generateSamplesInBatches: Simulate the samples in a single designated batch

Description Usage Arguments Value

View source: R/generatePCs.R

Description

This function generates the principal component (PC) location for all simulated cells in a single designated batch.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
generateSamplesInBatches(
  batchStructure,
  batch = "batch1",
  case_freqs,
  ctrl_freqs,
  cfcov,
  cf_scale,
  centroids,
  batch_shifts,
  sample_shifts,
  pc_cov_list,
  ncells
)

Arguments

batchStructure

The structure of the study design in which cases and controls are split into batches. These structures are output by the "distributeSample" functions (which can then be modified if specific structure is desired). If this parameter is kept as NULL, this function will automatically create a batchStructure with the "distributeSamples" function.

batch

The name of the batch that will be simulated

case_freqs

The cell state frequency distributions for all case samples. These are received as output from the "generateFreqs" function

ctrl_freqs

The cell state frequency distributions for all control samples. These are received as output from the "generateFreqs" function

cfcov

The cell state frequency variance-covariance across samples. This is obtained as output from the "estimateFreqVar" function

cf_scale

The magnitude of cell state frequency variation that cell states will exhibit across samples in the simulated dataset. Setting cf_scale = 1 will result in realistic levels of cell state frequency variation (as derived from parameter estimation of the input dataset). Increasing cf_scale results in higher variation, while decreasing cf_scale results in lower variation.

centroids

The mean PC values for each cell state. These are obtained as output from the "estimatePCVar" function.

batch_shifts

The simulated linear shifts in each batch for each cell state. These are received as output from the "generateBatchShifts" function.

sample_shifts

The simulated linear shifts in each sample for each cell state. These are received as output from the "generateSampleShifts" function.

pc_cov_list

A list containing the residual variance-covariance matrices for each cell state. These are obtained as output from the "estimatePCVar" function.

ncells

A vector containing the number of cells that will be simulated for each sample. The vector must be the same length as the number of total samples (ncases + nctrls).

Value

Returns a metadata table and the simulated PC locations for all simulated cells in the designated batch


immunogenomics/scpost documentation built on July 28, 2021, 4:03 a.m.