Hi Is is possible for mrlm in limma to treat duplicates the same way as gls.series does ? I.e using consensus correlation We have tried implementing this in our group ourselves but with no success thus far. Is there any reasons why this would not be good idea ? Any info greatfully received Cheers Jason -- -------------------------------- Jason Skelton Pathogen Microarrays Wellcome Trust Sanger Institute Hinxton Cambridge CB10 1SA Tel +44(0)1223 834244 Ext 7123 Fax +44(0)1223 494919