MCmat: MCmat

In adw96/DivNet: Diversity Estimation in Networked Ecological Communities

MCmat

Description

This function simulates MC step for an entire matrix. Should not need to be used by user directly; available to help with determining network estimation.

Usage

MCmat(
  Y,
  W,
  eY,
  N,
  Q,
  base,
  sigma,
  MCiter,
  stepsize = 1,
  perturbation = 0.05,
  network = "default",
  ncores = 1,
  ...
)

Arguments

Y	logratio matrix
W	corresponding count matrix
eY	current expected value of logratio matrix
N	number of samples, or nrow of Y
Q	number of OTUs, or ncol of W
base	OTU index used for base
sigma	current estimate of sigma
MCiter	number of MC samples to generate
stepsize	variance used for MH samples, defaults to 1. Tweak to adjust acceptance ratio
perturbation	size of purturbation used for to_log_ratios, defaults to 0.05
network	How to estimate network. Defaults to "default" (generalised inverse, aka naive). Other options include "diagonal", or a function that takes a sample covariance matrix and returns an estimate of the inverse covariance matrix (eg glasso or SpiecEasi)
ncores	number of cores to use, defaults to 1
...	additional arguments to be supplied to the network function

Author(s)

Bryan Martin

Amy Willis

adw96/DivNet documentation built on Oct. 2, 2023, 11:49 a.m.