--- a
+++ b/OmicsFold/man/normalise.clr.within.features.Rd
@@ -0,0 +1,32 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/MixMC.R
+\name{normalise.clr.within.features}
+\alias{normalise.clr.within.features}
+\title{Apply centered log-ratio normalisation within features only}
+\usage{
+normalise.clr.within.features(input, offset = 0)
+}
+\arguments{
+\item{input}{Data as proportions 0-1.}
+
+\item{offset}{Optional offset to apply to raw data to avoid logging of zero
+values. Only needed if any zeroes are present - should generally be set very
+small, e.g. 0.000001.}
+}
+\value{
+Data normalised by the within-feature CLR method
+}
+\description{
+Apply centered log-ratio (CLR) normalisation to other compositional data, but
+restrict normalisation to *within* features only. This is another method for
+converting the compositional data, i.e. proportional data in the range 0..1
+to Euclidean space which is most appropriate for the linear models. The
+implementation is the same as CLR, but on transposed input data (which is
+then transposed back to the input orientation). Note this is experimental,
+though it does give a sensible normalisation.
+}
+\examples{
+\dontrun{
+normalise.clr.within.features(data.proportional)
+}
+}