
Add a proximity transition column to differential proximity results
add_proximity_transition.RdClassifies the reference/rest and target group medians into proximity regimes
around zero ("anti" for < -tau, "random" for
[-tau, tau], and "coloc" for > tau) and combines them
into a transition column formatted as "<ref_state> -> <tgt_state>".
Usage
add_proximity_transition(
data,
tau = 0.15,
tgt_col = "med_tgt",
ref_col = NULL,
transition_col = "transition",
separator = " -> "
)Arguments
- data
A
data.frameortbl_dfreturned byDifferentialProximityAnalysisorFindAllProximityMarkers.- tau
Non-negative numeric threshold defining the neutral
"random"zone[-tau, tau]around 0. Default is0.15.- tgt_col
Name of the target median column. Default is
"med_tgt".- ref_col
Name of the reference median column. When
NULL(default), automatically uses"med_ref"or"med_rest"if present indata.- transition_col
Name of the output transition column. Default is
"transition".- separator
Separator placed between the reference and target states. Default is
" -> ".
Value
data with a character column named by
transition_col. An existing column of that name is replaced.
Details
Works on output tables from both DifferentialProximityAnalysis
(using med_ref and med_tgt) and
FindAllProximityMarkers (using med_rest and
med_tgt).
Examples
library(dplyr)
df <- tibble(
target = c("B", "CD4 T", "NK"),
med_tgt = c(0.8, 0.9, -0.4),
med_rest = c(0.0, 0.5, 0.0),
diff_median = med_tgt - med_rest
)
add_proximity_transition(df, tau = 0.15)
#> # A tibble: 3 × 5
#> target med_tgt med_rest diff_median transition
#> <chr> <dbl> <dbl> <dbl> <chr>
#> 1 B 0.8 0 0.8 random -> coloc
#> 2 CD4 T 0.9 0.5 0.4 coloc -> coloc
#> 3 NK -0.4 0 -0.4 random -> anti