tof_assess_clusters_knn: Assess a clustering result by calculating a cell's cluster...

In keyes-timothy/tidytof: Analyze High-dimensional Cytometry Data Using Tidy Data Principles

Assess a clustering result by calculating a cell's cluster assignment to that of its K nearest neighbors.

Description

This function evaluates the result of a clustering procedure by finding the cell's K nearest neighbors, determining which cluster the majority of them are assigned to, and checking if this matches the cell's own cluster assignment. If the cluster assignment of the majority of a cell's nearest neighbors does not match with the cell's own cluster assignment, the cell is flagged as potentially anomalous.

Usage

tof_assess_clusters_knn(
  tof_tibble,
  cluster_col,
  marker_cols = where(tof_is_numeric),
  num_neighbors = min(10, nrow(tof_tibble)),
  distance_function = c("euclidean", "cosine", "l2", "ip"),
  augment = FALSE
)

Arguments

tof_tibble	A 'tof_tbl' or 'tibble'.
cluster_col	An unquoted column name indicating which column in 'tof_tibble' stores the cluster ids for the cluster to which each cell belongs. Cluster labels can be produced via any method the user chooses - including manual gating, any of the functions in the 'tof_cluster_*' function family, or any other method.
marker_cols	Unquoted column names indicating which column in 'tof_tibble' should be interpreted as markers to be used in the mahalanobis distance calculation. Defaults to all numeric columns. Supports tidyselection.
num_neighbors	An integer indicating how many neighbors should be found during the nearest neighbor calculation.
distance_function	A string indicating which distance function should be used to perform the k nearest neighbor calculation. Options are "euclidean" (the default) and "cosine".
augment	A boolean value indicating if the output should column-bind the computed flags for each cell (see below) as new columns in 'tof_tibble' (TRUE) or if a tibble including only the computed flags should be returned (FALSE, the default).

Value

If augment = FALSE (the default), a tibble with 2 columns: ".knn_cluster" (a character vector indicating which cluster received the majority vote of each cell's k nearest neighbors) and "flagged_cell" (a boolean value indicating if the cell's cluster assignment matched the majority vote (TRUE) or not (FALSE)). If augment = TRUE, the same 2 columns will be column-bound to tof_tibble, and the resulting tibble will be returned.

Examples

sim_data <-
    dplyr::tibble(
        cd45 = c(rnorm(n = 1000, sd = 1.5), rnorm(n = 1000, mean = 2), rnorm(n = 1000, mean = -2)),
        cd38 = c(rnorm(n = 1000, sd = 1.5), rnorm(n = 1000, mean = 2), rnorm(n = 1000, mean = -2)),
        cd34 = c(rnorm(n = 1000, sd = 1.5), rnorm(n = 1000, mean = 2), rnorm(n = 1000, mean = -2)),
        cd19 = c(rnorm(n = 1000, sd = 1.5), rnorm(n = 1000, mean = 2), rnorm(n = 1000, mean = -2)),
        cluster_id = c(rep("a", 1000), rep("b", 1000), rep("c", 1000))
    )

knn_result <-
    sim_data |>
    tof_assess_clusters_knn(
        cluster_col = cluster_id,
        num_neighbors = 10
    )

keyes-timothy/tidytof documentation built on March 31, 2024, 12:01 p.m.