[Vegan-commits] r1428 - in pkg/vegan: R inst man

[noreply at r-forge.r-project.org]

Author: jarioksa
Date: 2010-12-31 08:53:23 +0100 (Fri, 31 Dec 2010)
New Revision: 1428

Modified:
   pkg/vegan/R/metaMDS.R
   pkg/vegan/inst/ChangeLog
   pkg/vegan/man/metaMDS.Rd
Log:
metaMDS tries to detect non-community data and adjust her behaviour

Modified: pkg/vegan/R/metaMDS.R
===================================================================
--- pkg/vegan/R/metaMDS.R	2010-12-18 19:50:03 UTC (rev 1427)
+++ pkg/vegan/R/metaMDS.R	2010-12-31 07:53:23 UTC (rev 1428)
@@ -4,6 +4,15 @@
               plot = FALSE, previous.best, old.wa = FALSE, ...) 
 {
     commname <- deparse(substitute(comm))
+    ## metaMDS was written for community data which should be all
+    ## positive. Check this here, and set arguments so that they are
+    ## suitable for non-negative data.
+    if (any(autotransform, noshare > 0, wascores) && any(comm < 0)) {
+        warning("'comm' has negative data: 'autotransform', 'noshare' and 'wascores' set to FALSE")
+        wascores <- FALSE
+        autotransform <- FALSE
+        noshare <- FALSE
+    }
     if (inherits(comm, "dist")) {
         dis <- comm
         if (is.null(attr(dis, "method")))

Modified: pkg/vegan/inst/ChangeLog
===================================================================
--- pkg/vegan/inst/ChangeLog	2010-12-18 19:50:03 UTC (rev 1427)
+++ pkg/vegan/inst/ChangeLog	2010-12-31 07:53:23 UTC (rev 1428)
@@ -6,7 +6,14 @@
 
 	* new version opened with the release of vegan 1.17-5 (submitted
 	Dec 17, 2010, released at CRAN on on Dec 18, 2010).
-	
+
+	* metaMDS: function was written for community data with all
+	entries non-negative, but it is used increasingly often with other
+	data types with a risk of failure (especially in wascores()). Now
+	data are checked for negative entries, and arguments wascores,
+	autotransform and noshare are set to FALSE with negative data
+	entries. Triggered by email queries and problem reports.
+
 Version 1.18-18 (closed December 17, 2010)
 
 	* mantel.correlog: proper treatment of ties in permutation

Modified: pkg/vegan/man/metaMDS.Rd
===================================================================
--- pkg/vegan/man/metaMDS.Rd	2010-12-18 19:50:03 UTC (rev 1427)
+++ pkg/vegan/man/metaMDS.Rd	2010-12-31 07:53:23 UTC (rev 1428)
@@ -254,7 +254,15 @@
   not save the used dissimilarity matrix, but \code{metaMDSredist}
   tries to reconstruct the used dissimilarities with original data
   transformation and possible \code{\link{stepacross}}.
+
+  The \code{metaMDS} function was designed to be used with community
+  data.  If you have other type of data, you should probably set some
+  arguments to non-default values: probably at least \code{wascores},
+  \code{autotransform} and \code{noshare} should be \code{FALSE}. If
+  you have negative data entries, \code{metaMDS} will set the previous
+  to \code{FALSE} with a warning.
 }
+
 \section{Warning}{
   The calculation of \code{\link{wascores}} for species was changed in
   \pkg{vegan} version 1.12-6. They are now based on the community data