Skip to contents

Run MDIC3 cell-cell communication inference

Source: R/RunMDIC3.R
RunMDIC3.Rd

Run MDIC3 cell-cell communication inference

Usage

RunMDIC3(object, ...)

# S3 method for class 'Seurat'
RunMDIC3(
  object,
  group.by,
  grn = NULL,
  grn_method = c("grnboost2", "gniplr", "correlation"),
  grn_backend = c("cpp", "python"),
  assay = NULL,
  layer = "data",
  backend = c("cpp", "python"),
  envname = "scop_env",
  conda = "auto",
  verbose = TRUE,
  ...
)

# S3 method for class 'matrix'
RunMDIC3(object, ...)

# Default S3 method
RunMDIC3(
  object,
  labels,
  grn = NULL,
  grn_method = c("grnboost2", "gniplr", "correlation"),
  grn_backend = c("cpp", "python"),
  genes_in = c("rows", "columns"),
  backend = c("cpp", "python"),
  envname = "scop_env",
  conda = "auto",
  verbose = TRUE,
  ...
)

Arguments

object

A Seurat object or a gene-by-cell expression matrix.

...

Additional arguments passed to [RunGRNBoost2()] or [RunGNIPLR()] when those methods are selected through `grn_method`.

group.by

Metadata column used as cell labels when `object` is a Seurat object.

grn

Gene-by-gene GRN matrix aligned to expression genes, or a data frame with columns `TF`, `target`, and `importance`.

grn_method

GRN inference method used when `grn = NULL`. `"grnboost2"` calls [RunGRNBoost2()]; `"gniplr"` calls [RunGNIPLR()]; `"correlation"` uses absolute expression correlation as a lightweight GRN approximation.

grn_backend

Backend passed to [RunGRNBoost2()] when `grn_method = "grnboost2"`.

assay

Assay used when `object` is a Seurat object.

layer

Assay layer used when `object` is a Seurat object.

backend

Runtime backend for the MDIC3 scoring step. Supports `"cpp"` and `"python"`.

envname

Python environment used when `backend = "python"`.

conda

Conda-compatible executable used when `backend = "python"`.

verbose

Whether to print the message. Default is TRUE.

labels

Cell labels used when `object` is a matrix.

genes_in

Matrix orientation for matrix inputs. `"rows"` means genes x cells; `"columns"` means cells x genes.

Value

A Seurat object with results stored in `object@tools[["MDIC3"]]`, or a list for matrix input.

Examples

data(pancreas_sub)
expr <- GetAssayData5(
  pancreas_sub,
  assay = SeuratObject::DefaultAssay(pancreas_sub),
  layer = "counts"
)
expr <- as.matrix(expr[, seq_len(8)])
expr <- expr[
  names(sort(apply(expr, 1, stats::var), decreasing = TRUE))[seq_len(5)],
]
labels <- as.character(pancreas_sub$SubCellType[seq_len(8)])

mdic3 <- RunMDIC3(
  expr,
  labels = labels,
  grn_method = "gniplr",
  correlation_threshold = 0,
  lasso_degree = 1,
  max_lag = 1,
  max_edges_per_target = 2
)
#> ℹ [2026-07-28 03:33:44] Running GNIPLR with `backend = cpp` on 5 genes and 8 cells
#> ℹ [2026-07-28 03:33:44] Running MDIC3 with `backend = cpp` and `grn_method = gniplr` on 5 genes and 8 cells
mdic3$celltype_communication
#>                Ductal Ngn3-high-EP      Beta Ngn3-low-EP
#> Ductal       0.000000    0.0000000 1.5942511   0.0000000
#> Ngn3-high-EP 0.000000    0.0000000 0.5611239   0.0000000
#> Beta         1.070243    0.8566177 2.2642573   0.6299158
#> Ngn3-low-EP  0.000000    0.0000000 0.7640191   0.0000000