimPICcount(),
  pm.distr = "weibull",
  method = "single",
  verbose = TRUE
)

Arguments

counts	either a sparse peak by cell count matrix, or a SingleCellExperiment object containing count data to estimate parameters.
object	simPICcount object to store estimated parameters and counts.
pm.distr	statistical distribution for estimating peak mean parameters. Available distributions: gamma, weibull, lngamma, pareto. Default is weibull.
method	to use for simulation. Single for simulating one cell-type or groups for simulating distinct cell-types.
verbose	logical variable. Prints the simulation progress if TRUE.

Value

simPICcount object containing all estimated parameters.

Examples

counts <- readRDS(system.file("extdata", "test.rds", package = "simPIC"))
est <- newsimPICcount()
est <- simPICestimate(counts, pm.distr = "weibull")

---

Estimate simPIC library size parameters.

Description

Estimate the library size parameters for simPIC simulation.

Usage

simPICestimateLibSize(counts, object, verbose)

Arguments

counts	count matrix.
object	simPICcount object to store estimated values.
verbose	logical. To print messages or not.

Details

Parameters for the lognormal distribution are estimated by fitting the library sizes using fitdist. All the fitting methods are tried and the fit with the best Cramer-von Mises statistic is selected.

Value

simPICcount object with estimated library size parameters.

---

Estimate simPIC peak means

Description

Estimate peak mean parameters for simPIC simulation

Usage

simPICestimatePeakMean(norm.counts, object, pm.distr, verbose)

Arguments

norm.counts	library size normalised counts matrix.
object	simPICcount object to store estimated values.
pm.distr	distribution parameter for peak means.
verbose	logical. To print progress messages or not.

Details

Parameters for gamma distribution are estimated by fitting the mean normalised counts using fitdist. All the fitting methods are tried and the fit with the best Cramer-von Mises statistic is selected.

Value

simPICcount object containing all estimated parameters

---

Estimate sparsity.

Description

This function estimates the sparsity of cells based on a normalized counts matrix and updates the parameters of a simPIC object accordingly.

Usage

simPICestimateSparsity(norm.counts, object, verbose)

Arguments

norm.counts	A normalized count matrix to estimate parameters from.
object	simPICcount object to store estimated parameters.
verbose	logical. To print messages or not.

Value

simPICcount object with updated sparsity parameter.

---

Get a single simPICcount parameter

Description

Get the value of a single variable from input simPICcount object.

Usage

simPICget(object, name)

Arguments

object	input simPICcount object.
name	name of the parameter.

Value

Value of the input parameter.

Examples

object <- newsimPICcount()
nPeaks <- simPICget(object, "nPeaks")

---

Get parameters

Description

Get multiple parameter values from a simPIC object.

Usage

simPICgetparameters(object, names)

Arguments

object	input object to get values from.
names	vector of names of the parameters to get.

Value

List with the values of the selected parameters.

Examples

object <- newsimPICcount()
simPICgetparameters(object, c("nPeaks", "nCells", "peak.mean.shape"))

---

Simulate cell means

Description

Simulate a peak by cell matrix given the mean accessibility for each peak in each cell. Cells start with the mean accessibility for the group they belong to (when simulating groups). The selected means are adjusted for each cell's expected library size.

Usage

simPICsimSingleCellMeans(object, sim)

simPICsimulateGroupCellMeans(object, sim)

Arguments

object	simPIC object with simulation parameters.
sim	SingleCellExperiment to add cell means to.

Value

SingleCellExperiment with added cell means.

---

simPIC simulation

Description

Simulate peak by cell count matrix from a sparse single-cell ATAC-seq peak by cell input using simPIC methods.

Usage

simPICsimulate(
  object = newsimPICcount(),
  pm.distr = "weibull",
  method = c("single", "groups"),
  verbose = TRUE,
  ...
)

simPICsimulatesingle(object = newsimPICcount(), verbose = TRUE, ...)

simPICsimulatemulti(
  object = newsimPICcount(),
  pm.distr = "weibull",
  method = c("groups"),
  verbose = TRUE,
  ...
)

Arguments

object	simPICcount object with simulation parameters. See simPICcount for details.
pm.distr	distribution parameter for peak means. Available distributions: gamma, weibull, lngamma, pareto. Default is weibull.
method	to use for simulation. Single for simulating one cell-type or groups for simulating distinct cell-types.
verbose	logical variable. Prints the simulation progress if TRUE.
...	Any additional parameter settings to override what is provided in simPICcount object.

Details

simPIC provides the option to manually adjust each of the simPICcount object parameters by calling setsimPICparameters.

The simulation involves following steps:

1. Set up simulation parameters

2. Set up SingleCellExperiment object

3. Simulate library sizes

4. Simulate sparsity

5. Simulate peak means

6. Create final synthetic counts

The final output is a SingleCellExperiment object that contains the simulated count matrix. The parameters are stored in the colData (for cell specific information), rowData (for peak specific information) or assays (for peak by cell matrix) slots. This additional information includes:

Value

SingleCellExperiment object containing the simulated counts.

Examples

# default simulation
sim <- simPICsimulate(pm.distr = "weibull")

---

Simulate BCV means

Description

Simulate means for each peak in each cell that are adjusted to follow a mean-variance trend using Biological Coefficient of Variation taken from and inverse gamma distribution.

Usage

simPICsimulateBCVMeans(object, sim)

Arguments

object	simPICcount object with simulation parameters.
sim	SingleCellExperiment to add BCV means to.

Value

SingleCellExperiment with simulated BCV means.

---

Simulate simPIC library sizes

Description

Generate library sizes for cells in simPIC simulation based on the estimated values of mus and sigmas.

Usage

simPICsimulateLibSize(object, sim, verbose)

Arguments

object	simPICcount object with simulation parameters.
sim	SingleCellExperiment object containing simulation parameters.
verbose	logical. To print progress messages.

Value

SingleCellExperiment object with simulated library sizes.

---

Simulate group differential accessibility

Description

Simulate differential accessibility. Differential accessibility factors for each group are produced using getLNormFactors and these are added along with updated means for each group. For paths care is taken to make sure they are simulated in the correct order.

Usage

simPICsimulatemultiDA(object, sim)

Arguments

object	simPICcount object with simulation parameters.
sim	SingleCellExperiment to add differential accessibility to.

Value

SingleCellExperiment with simulated differential accessibility.

---

Simulate simPIC peak means.

Description

Generate peak means for cells in simPIC simulation based on the estimated values of shape and rate parameters.

Usage

simPICsimulatePeakMean(object, sim, pm.distr, verbose)

Arguments

object	simPICcount object with simulation parameters.
sim	SingleCellExperiment object containing simulation parameters.
pm.distr	distribution parameter for peak means. Available distributions: gamma, weibull, lngamma, pareto. Default is weibull.
verbose	logical. Whether to print progress messages.

Value

SingleCellExperiment object with simulated peak means.

---

Simulate true counts.

Description

Counts are simulated from a poisson distribution where each peak has a mean, expected library size and proportion of accessible chromatin.

Usage

simPICsimulateTrueCounts(object, sim)

Arguments

object	simPICcount object with simulation parameters.
sim	SingleCellExperiment object containing simulation parameters.

Value

SingleCellExperiment object with simulated true counts.

---

Simulate true counts groups.

Description

Counts are simulated from a poisson distribution where each peak has a mean, expected library size and proportion of accessible chromatin.

Usage

simPICsimulateTrueCountsGroups(object, sim)

Arguments

object	simPICcount object with simulation parameters.
sim	SingleCellExperiment object containing simulation parameters.

Value

SingleCellExperiment object with simulated true counts.

---