generic_dist: Compute pairwise generic distances

Description Usage Arguments Value

Description

A wrapper for dist function.

Usage

1
2
3
4
generic_dist(X, method = c("manhattan", "euclidean", "exp manhattan",
  "exp euclidean", "maximum", "canberra", "binary", "minkowski",
  "jaccard"), log_trans = TRUE, log_base = 10, min_row_sum = 100,
  min_row_prevalence = 5)

Arguments

X

A data matrix, e.g. gene expression

method

a character string indicating which distance method to use.

log_trans

A boolean indicating whether to log transform the data prio to distance computation (log(X + 1)). Default is FALSE.

log_base

A number indicating base for log transformation. Default is 10.

min_row_sum

A scalar indicating the minimum feature (species) summ across columns (samples) to be kept for distance computation. Only used for Jaccard distance.

min_row_prevalence

An integer indicating the minimum prevalence (non-zero occurance) of a feature (species) across samples to be kept for distance computation. Only used for Jaccard distance.

Value

A dissimilarity matrix, D.


nlhuong/buds documentation built on May 17, 2019, 3:13 a.m.