resampling: Random sampling to do KNN-Louvain/leiden clustering

In renjun0324/KKLClustering: Large-scale single-cell clustering algorithm based on optimal KNN graph structure

Random sampling to do KNN-Louvain/leiden clustering

Description

This function is used to randomly sample each cluster multiple times to select the most powerful KNN graph structure.

Usage

resampling(
  dat = dat,
  outlier_kmeans = NULL,
  knn_range = c(3:70),
  cluster_method = "louvain",
  resolution = 1,
  iter = 30,
  is_weight = TRUE,
  python_path = "/usr/bin/python3",
  seed = 723
)

Arguments

dat	expression matrix with cell * feature
outlier_kmeans	the result from pre_partitioning function
knn_range	the range of the number of neighbors in the KNN graph structure
cluster_method	louvain or leiden in resampling
resolution	clustering parameters settings in resampling
iter	the number of iterations in resampling
is_weight	Whether to use distance weights when constructing the KNN graph
python_path	which python
seed	random seed

Details

The currently available indices are:

• Ball_Hall

• Banfeld_Raftery

• C_index

• Calinski_Harabasz

• Davies_Bouldin

• Det_Ratio

• Dunn

• Gamma

• G_plus

• Ksq_DetW

• Log_Det_Ratio

• Log_SS_Ratio

• McClain_Rao

• PBM

• Point_Biserial

• Ray_Turi

• Ratkowsky_Lance

• Scott_Symons

• SD_Scat

• SD_Dis

• S_Dbw

• Silhouette

• Tau

• Trace_W

• Trace_WiB

• Wemmert_Gancarski

• Xie_Beni

• GDI11

• GDI12

• GDI13

• GDI21

• GDI22

• GDI23

• GDI31

• GDI32

• GDI33

• GDI41

• GDI42

• GDI43

• GDI51

• GDI52

• GDI53

renjun0324/KKLClustering documentation built on Oct. 8, 2024, 7:39 p.m.