Differential analysis (colocalization)

Runs differential analysis on Moran's Z colocalization scores generated with the Pixelator data processing pipeline.

Usage

RunDCA(object, ...)

# S3 method for data.frame
RunDCA(
  object,
  target,
  reference,
  contrast_column,
  group_vars = NULL,
  alternative = c("two.sided", "less", "greater"),
  conf_int = TRUE,
  p_adjust_method = c("bonferroni", "holm", "hochberg", "hommel", "BH", "BY", "fdr"),
  cl = NULL,
  verbose = TRUE,
  ...
)

# S3 method for Seurat
RunDCA(
  object,
  target,
  reference,
  contrast_column,
  assay = NULL,
  group_vars = NULL,
  alternative = c("two.sided", "less", "greater"),
  conf_int = TRUE,
  p_adjust_method = c("bonferroni", "holm", "hochberg", "hommel", "BH", "BY", "fdr"),
  cl = NULL,
  verbose = TRUE,
  ...
)

Arguments

object: An object containing colocalization scores
...: Not yet implemented
target: The name of the target group
reference: The name of the reference group
contrast_column: The name of the column where the group labels are stored. This column must include target and reference.
group_vars: An optional character vector with column names to group the tests by.
alternative: One of 'two.sided', 'less' or 'greater' (see ?wilcox.test for details)
conf_int: Should confidence intervals be computed? (see ?wilcox.test for details)
p_adjust_method: One of "bonferroni", "holm", "hochberg", "hommel", "BH", "BY" or "fdr". (see ?p.adjust for details)
cl: A cluster object created by makeCluster, or an integer to indicate number of child-processes (integer values are ignored on Windows) for parallel evaluations. See Details on performance in the documentation for pbapply. The default is NULL, which means that no parallelization is used.
verbose: Print messages
assay: Name of assay to use

Value

A tbl_df object with test results

Details

If you are working with a Seurat object containing a CellGraphAssay, the polarization scores are accessed directly from the CellGraphAssay. A character vector or factor must be selected with contrast_column from the input data (or @meta.data slot from a Seurat object) which holds the groups to run the test for. The target and reference parameters should refer to the names of the two groups used for the comparison and these names should be present in the contrast_column.

Additional groups

The test is always computed between target and reference, but it is possible to add additional grouping variables with group_vars. If group_vars is used, the test will be computed within each combination of groups. For instance, if we have annotated cells into cell type populations across two conditions defined by target and reference, we can pass the name of a cell annotation column with group_vars to run the test for each cell type.

Examples

library(pixelatorR)
library(dplyr)

pxl_file <- system.file("extdata/five_cells",
                        "five_cells.pxl",
                        package = "pixelatorR")
# Seurat objects
seur1 <- seur2 <- ReadMPX_Seurat(pxl_file)
#> ✔ Created a 'Seurat' object with 5 cells and 80 targeted surface proteins
seur1$sample <- "Sample1"
seur2$sample <- "Sample2"
seur_merged <- merge(seur1, seur2, add.cell.ids = c("A", "B"))

# Subset data to run test on a few markers
seur_merged <- subset(seur_merged,
                      features = c("CD3", "CD4", "CD8", "CD19",
                                   "CD20", "CD45RA", "CD45RO"))

# Run DCA
dca_markers <- RunDCA(seur_merged, contrast_column = "sample",
                      target = "Sample1", reference = "Sample2")
#> ℹ Splitting data by: marker_1, marker_2
#> ℹ Running 10 tests
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
#> Warning: cannot compute exact p-value with ties
#> Warning: cannot compute exact confidence intervals with ties
dca_markers
#> # A tibble: 10 × 15
#>      estimate data_type target  reference    n1    n2 statistic     p conf.low
#>         <dbl> <chr>     <chr>   <chr>     <int> <int>     <dbl> <dbl>    <dbl>
#>  1  0         pearson_z Sample1 Sample2       4     4       8       1   -10.8 
#>  2  0         pearson_z Sample1 Sample2       5     5      12.5     1    -4.61
#>  3 -0.0000165 pearson_z Sample1 Sample2       5     5      12.5     1   -14.6 
#>  4  0         pearson_z Sample1 Sample2       5     5      12.5     1   -11.2 
#>  5  0         pearson_z Sample1 Sample2       4     4       8       1    -9.33
#>  6  0         pearson_z Sample1 Sample2       4     4       8       1   -12.7 
#>  7  0         pearson_z Sample1 Sample2       4     4       8       1   -13.0 
#>  8  0         pearson_z Sample1 Sample2       5     5      12.5     1   -14.7 
#>  9  0         pearson_z Sample1 Sample2       5     5      12.5     1    -6.46
#> 10 -0.0000464 pearson_z Sample1 Sample2       5     5      12.5     1    -8.07
#> # ℹ 6 more variables: conf.high <dbl>, method <chr>, alternative <chr>,
#> #   marker_1 <chr>, marker_2 <chr>, p_adj <dbl>