10/23/2006	Updated defaults for tuning (epsilon), values for
0.0.2	   	added writeToFile method
0.0.3           Constrain r to [0, 1]
0.0.4           Got rid of InitialMCMC and XdeMCMC classes/methods.
0.0.5           defaultStartingValues<- sets specifiedInitialValues to FALSE.  firstIteration<- sets specifiedInitialValues to TRUE.

0.0.6           - added GeneMeta to depends in order to use zScoresMethod
		- C code no longer restrics r and rho to be nonnegative
		- this version does not pass R CMD check and is too big (5 MB) for BioC		

0.0.7           - move non-essential dependencies to Suggests in the DESCRIPTION file
		- changed example for concordance
		- the following error occurs when running xde() on my mac:
   				 *** caught bus error ***	
				  address 0x0, cause 'invalid alignment
                
                Tthe following warnings occurred with R CMD check:
		* checking for CRLF line endings in C/C++/Fortran sources/headers ... WARNING
		Found the following sources/headers with CRLF line endings:
		  src/Random.cpp
 		  src/Random.h	

		  The CRLF line ending problem was corrected by running dos2unix:
		 dos2unix Random.ccp 
		 dos2unix Random.h 

		 - After the CRLF fix, xde() runs fine on the mac

		 - 0.0.7 passes R CMD check without warnings

		 - added executable cXde that removes *.o and *.so
                   files from XDE/src before running R CMD check

0.0.8            - modified validXdeSet function
		 - put a check for RColorBrewer in pairs()

0.0.9           Add method to not model the nu's (resets various parameters in the XdeSet object) 
		Add method to standardize the variance of the columns in the ExpressionSets and to mean center the genes
		Export validXdeSet
		Added method for filtering by integrative correlation
		.integ.
		- modified NAMESPACE to export MergeMaid methods/classses
		- plot(object) makes traceplots of the various parameters
		- added studySplit method for skeptic's validation


Next Release: the .tar.gz package is 5MB.  Need to use examples that
take < 1MB and pass R CMD check. Might help to make the vignette take
less memory

Need to generalize integrativeCorrelationFilter to an arbritary number
of studies.  Would be better if this were handled in MergeMaid.
mergeExprs does not work for a list of ExpressionSets. Should work for
a mergeExpressionSet object that contains a list of ExpressionSets in
@data

Should we change the class to a simple list of ExpressionSets with
validity checks on the dimnames?  Could remove MergeMaid from Depends.

				  
