getNeighborMatrix: Establish rough neighbor matching between samples given their...

In conos: Clustering on Network of Samples

Establish rough neighbor matching between samples given their projections in a common space

Description

Establish rough neighbor matching between samples given their projections in a common space

Usage

getNeighborMatrix(
  p1,
  p2,
  k,
  k1 = k,
  matching = "mNN",
  metric = "angular",
  l2.sigma = 1e+05,
  cor.base = 1,
  min.similarity = 1e-05
)

Arguments

p1	projection of sample 1
p2	projection of sample 2
k	neighborhood radius
k1	neighborhood radius
matching	string mNN (default) or NN (default='mNN')
metric	string Distance type (default: "angular", can also be 'L2')
l2.sigma	numeric L2 distances get transformed as exp(-d/sigma) using this value (default=1e5)
cor.base	numeric (default=1)
min.similarity	minimal similarity between two cells, required to have an edge

Value

matrix with the similarity (!) values corresponding to weight (1-d for angular, and exp(-d/l2.sigma) for L2)

conos documentation built on May 29, 2024, 9:56 a.m.