update testNhoods docs with details re: contrasts behaviours in #304

R/testNhoods.R

@@ -71,11 +71,18 @@
 #' function \code{fitGLMM} can be used to fit a mixed effect model to each
 #' nhood (see \code{fitGLMM} docs for details).
 #'
-#' Parallelisation is currently only enabled for the NB-LMM and uses the BiocParallel paradigm. In
-#' general the GLM implementation in \code{glmQLFit} is sufficiently fast that it does not require
+#' Parallelisation is currently only enabled for the NB-GLMM and uses the BiocParallel paradigm at
+#' the level of R, and OpenMP to allow multi-threading of RCpp code. In general the GLM implementation
+#' in \code{glmQLFit} is sufficiently fast that it does not require
 #' parallelisation. Parallelisation requires the user to pass a \linkS4class{BiocParallelParam} object
-#' with the parallelisation arguments contained therein. This relies on the user to specify how
-#' parallelisation - for details see the \code{BiocParallel} package.
+#' with the parallelisation arguments contained therein. This relies on the user specifying how to
+#' parallelise - for details see the \code{BiocParallel} package.
+#'
+#' \code{model.contrasts} are used to define specific comparisons for DA testing. Currently,
+#' \code{testNhoods} will take the last formula variable for comparisons, however, contrasts
+#' need this to be the first variable. A future update will harmonise these behaviours for
+#' consistency. While it is strictly feasible to compute multiple contrasts at once, the
+#' recommendation, for ease of interpretability, is to compute one at a time.
 #'
 #' @return A \code{data.frame} of model results, which contain:
 #' \describe{