preProcess: preProcess

In lch14forever/BEEM: BEEM: Estimating Lotka-Volterra models from time-course microbiome sequencing data

preProcess

Description

Function to preprocess input data. Input counts are normalized with CSS within each subject across all time points and then normalized across subjects with method described by David et. al.

Usage

preProcess(
  counts,
  metadata,
  rsp,
  dev = 100,
  scaling = 5000,
  smooth_data = TRUE,
  ncpu = 10,
  norder = 3,
  finite_diff = FALSE,
  forceBreak = NULL
)

Arguments

counts	input data following MDSINE's OTU table (i.e. variables in rows and samples in columns)
metadata	metadata following MDSINE's metadata format
rsp	reference species used for alr transformation
dev	deviation (dev * mad) from the median to be considered as outliers
scaling	scale the normalized counts (default 5000)
smooth_data	data will be smoothed before initial normalization (default: TRUE)
ncpu	number of CPUs (default: 10)
norder	order of spline basis (default: 3)
finite_diff	use finite difference method to calculate gradients (default FALSE)
forceBreak	force to break the trajectory to handle pulsed perturbation (or species invasion) (default: NULL)

lch14forever/BEEM documentation built on April 5, 2025, 11:24 p.m.