Visualize the layer modeling correlation of statistics

Source: R/layer_stat_cor_plot.R

This function makes a heatmap from the layer_stat_cor() correlation matrix between a given set of cell cluster/type statistics derived from scRNA-seq or snRNA-seq data (among other types) and the layer statistics from the Human DLPFC Visium data (when using the default arguments).

layer_stat_cor_plot(
  cor_stats_layer,
  max = 0.81,
  min = -max,
  layerHeights = NULL,
  cex = 1.2
)

Arguments

cor_stats_layer

The output of layer_stat_cor().

max

A numeric(1) specifying the highest correlation value for the color scale (should be between 0 and 1).

min

A numeric(1) specifying the lowest correlation value for the color scale (should be between 0 and -1).

layerHeights

A numeric() vector of length equal to ncol(cor_stats_layer) + 1 that starts at 0 specifying where to plot the y-axis breaks which can be used for re-creating the length of each brain layer. Gets passed to layer_matrix_plot().

cex

Passed to layer_matrix_plot().

Value

A heatmap for the correlation matrix between statistics.

Details

Check https://github.com/LieberInstitute/HumanPilot/blob/master/Analysis/Layer_Guesses/dlpfc_snRNAseq_annotation.R for a full analysis from which this family of functions is derived from.

See also

layer_matrix_plot annotate_registered_clusters

Other Layer correlation functions: annotate_registered_clusters(), layer_stat_cor()

Author

Andrew E Jaffe, Leonardo Collado-Torres

Examples


## Obtain the necessary data
if (!exists("modeling_results")) {
    modeling_results <- fetch_data(type = "modeling_results")
}
#> 2024-04-19 19:44:12.43869 loading file /github/home/.cache/R/BiocFileCache/2263e7e4ce8_Human_DLPFC_Visium_modeling_results.Rdata%3Fdl%3D1

## Compute the correlations
cor_stats_layer <- layer_stat_cor(
    tstats_Human_DLPFC_snRNAseq_Nguyen_topLayer,
    modeling_results,
    model_type = "enrichment"
)

## Visualize the correlation matrix
layer_stat_cor_plot(cor_stats_layer, max = max(cor_stats_layer))


## Annotate then re-plot
rownames(cor_stats_layer) <- paste0(
    rownames(cor_stats_layer),
    " - ",
    annotate_registered_clusters(cor_stats_layer)$layer_label
)
layer_stat_cor_plot(cor_stats_layer, max = max(cor_stats_layer))


## Restrict the range of colors further
layer_stat_cor_plot(cor_stats_layer, max = 0.25)


## Repeat with just the top 10 layer marker genes
layer_stat_cor_plot(layer_stat_cor(
    tstats_Human_DLPFC_snRNAseq_Nguyen_topLayer,
    modeling_results,
    model_type = "enrichment",
    top_n = 10
), max = 0.25)


## Now with the "pairwise" modeling results and also top_n = 10
layer_stat_cor_plot(layer_stat_cor(
    tstats_Human_DLPFC_snRNAseq_Nguyen_topLayer,
    modeling_results,
    model_type = "pairwise",
    top_n = 10
), max = 0.25)