Convert colocalization score table to an Assay or Assay5

Usage

ColocalizationScoresToAssay(object, ...)

# S3 method for class 'data.frame'
ColocalizationScoresToAssay(
  object,
  values_from = c("pearson_z", "pearson"),
  ...
)

# S3 method for class 'MPXAssay'
ColocalizationScoresToAssay(
  object,
  values_from = c("pearson_z", "pearson"),
  ...
)

# S3 method for class 'CellGraphAssay'
ColocalizationScoresToAssay(
  object,
  values_from = c("pearson_z", "pearson"),
  ...
)

# S3 method for class 'CellGraphAssay5'
ColocalizationScoresToAssay(
  object,
  values_from = c("pearson_z", "pearson"),
  ...
)

# S3 method for class 'Seurat'
ColocalizationScoresToAssay(
  object,
  assay = NULL,
  new_assay = NULL,
  values_from = c("pearson_z", "pearson"),
  ...
)

Arguments

object: An object with colocalization scores
...: Not yet implemented
values_from: What column to pick colocalization scores from. One of "pearson" or "pearson_z"
assay: Name of the CellGraphAssay to pull polarization scores from
new_assay: Name of the Assay to store the polarization scores in

Behavior

Takes an object with colocalization scores in long format and returns an object with colocalization scores in a wide format. The colocalization score table includes various colocalization scores along with p-values for each pair markers and component.

The wide format is an array-like object with dimensions (markers_1 * marker_2) x components, where each cell is filled with a polarization score. Scores that are missing from the colocalization score table are replaced with 0's.

Different outputs are returned depending on the input object type:

tibble/data.frame: returns a matrix with marker pairs in rows and components in columns
CelGraphAssay: returns an Assay with marker pairs in rows and components in columns
Seurat object: returns the Seurat object with a new Assay with marker pairs in rows and components in columns

As many functions provided in Seurat works on Assay objects, it is sometimes convenient to make this conversion. For instance, if we want to compute a UMAP on the colocalization scores with RunUMAP, we need the values to be formatted in an Assay. This also makes it possible to use various visualization functions such as VlnPlot or FeaturePlor to show the distribution of colocalization scores.

Examples

library(pixelatorR)
library(SeuratObject)

# Load example data as a Seurat object
pxl_file <- system.file("extdata/five_cells",
  "five_cells.pxl",
  package = "pixelatorR"
)
col_scores <- ReadMPX_colocalization(pxl_file)
#> ℹ Loading item(s) from: /private/var/folders/gw/bdcqhnvs0m9gs_mq8n51jtbc0000gn/T/Rtmpn9YTAb/temp_libpath1398336799173/pixelatorR/extdata/five_cells/five_cells.pxl
#> →   Loading colocalization data
#> ✔ Returning a 'tbl_df' object

# ColocalizationScoresToAssay returns a matrix for a tbl_df
col_scores_mat <- ColocalizationScoresToAssay(col_scores)
col_scores_mat[1:4, 1:5]
#>            RCVCMP0000217 RCVCMP0000118 RCVCMP0000487 RCVCMP0000655
#> ACTB/B2M      -0.9401321      0.000000     -2.879992     4.5740882
#> ACTB/CD102     1.9241318      0.000000      3.157566    -0.4834448
#> B2M/CD102      4.7268224      2.190192     -3.101690     2.3802011
#> ACTB/CD11a     2.1011979      0.000000     -1.804651     8.0947966
#>            RCVCMP0000263
#> ACTB/B2M       1.9368818
#> ACTB/CD102     0.5934362
#> B2M/CD102     -5.2602955
#> ACTB/CD11a     5.4727656

# Create a Seurat object
seur <- ReadMPX_Seurat(pxl_file)
#> ✔ Created a 'Seurat' object with 5 cells and 80 targeted surface proteins

# Fetch CellGraphAssay and create new polarization
# scores Assay
cg_assay <- seur[["mpxCells"]]
class(cg_assay)
#> [1] "CellGraphAssay5"
#> attr(,"package")
#> [1] "pixelatorR"
col_assay <- ColocalizationScoresToAssay(cg_assay)
class(col_assay)
#> [1] "Assay5"
#> attr(,"package")
#> [1] "SeuratObject"

# Convert colocalization scores within a Seurat object
seur <- ColocalizationScoresToAssay(seur)

# After conversion, we now have a new assay called "colocalization"
seur[["colocalization"]]
#> Assay (v5) data with 3160 features for 5 cells
#> First 10 features:
#>  ACTB/B2M, ACTB/CD102, B2M/CD102, ACTB/CD11a, B2M/CD11a, CD102/CD11a,
#> ACTB/CD11b, B2M/CD11b, CD102/CD11b, CD11a/CD11b 
#> Layers:
#>  data 

# Switch default assay to polarization
DefaultAssay(seur) <- "colocalization"

# Visualize colocalization scores with Seurat
# VlnPlot(seur, features = "CD19") +
#   ggplot2::labs(y = "Colocalization score")

Convert colocalization score table to an Assay or Assay5

Usage

Arguments

Behavior

See also

Examples