Package 'CellDEEP'

K-means Based Cell Pooling for Seurat Objects

Description

Pools cells into "pseudocells" by applying k-means clustering to PCA embeddings. This reduces data sparsity while maintaining the biological grouping of sample, cluster, and condition.

Usage

CellDEEP.Kmean(
  dataset,
  n_cells = 10,
  nstart = 100,
  assay_name = "RNA",
  readcounts = "mean",
  min_cells_per_subgroup = 25
)

Arguments

dataset	A Seurat object. Must have PCA reductions calculated.
n_cells	Integer. Target number of cells to pool into each pseudocell.
nstart	Integer. Number of random sets to start with in kmeans.
assay_name	Character. The assay to pull counts from (default "RNA").
readcounts	Character. Aggregation method: "mean" (rounded average), "sum", "10X" (mean * 10).
min_cells_per_subgroup	Integer. Minimum cells required in each sample-cluster subgroup to perform pooling (default 25).

Value

A new Seurat object where each "cell" is a pooled group of original cells.

Note

This function requires that PCA has already been run on the input dataset, as it uses the "pca" reduction for clustering.

Examples

data("sim")
pool_input <- prepare_data(
  sim,
  sample_id = "DonorID",
  group_id = "Status",
  cluster_id = "cluster_id"
)

pooled_kmean <- CellDEEP.Kmean(
  pool_input,
  readcounts = "sum",
  n_cells = 3,
  min_cells_per_subgroup = 1,
  assay_name = "RNA"
)
pooled_kmean

---

Random Cell Pooling for Seurat Objects

Description

Pools cells into pseudocells by random selection within biological groups. Includes a minimum threshold filter of 25 cells per subgroup to ensure pooling quality.

Usage

CellDEEP.Random(
  dataset,
  n_cells = 10,
  assay_name = "RNA",
  min_cells_per_subgroup = 25,
  readcounts = "mean"
)

Arguments

dataset	A Seurat object.
n_cells	Integer. The number of cells to pool into each pseudocell.
assay_name	Character. The assay to use for counts (default "RNA").
min_cells_per_subgroup	Integer. Minimum cells required in each sample-cluster subgroup to perform pooling (default 25).
readcounts	Character. Method to aggregate counts: "sum" or "mean".

Value

A new Seurat object containing the aggregated pseudocells.

Note

Subgroups (sample-cluster combinations) with fewer than 25 cells are automatically skipped. The function also generates a DimPlot to visualize the random pooling across samples.

Examples

data("sim")
pool_input <- prepare_data(
  sim,
  sample_id = "DonorID",
  group_id = "Status",
  cluster_id = "cluster_id"
)

pooled_random <- CellDEEP.Random(
  pool_input,
  readcounts = "sum",
  n_cells = 3,
  min_cells_per_subgroup = 1,
  assay_name = "RNA"
)
pooled_random

---

Differential Expression with Optional Cell Pooling

Description

It can run Seurat DE directly or first aggregate cells into metacells using CellDEEP pooling.

Usage

FindMarker.CellDEEP(
  object,
  ident.1 = NULL,
  ident.2 = NULL,
  group.by = "group_id",
  sample_id = NULL,
  group_id = NULL,
  cluster_id = NULL,
  prepare = TRUE,
  test.use = "wilcox",
  Pool = TRUE,
  readcounts = "sum",
  n_cells = 10,
  assay = "RNA",
  min_cells_per_subgroup = 25,
  cell_selection = "kmean",
  name.only = TRUE,
  logfc.threshold = 0.25,
  min.pct = 0.01,
  p_cutoff = 0.05,
  full_list = FALSE,
  ...
)

Arguments

object	A Seurat object.
ident.1	Character. First identity group to compare.
ident.2	Character. Second identity group to compare.
group.by	Character. Metadata column used for grouping (default "group_id").
sample_id	Character. Input metadata column for sample IDs.
group_id	Character. Input metadata column for group IDs.
cluster_id	Character. Input metadata column for cluster IDs.
prepare	Logical. If TRUE, run prepare_data first.
test.use	Character. DE test to use.
Pool	Logical. If TRUE, perform CellDEEP pooling before DE (default TRUE).
readcounts	Character. Pool aggregation method: "sum", "mean", or "10X".
n_cells	Integer. Target number of cells per pool.
assay	Character. Assay to use (default "RNA").
min_cells_per_subgroup	Integer. Minimum cells in each sample-cluster subgroup required for pooling.
cell_selection	Character. Pooling strategy: "kmean" or "random".
name.only	Logical. If TRUE, return gene names only.
logfc.threshold	Numeric. Minimum log fold-change.
min.pct	Numeric. Minimum detection rate.
p_cutoff	Numeric. Adjusted p-value threshold.
full_list	Logical. If TRUE, return all genes regardless of p-value.
...	Additional arguments passed to Seurat::FindMarkers.

Value

A vector of gene names or a DE data.frame.

---

Standardize Seurat Metadata for CellDEEP

Description

Standardizes metadata columns to sample_id, group_id, and cluster_id so CellDEEP functions can run consistently.

Usage

prepare_data(
  Subset.Seurat,
  assay = "RNA",
  sample_id,
  group_id,
  cluster_id,
  file_path = NULL
)

Arguments

Subset.Seurat	A Seurat object.
assay	Character. Assay to use (default "RNA").
sample_id	Character. Metadata column name for sample IDs.
group_id	Character. Metadata column name for group IDs.
cluster_id	Character. Metadata column name for cluster IDs.
file_path	Character. Reserved for compatibility.

Value

A Seurat object with standardized metadata fields.

---

Perform Differential Expression and Filter Results

Description

A wrapper for Seurat::FindMarkers that simplifies the extraction of Differentially Expressed (DE) genes. It supports p-value filtering and can return either gene names or a full results table.

Usage

return.DE(
  dataset,
  test.use = "wilcox",
  DE.ident.1,
  DE.ident.2,
  DE.group,
  assay = "RNA",
  p_cutoff = 0.05,
  name.only = TRUE,
  logfc.threshold = 0.25,
  min.pct = 0.01,
  full_list = FALSE,
  ...
)

Arguments

dataset	A Seurat object.
test.use	Character. DE test to use (default "wilcox").
DE.ident.1	Identifier(s) for the first group of cells.
DE.ident.2	Identifier(s) for the second group of cells.
DE.group	Character. Metadata column to group by.
assay	Character. Assay to use (default "RNA").
p_cutoff	Numeric. Adjusted p-value threshold (default 0.05).
name.only	Logical. If TRUE, return gene names only.
logfc.threshold	Numeric. Minimum log fold change (default 0.1).
min.pct	Numeric. Minimum fraction of cells expressing a gene.
full_list	Logical. If TRUE, return all genes and skip p-value filter.
...	Extra arguments passed to Seurat::FindMarkers.

Value

A character vector of genes or a marker data.frame.

---

Sample simulated cells from muscat package

Description

A dataset containing 200 simulated cells(100 per group) for demonstrating CellDEEP functions. Can be found at doi:10.5281/zenodo.18863779

Usage

data(sim)

Format

A Seurat object

Source

simulated data with muscat package

---